git/builtin/gc.c

1635 строки
40 KiB
C
Исходник Обычный вид История

/*
* git gc builtin command
*
* Cleanup unreachable files and optimize the repository.
*
* Copyright (c) 2007 James Bowes
*
* Based on git-gc.sh, which is
*
* Copyright (c) 2006 Shawn O. Pearce
*/
#include "builtin.h"
#include "repository.h"
#include "config.h"
#include "tempfile.h"
#include "lockfile.h"
#include "parse-options.h"
#include "run-command.h"
#include "sigchain.h"
#include "strvec.h"
#include "commit.h"
#include "commit-graph.h"
#include "packfile.h"
#include "object-store.h"
#include "pack.h"
#include "pack-objects.h"
#include "blob.h"
#include "tree.h"
#include "promisor-remote.h"
#include "refs.h"
maintenance: add prefetch task When working with very large repositories, an incremental 'git fetch' command can download a large amount of data. If there are many other users pushing to a common repo, then this data can rival the initial pack-file size of a 'git clone' of a medium-size repo. Users may want to keep the data on their local repos as close as possible to the data on the remote repos by fetching periodically in the background. This can break up a large daily fetch into several smaller hourly fetches. The task is called "prefetch" because it is work done in advance of a foreground fetch to make that 'git fetch' command much faster. However, if we simply ran 'git fetch <remote>' in the background, then the user running a foreground 'git fetch <remote>' would lose some important feedback when a new branch appears or an existing branch updates. This is especially true if a remote branch is force-updated and this isn't noticed by the user because it occurred in the background. Further, the functionality of 'git push --force-with-lease' becomes suspect. When running 'git fetch <remote> <options>' in the background, use the following options for careful updating: 1. --no-tags prevents getting a new tag when a user wants to see the new tags appear in their foreground fetches. 2. --refmap= removes the configured refspec which usually updates refs/remotes/<remote>/* with the refs advertised by the remote. While this looks confusing, this was documented and tested by b40a50264ac (fetch: document and test --refmap="", 2020-01-21), including this sentence in the documentation: Providing an empty `<refspec>` to the `--refmap` option causes Git to ignore the configured refspecs and rely entirely on the refspecs supplied as command-line arguments. 3. By adding a new refspec "+refs/heads/*:refs/prefetch/<remote>/*" we can ensure that we actually load the new values somewhere in our refspace while not updating refs/heads or refs/remotes. By storing these refs here, the commit-graph job will update the commit-graph with the commits from these hidden refs. 4. --prune will delete the refs/prefetch/<remote> refs that no longer appear on the remote. 5. --no-write-fetch-head prevents updating FETCH_HEAD. We've been using this step as a critical background job in Scalar [1] (and VFS for Git). This solved a pain point that was showing up in user reports: fetching was a pain! Users do not like waiting to download the data that was created while they were away from their machines. After implementing background fetch, the foreground fetch commands sped up significantly because they mostly just update refs and download a small amount of new data. The effect is especially dramatic when paried with --no-show-forced-udpates (through fetch.showForcedUpdates=false). [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/FetchStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:31 +03:00
#include "remote.h"
#include "object-store.h"
maintenance: add start/stop subcommands Add new subcommands to 'git maintenance' that start or stop background maintenance using 'cron', when available. This integration is as simple as I could make it, barring some implementation complications. The schedule is laid out as follows: 0 1-23 * * * $cmd maintenance run --schedule=hourly 0 0 * * 1-6 $cmd maintenance run --schedule=daily 0 0 * * 0 $cmd maintenance run --schedule=weekly where $cmd is a properly-qualified 'git for-each-repo' execution: $cmd=$path/git --exec-path=$path for-each-repo --config=maintenance.repo where $path points to the location of the Git executable running 'git maintenance start'. This is critical for systems with multiple versions of Git. Specifically, macOS has a system version at '/usr/bin/git' while the version that users can install resides at '/usr/local/bin/git' (symlinked to '/usr/local/libexec/git-core/git'). This will also use your locally-built version if you build and run this in your development environment without installing first. This conditional schedule avoids having cron launch multiple 'git for-each-repo' commands in parallel. Such parallel commands would likely lead to the 'hourly' and 'daily' tasks competing over the object database lock. This could lead to to some tasks never being run! Since the --schedule=<frequency> argument will run all tasks with _at least_ the given frequency, the daily runs will also run the hourly tasks. Similarly, the weekly runs will also run the daily and hourly tasks. The GIT_TEST_CRONTAB environment variable is not intended for users to edit, but instead as a way to mock the 'crontab [-l]' command. This variable is set in test-lib.sh to avoid a future test from accidentally running anything with the cron integration from modifying the user's schedule. We use GIT_TEST_CRONTAB='test-tool crontab <file>' in our tests to check how the schedule is modified in 'git maintenance (start|stop)' commands. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:18 +03:00
#include "exec-cmd.h"
#define FAILED_RUN "failed to run %s"
static const char * const builtin_gc_usage[] = {
N_("git gc [<options>]"),
NULL
};
static int pack_refs = 1;
static int prune_reflogs = 1;
gc: default aggressive depth to 50 This commit message is long and has lots of background and numbers. The summary is: the current default of 250 doesn't save much space, and costs CPU. It's not a good tradeoff. Read on for details. The "--aggressive" flag to git-gc does three things: 1. use "-f" to throw out existing deltas and recompute from scratch 2. use "--window=250" to look harder for deltas 3. use "--depth=250" to make longer delta chains Items (1) and (2) are good matches for an "aggressive" repack. They ask the repack to do more computation work in the hopes of getting a better pack. You pay the costs during the repack, and other operations see only the benefit. Item (3) is not so clear. Allowing longer chains means fewer restrictions on the deltas, which means potentially finding better ones and saving some space. But it also means that operations which access the deltas have to follow longer chains, which affects their performance. So it's a tradeoff, and it's not clear that the tradeoff is even a good one. The existing "250" numbers for "--aggressive" come originally from this thread: http://public-inbox.org/git/alpine.LFD.0.9999.0712060803430.13796@woody.linux-foundation.org/ where Linus says: So when I said "--depth=250 --window=250", I chose those numbers more as an example of extremely aggressive packing, and I'm not at all sure that the end result is necessarily wonderfully usable. It's going to save disk space (and network bandwidth - the delta's will be re-used for the network protocol too!), but there are definitely downsides too, and using long delta chains may simply not be worth it in practice. There are some numbers in that thread, but they're mostly focused on the improved window size, and measure the improvement from --depth=250 and --window=250 together. E.g.: http://public-inbox.org/git/9e4733910712062006l651571f3w7f76ce64c6650dff@mail.gmail.com/ talks about the improved run-time of "git-blame", which comes from the reduced pack size. But most of that reduction is coming from --window=250, whereas most of the extra costs come from --depth=250. There's a link in that thread showing that increasing the depth beyond 50 doesn't seem to help much with the size: https://vcscompare.blogspot.com/2008/06/git-repack-parameters.html but again, no discussion of the timing impact. In an earlier thread from Ted Ts'o which discussed setting the non-aggressive default (from 10 to 50): http://public-inbox.org/git/20070509134958.GA21489%40thunk.org/ we have more numbers, with the conclusion that going past 50 does not help size much, and hurts the speed of normal operations. So from that, we might guess that 50 is actually a sweet spot, even for aggressive, if we interpret aggressive to "spend time now to make a better pack". It is not clear that "--depth=250" is actually a better pack. It may be slightly _smaller_, but it carries a run-time penalty. Here are some more recent timings I did to verify that. They show three things: - the size of the resulting pack (so disk saved to store, bandwidth saved on clones/fetches) - the cost of "rev-list --objects --all", which shows the effect of the delta chains on trees (commits typically don't delta, and the command doesn't touch the blobs at all) - the cost of "log -Sfoo", which will additionally access each blob All cases were repacked with "git repack -adf --depth=$d --window=250" (so basically, what would happen if we tweaked the "gc --aggressive" default depth). The timings are all wall-clock best-of-3. The machine itself has plenty of RAM compared to the repositories (which is probably typical of most workstations these days), so we're really measuring CPU usage, as the whole thing will be in disk cache after the first run. The core.deltaBaseCacheLimit is at its default of 96MiB. It's possible that tweaking it would have some impact on the tests, as some of them (especially "log -S" on a large repo) are likely to overflow that. But bumping that carries a run-time memory cost, so for these tests, I focused on what we could do just with the on-disk pack tradeoffs. Each test is done for four depths: 250 (the current value), 50 (the current default that tested well previously), 100 (to show something on the larger side, which previous tests showed was not a good tradeoff), and 10 (the very old default, which previous tests showed was worse than 50). Here are the numbers for linux.git: depth | size | % | rev-list | % | log -Sfoo | % -------+-------+-------+----------+--------+-----------+------- 250 | 967MB | n/a | 48.159s | n/a | 378.088 | n/a 100 | 971MB | +0.4% | 41.471s | -13.9% | 342.060 | -9.5% 50 | 979MB | +1.2% | 37.778s | -21.6% | 311.040s | -17.7% 10 | 1.1GB | +6.6% | 32.518s | -32.5% | 279.890s | -25.9% and for git.git: depth | size | % | rev-list | % | log -Sfoo | % -------+-------+-------+----------+--------+-----------+------- 250 | 48MB | n/a | 2.215s | n/a | 20.922s | n/a 100 | 49MB | +0.5% | 2.140s | -3.4% | 17.736s | -15.2% 50 | 49MB | +1.7% | 2.099s | -5.2% | 15.418s | -26.3% 10 | 53MB | +9.3% | 2.001s | -9.7% | 12.677s | -39.4% You can see that that the CPU savings for regular operations improves as we decrease the depth. The savings are less for "rev-list" on a smaller repository than they are for blob-accessing operations, or even rev-list on a larger repository. This may mean that a larger delta cache would help (though setting core.deltaBaseCacheLimit by itself doesn't). But we can also see that the space savings are not that great as the depth goes higher. Saving 5-10% between 10 and 50 is probably worth the CPU tradeoff. Saving 1% to go from 50 to 100, or another 0.5% to go from 100 to 250 is probably not. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2016-08-11 19:13:09 +03:00
static int aggressive_depth = 50;
static int aggressive_window = 250;
static int gc_auto_threshold = 6700;
static int gc_auto_pack_limit = 50;
static int detach_auto = 1;
static timestamp_t gc_log_expire_time;
static const char *gc_log_expire = "1.day.ago";
static const char *prune_expire = "2.weeks.ago";
static const char *prune_worktrees_expire = "3.months.ago";
static unsigned long big_pack_threshold;
static unsigned long max_delta_cache_size = DEFAULT_DELTA_CACHE_SIZE;
static struct strvec pack_refs_cmd = STRVEC_INIT;
static struct strvec reflog = STRVEC_INIT;
static struct strvec repack = STRVEC_INIT;
static struct strvec prune = STRVEC_INIT;
static struct strvec prune_worktrees = STRVEC_INIT;
static struct strvec rerere = STRVEC_INIT;
tempfile: auto-allocate tempfiles on heap The previous commit taught the tempfile code to give up ownership over tempfiles that have been renamed or deleted. That makes it possible to use a stack variable like this: struct tempfile t; create_tempfile(&t, ...); ... if (!err) rename_tempfile(&t, ...); else delete_tempfile(&t); But doing it this way has a high potential for creating memory errors. The tempfile we pass to create_tempfile() ends up on a global linked list, and it's not safe for it to go out of scope until we've called one of those two deactivation functions. Imagine that we add an early return from the function that forgets to call delete_tempfile(). With a static or heap tempfile variable, the worst case is that the tempfile hangs around until the program exits (and some functions like setup_shallow_temporary rely on this intentionally, creating a tempfile and then leaving it for later cleanup). But with a stack variable as above, this is a serious memory error: the variable goes out of scope and may be filled with garbage by the time the tempfile code looks at it. Let's see if we can make it harder to get this wrong. Since many callers need to allocate arbitrary numbers of tempfiles, we can't rely on static storage as a general solution. So we need to turn to the heap. We could just ask all callers to pass us a heap variable, but that puts the burden on them to call free() at the right time. Instead, let's have the tempfile code handle the heap allocation _and_ the deallocation (when the tempfile is deactivated and removed from the list). This changes the return value of all of the creation functions. For the cleanup functions (delete and rename), we'll add one extra bit of safety: instead of taking a tempfile pointer, we'll take a pointer-to-pointer and set it to NULL after freeing the object. This makes it safe to double-call functions like delete_tempfile(), as the second call treats the NULL input as a noop. Several callsites follow this pattern. The resulting patch does have a fair bit of noise, as each caller needs to be converted to handle: 1. Storing a pointer instead of the struct itself. 2. Passing the pointer instead of taking the struct address. 3. Handling a "struct tempfile *" return instead of a file descriptor. We could play games to make this less noisy. For example, by defining the tempfile like this: struct tempfile { struct heap_allocated_part_of_tempfile { int fd; ...etc } *actual_data; } Callers would continue to have a "struct tempfile", and it would be "active" only when the inner pointer was non-NULL. But that just makes things more awkward in the long run. There aren't that many callers, so we can simply bite the bullet and adjust all of them. And the compiler makes it easy for us to find them all. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-09-05 15:15:08 +03:00
static struct tempfile *pidfile;
static struct lock_file log_lock;
static struct string_list pack_garbage = STRING_LIST_INIT_DUP;
static void clean_pack_garbage(void)
{
int i;
for (i = 0; i < pack_garbage.nr; i++)
unlink_or_warn(pack_garbage.items[i].string);
string_list_clear(&pack_garbage, 0);
}
static void report_pack_garbage(unsigned seen_bits, const char *path)
{
if (seen_bits == PACKDIR_FILE_IDX)
string_list_append(&pack_garbage, path);
}
static void process_log_file(void)
{
struct stat st;
if (fstat(get_lock_file_fd(&log_lock), &st)) {
/*
* Perhaps there was an i/o error or another
* unlikely situation. Try to make a note of
* this in gc.log along with any existing
* messages.
*/
int saved_errno = errno;
fprintf(stderr, _("Failed to fstat %s: %s"),
tempfile: auto-allocate tempfiles on heap The previous commit taught the tempfile code to give up ownership over tempfiles that have been renamed or deleted. That makes it possible to use a stack variable like this: struct tempfile t; create_tempfile(&t, ...); ... if (!err) rename_tempfile(&t, ...); else delete_tempfile(&t); But doing it this way has a high potential for creating memory errors. The tempfile we pass to create_tempfile() ends up on a global linked list, and it's not safe for it to go out of scope until we've called one of those two deactivation functions. Imagine that we add an early return from the function that forgets to call delete_tempfile(). With a static or heap tempfile variable, the worst case is that the tempfile hangs around until the program exits (and some functions like setup_shallow_temporary rely on this intentionally, creating a tempfile and then leaving it for later cleanup). But with a stack variable as above, this is a serious memory error: the variable goes out of scope and may be filled with garbage by the time the tempfile code looks at it. Let's see if we can make it harder to get this wrong. Since many callers need to allocate arbitrary numbers of tempfiles, we can't rely on static storage as a general solution. So we need to turn to the heap. We could just ask all callers to pass us a heap variable, but that puts the burden on them to call free() at the right time. Instead, let's have the tempfile code handle the heap allocation _and_ the deallocation (when the tempfile is deactivated and removed from the list). This changes the return value of all of the creation functions. For the cleanup functions (delete and rename), we'll add one extra bit of safety: instead of taking a tempfile pointer, we'll take a pointer-to-pointer and set it to NULL after freeing the object. This makes it safe to double-call functions like delete_tempfile(), as the second call treats the NULL input as a noop. Several callsites follow this pattern. The resulting patch does have a fair bit of noise, as each caller needs to be converted to handle: 1. Storing a pointer instead of the struct itself. 2. Passing the pointer instead of taking the struct address. 3. Handling a "struct tempfile *" return instead of a file descriptor. We could play games to make this less noisy. For example, by defining the tempfile like this: struct tempfile { struct heap_allocated_part_of_tempfile { int fd; ...etc } *actual_data; } Callers would continue to have a "struct tempfile", and it would be "active" only when the inner pointer was non-NULL. But that just makes things more awkward in the long run. There aren't that many callers, so we can simply bite the bullet and adjust all of them. And the compiler makes it easy for us to find them all. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-09-05 15:15:08 +03:00
get_tempfile_path(log_lock.tempfile),
strerror(saved_errno));
fflush(stderr);
commit_lock_file(&log_lock);
errno = saved_errno;
} else if (st.st_size) {
/* There was some error recorded in the lock file */
commit_lock_file(&log_lock);
} else {
/* No error, clean up any old gc.log */
unlink(git_path("gc.log"));
rollback_lock_file(&log_lock);
}
}
static void process_log_file_at_exit(void)
{
fflush(stderr);
process_log_file();
}
static void process_log_file_on_signal(int signo)
{
process_log_file();
sigchain_pop(signo);
raise(signo);
}
gc: handle & check gc.reflogExpire config Don't redundantly run "git reflog expire --all" when gc.reflogExpire and gc.reflogExpireUnreachable are set to "never", and die immediately if those configuration valuer are bad. As an earlier "assert lack of early exit" change to the tests for "git reflog expire" shows, an early check of gc.reflogExpire{Unreachable,} isn't wanted in general for "git reflog expire", but it makes sense for "gc" because: 1) Similarly to 8ab5aa4bd8 ("parseopt: handle malformed --expire arguments more nicely", 2018-04-21) we'll now die early if the config variables are set to invalid values. We run "pack-refs" before "reflog expire", which can take a while, only to then die on an invalid gc.reflogExpire{Unreachable,} configuration. 2) Not invoking the command at all means it won't show up in trace output, which makes what's going on more obvious when the two are set to "never". 3) As a later change documents we lock the refs when looping over the refs to expire, even in cases where we end up doing nothing due to this config. For the reasons noted in the earlier "assert lack of early exit" change I don't think it's worth it to bend over backwards in "git reflog expire" itself to carefully detect if we'll really do nothing given the combination of all its possible options and skip that locking, but that's easy to detect here in "gc" where we'll only run "reflog expire" in a relatively simple mode. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-03-28 19:14:34 +03:00
static int gc_config_is_timestamp_never(const char *var)
{
const char *value;
timestamp_t expire;
if (!git_config_get_value(var, &value) && value) {
if (parse_expiry_date(value, &expire))
die(_("failed to parse '%s' value '%s'"), var, value);
return expire == 0;
}
return 0;
}
static void gc_config(void)
{
const char *value;
if (!git_config_get_value("gc.packrefs", &value)) {
if (value && !strcmp(value, "notbare"))
pack_refs = -1;
else
pack_refs = git_config_bool("gc.packrefs", value);
}
gc: handle & check gc.reflogExpire config Don't redundantly run "git reflog expire --all" when gc.reflogExpire and gc.reflogExpireUnreachable are set to "never", and die immediately if those configuration valuer are bad. As an earlier "assert lack of early exit" change to the tests for "git reflog expire" shows, an early check of gc.reflogExpire{Unreachable,} isn't wanted in general for "git reflog expire", but it makes sense for "gc" because: 1) Similarly to 8ab5aa4bd8 ("parseopt: handle malformed --expire arguments more nicely", 2018-04-21) we'll now die early if the config variables are set to invalid values. We run "pack-refs" before "reflog expire", which can take a while, only to then die on an invalid gc.reflogExpire{Unreachable,} configuration. 2) Not invoking the command at all means it won't show up in trace output, which makes what's going on more obvious when the two are set to "never". 3) As a later change documents we lock the refs when looping over the refs to expire, even in cases where we end up doing nothing due to this config. For the reasons noted in the earlier "assert lack of early exit" change I don't think it's worth it to bend over backwards in "git reflog expire" itself to carefully detect if we'll really do nothing given the combination of all its possible options and skip that locking, but that's easy to detect here in "gc" where we'll only run "reflog expire" in a relatively simple mode. Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-03-28 19:14:34 +03:00
if (gc_config_is_timestamp_never("gc.reflogexpire") &&
gc_config_is_timestamp_never("gc.reflogexpireunreachable"))
prune_reflogs = 0;
git_config_get_int("gc.aggressivewindow", &aggressive_window);
git_config_get_int("gc.aggressivedepth", &aggressive_depth);
git_config_get_int("gc.auto", &gc_auto_threshold);
git_config_get_int("gc.autopacklimit", &gc_auto_pack_limit);
git_config_get_bool("gc.autodetach", &detach_auto);
git_config_get_expiry("gc.pruneexpire", &prune_expire);
git_config_get_expiry("gc.worktreepruneexpire", &prune_worktrees_expire);
git_config_get_expiry("gc.logexpiry", &gc_log_expire);
git_config_get_ulong("gc.bigpackthreshold", &big_pack_threshold);
git_config_get_ulong("pack.deltacachesize", &max_delta_cache_size);
git_config(git_default_config, NULL);
}
static int too_many_loose_objects(void)
{
/*
* Quickly check if a "gc" is needed, by estimating how
* many loose objects there are. Because SHA-1 is evenly
* distributed, we can check only one and get a reasonable
* estimate.
*/
DIR *dir;
struct dirent *ent;
int auto_threshold;
int num_loose = 0;
int needed = 0;
gc: convert to using the_hash_algo There's been a lot of changing of the hardcoded "40" values to the_hash_algo->hexsz, but we've so far missed this one where we hardcoded 38 for the loose object file length. This is because a SHA-1 like abcde[...] gets turned into objects/ab/cde[...]. There's no reason to suppose the same won't be the case for SHA-256, and reading between the lines in hash-function-transition.txt the format is planned to be the same. In the future we may want to further modify this code for the hash function transition. There's a potential pathological case here where we'll only consider the loose objects for the currently active hash, but objects for that hash will share a directory storage with the other hash. Thus we could theoretically have e.g. 1k SHA-1 loose objects, and 1 million SHA-256 objects. Then not notice that we need to pack them because we're currently using SHA-1, even though our FS may be straining under the stress of such humongous directories. So assuming that "gc" eventually learns to pack up both SHA-1 and SHA-256 objects regardless of what the current the_hash_algo is, perhaps this check should be changed to consider all files in objects/17/ matching [0-9a-f] 38 or 62 characters in length (i.e. both SHA-1 and SHA-256). But none of that is something we need to worry about now, and supporting both 38 and 62 characters depending on "the_hash_algo" removes another case of SHA-1 hardcoding. As noted in [1] I'm making no effort to somehow remove the hardcoding for "2" as in "use the first two hexdigits for the directory name". There's no indication that that'll ever change, and somehow generalizing it here would be a drop in the ocean, so there's no point in doing that. It also couldn't be done without coming up with some generalized version of the magical "objects/17" directory. See [2] for a discussion of that directory. 1. https://public-inbox.org/git/874l84ber7.fsf@evledraar.gmail.com/ 2. https://public-inbox.org/git/87k1mta9x5.fsf@evledraar.gmail.com/ Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-03-15 18:59:53 +03:00
const unsigned hexsz_loose = the_hash_algo->hexsz - 2;
dir = opendir(git_path("objects/17"));
if (!dir)
return 0;
auto_threshold = DIV_ROUND_UP(gc_auto_threshold, 256);
while ((ent = readdir(dir)) != NULL) {
gc: convert to using the_hash_algo There's been a lot of changing of the hardcoded "40" values to the_hash_algo->hexsz, but we've so far missed this one where we hardcoded 38 for the loose object file length. This is because a SHA-1 like abcde[...] gets turned into objects/ab/cde[...]. There's no reason to suppose the same won't be the case for SHA-256, and reading between the lines in hash-function-transition.txt the format is planned to be the same. In the future we may want to further modify this code for the hash function transition. There's a potential pathological case here where we'll only consider the loose objects for the currently active hash, but objects for that hash will share a directory storage with the other hash. Thus we could theoretically have e.g. 1k SHA-1 loose objects, and 1 million SHA-256 objects. Then not notice that we need to pack them because we're currently using SHA-1, even though our FS may be straining under the stress of such humongous directories. So assuming that "gc" eventually learns to pack up both SHA-1 and SHA-256 objects regardless of what the current the_hash_algo is, perhaps this check should be changed to consider all files in objects/17/ matching [0-9a-f] 38 or 62 characters in length (i.e. both SHA-1 and SHA-256). But none of that is something we need to worry about now, and supporting both 38 and 62 characters depending on "the_hash_algo" removes another case of SHA-1 hardcoding. As noted in [1] I'm making no effort to somehow remove the hardcoding for "2" as in "use the first two hexdigits for the directory name". There's no indication that that'll ever change, and somehow generalizing it here would be a drop in the ocean, so there's no point in doing that. It also couldn't be done without coming up with some generalized version of the magical "objects/17" directory. See [2] for a discussion of that directory. 1. https://public-inbox.org/git/874l84ber7.fsf@evledraar.gmail.com/ 2. https://public-inbox.org/git/87k1mta9x5.fsf@evledraar.gmail.com/ Signed-off-by: Ævar Arnfjörð Bjarmason <avarab@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2019-03-15 18:59:53 +03:00
if (strspn(ent->d_name, "0123456789abcdef") != hexsz_loose ||
ent->d_name[hexsz_loose] != '\0')
continue;
if (++num_loose > auto_threshold) {
needed = 1;
break;
}
}
closedir(dir);
return needed;
}
static struct packed_git *find_base_packs(struct string_list *packs,
unsigned long limit)
{
struct packed_git *p, *base = NULL;
for (p = get_all_packs(the_repository); p; p = p->next) {
if (!p->pack_local)
continue;
if (limit) {
if (p->pack_size >= limit)
string_list_append(packs, p->pack_name);
} else if (!base || base->pack_size < p->pack_size) {
base = p;
}
}
if (base)
string_list_append(packs, base->pack_name);
return base;
}
static int too_many_packs(void)
{
struct packed_git *p;
int cnt;
if (gc_auto_pack_limit <= 0)
return 0;
for (cnt = 0, p = get_all_packs(the_repository); p; p = p->next) {
if (!p->pack_local)
continue;
if (p->pack_keep)
continue;
/*
* Perhaps check the size of the pack and count only
* very small ones here?
*/
cnt++;
}
return gc_auto_pack_limit < cnt;
}
static uint64_t total_ram(void)
{
#if defined(HAVE_SYSINFO)
struct sysinfo si;
if (!sysinfo(&si))
return si.totalram;
#elif defined(HAVE_BSD_SYSCTL) && (defined(HW_MEMSIZE) || defined(HW_PHYSMEM))
int64_t physical_memory;
int mib[2];
size_t length;
mib[0] = CTL_HW;
# if defined(HW_MEMSIZE)
mib[1] = HW_MEMSIZE;
# else
mib[1] = HW_PHYSMEM;
# endif
length = sizeof(int64_t);
if (!sysctl(mib, 2, &physical_memory, &length, NULL, 0))
return physical_memory;
#elif defined(GIT_WINDOWS_NATIVE)
MEMORYSTATUSEX memInfo;
memInfo.dwLength = sizeof(MEMORYSTATUSEX);
if (GlobalMemoryStatusEx(&memInfo))
return memInfo.ullTotalPhys;
#endif
return 0;
}
static uint64_t estimate_repack_memory(struct packed_git *pack)
{
unsigned long nr_objects = approximate_object_count();
size_t os_cache, heap;
if (!pack || !nr_objects)
return 0;
/*
* First we have to scan through at least one pack.
* Assume enough room in OS file cache to keep the entire pack
* or we may accidentally evict data of other processes from
* the cache.
*/
os_cache = pack->pack_size + pack->index_size;
/* then pack-objects needs lots more for book keeping */
heap = sizeof(struct object_entry) * nr_objects;
/*
* internal rev-list --all --objects takes up some memory too,
* let's say half of it is for blobs
*/
heap += sizeof(struct blob) * nr_objects / 2;
/*
* and the other half is for trees (commits and tags are
* usually insignificant)
*/
heap += sizeof(struct tree) * nr_objects / 2;
/* and then obj_hash[], underestimated in fact */
heap += sizeof(struct object *) * nr_objects;
/* revindex is used also */
heap += sizeof(struct revindex_entry) * nr_objects;
/*
* read_sha1_file() (either at delta calculation phase, or
* writing phase) also fills up the delta base cache
*/
heap += delta_base_cache_limit;
/* and of course pack-objects has its own delta cache */
heap += max_delta_cache_size;
return os_cache + heap;
}
static int keep_one_pack(struct string_list_item *item, void *data)
{
strvec_pushf(&repack, "--keep-pack=%s", basename(item->string));
return 0;
}
static void add_repack_all_option(struct string_list *keep_pack)
{
if (prune_expire && !strcmp(prune_expire, "now"))
strvec_push(&repack, "-a");
else {
strvec_push(&repack, "-A");
if (prune_expire)
strvec_pushf(&repack, "--unpack-unreachable=%s", prune_expire);
}
if (keep_pack)
for_each_string_list(keep_pack, keep_one_pack, NULL);
}
static void add_repack_incremental_option(void)
{
strvec_push(&repack, "--no-write-bitmap-index");
}
static int need_to_gc(void)
{
/*
* Setting gc.auto to 0 or negative can disable the
* automatic gc.
*/
if (gc_auto_threshold <= 0)
return 0;
/*
* If there are too many loose objects, but not too many
* packs, we run "repack -d -l". If there are too many packs,
* we run "repack -A -d -l". Otherwise we tell the caller
* there is no need.
*/
if (too_many_packs()) {
struct string_list keep_pack = STRING_LIST_INIT_NODUP;
if (big_pack_threshold) {
find_base_packs(&keep_pack, big_pack_threshold);
if (keep_pack.nr >= gc_auto_pack_limit) {
big_pack_threshold = 0;
string_list_clear(&keep_pack, 0);
find_base_packs(&keep_pack, 0);
}
} else {
struct packed_git *p = find_base_packs(&keep_pack, 0);
uint64_t mem_have, mem_want;
mem_have = total_ram();
mem_want = estimate_repack_memory(p);
/*
* Only allow 1/2 of memory for pack-objects, leave
* the rest for the OS and other processes in the
* system.
*/
if (!mem_have || mem_want < mem_have / 2)
string_list_clear(&keep_pack, 0);
}
add_repack_all_option(&keep_pack);
string_list_clear(&keep_pack, 0);
} else if (too_many_loose_objects())
add_repack_incremental_option();
else
return 0;
if (run_hook_le(NULL, "pre-auto-gc", NULL))
return 0;
return 1;
}
/* return NULL on success, else hostname running the gc */
static const char *lock_repo_for_gc(int force, pid_t* ret_pid)
{
struct lock_file lock = LOCK_INIT;
char my_host[HOST_NAME_MAX + 1];
struct strbuf sb = STRBUF_INIT;
struct stat st;
uintmax_t pid;
FILE *fp;
int fd;
char *pidfile_path;
tempfile: auto-allocate tempfiles on heap The previous commit taught the tempfile code to give up ownership over tempfiles that have been renamed or deleted. That makes it possible to use a stack variable like this: struct tempfile t; create_tempfile(&t, ...); ... if (!err) rename_tempfile(&t, ...); else delete_tempfile(&t); But doing it this way has a high potential for creating memory errors. The tempfile we pass to create_tempfile() ends up on a global linked list, and it's not safe for it to go out of scope until we've called one of those two deactivation functions. Imagine that we add an early return from the function that forgets to call delete_tempfile(). With a static or heap tempfile variable, the worst case is that the tempfile hangs around until the program exits (and some functions like setup_shallow_temporary rely on this intentionally, creating a tempfile and then leaving it for later cleanup). But with a stack variable as above, this is a serious memory error: the variable goes out of scope and may be filled with garbage by the time the tempfile code looks at it. Let's see if we can make it harder to get this wrong. Since many callers need to allocate arbitrary numbers of tempfiles, we can't rely on static storage as a general solution. So we need to turn to the heap. We could just ask all callers to pass us a heap variable, but that puts the burden on them to call free() at the right time. Instead, let's have the tempfile code handle the heap allocation _and_ the deallocation (when the tempfile is deactivated and removed from the list). This changes the return value of all of the creation functions. For the cleanup functions (delete and rename), we'll add one extra bit of safety: instead of taking a tempfile pointer, we'll take a pointer-to-pointer and set it to NULL after freeing the object. This makes it safe to double-call functions like delete_tempfile(), as the second call treats the NULL input as a noop. Several callsites follow this pattern. The resulting patch does have a fair bit of noise, as each caller needs to be converted to handle: 1. Storing a pointer instead of the struct itself. 2. Passing the pointer instead of taking the struct address. 3. Handling a "struct tempfile *" return instead of a file descriptor. We could play games to make this less noisy. For example, by defining the tempfile like this: struct tempfile { struct heap_allocated_part_of_tempfile { int fd; ...etc } *actual_data; } Callers would continue to have a "struct tempfile", and it would be "active" only when the inner pointer was non-NULL. But that just makes things more awkward in the long run. There aren't that many callers, so we can simply bite the bullet and adjust all of them. And the compiler makes it easy for us to find them all. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-09-05 15:15:08 +03:00
if (is_tempfile_active(pidfile))
/* already locked */
return NULL;
if (xgethostname(my_host, sizeof(my_host)))
xsnprintf(my_host, sizeof(my_host), "unknown");
pidfile_path = git_pathdup("gc.pid");
fd = hold_lock_file_for_update(&lock, pidfile_path,
LOCK_DIE_ON_ERROR);
if (!force) {
static char locking_host[HOST_NAME_MAX + 1];
static char *scan_fmt;
int should_exit;
if (!scan_fmt)
scan_fmt = xstrfmt("%s %%%ds", "%"SCNuMAX, HOST_NAME_MAX);
fp = fopen(pidfile_path, "r");
memset(locking_host, 0, sizeof(locking_host));
should_exit =
fp != NULL &&
!fstat(fileno(fp), &st) &&
/*
* 12 hour limit is very generous as gc should
* never take that long. On the other hand we
* don't really need a strict limit here,
* running gc --auto one day late is not a big
* problem. --force can be used in manual gc
* after the user verifies that no gc is
* running.
*/
time(NULL) - st.st_mtime <= 12 * 3600 &&
fscanf(fp, scan_fmt, &pid, locking_host) == 2 &&
/* be gentle to concurrent "gc" on remote hosts */
(strcmp(locking_host, my_host) || !kill(pid, 0) || errno == EPERM);
if (fp != NULL)
fclose(fp);
if (should_exit) {
if (fd >= 0)
rollback_lock_file(&lock);
*ret_pid = pid;
free(pidfile_path);
return locking_host;
}
}
strbuf_addf(&sb, "%"PRIuMAX" %s",
(uintmax_t) getpid(), my_host);
write_in_full(fd, sb.buf, sb.len);
strbuf_release(&sb);
commit_lock_file(&lock);
tempfile: auto-allocate tempfiles on heap The previous commit taught the tempfile code to give up ownership over tempfiles that have been renamed or deleted. That makes it possible to use a stack variable like this: struct tempfile t; create_tempfile(&t, ...); ... if (!err) rename_tempfile(&t, ...); else delete_tempfile(&t); But doing it this way has a high potential for creating memory errors. The tempfile we pass to create_tempfile() ends up on a global linked list, and it's not safe for it to go out of scope until we've called one of those two deactivation functions. Imagine that we add an early return from the function that forgets to call delete_tempfile(). With a static or heap tempfile variable, the worst case is that the tempfile hangs around until the program exits (and some functions like setup_shallow_temporary rely on this intentionally, creating a tempfile and then leaving it for later cleanup). But with a stack variable as above, this is a serious memory error: the variable goes out of scope and may be filled with garbage by the time the tempfile code looks at it. Let's see if we can make it harder to get this wrong. Since many callers need to allocate arbitrary numbers of tempfiles, we can't rely on static storage as a general solution. So we need to turn to the heap. We could just ask all callers to pass us a heap variable, but that puts the burden on them to call free() at the right time. Instead, let's have the tempfile code handle the heap allocation _and_ the deallocation (when the tempfile is deactivated and removed from the list). This changes the return value of all of the creation functions. For the cleanup functions (delete and rename), we'll add one extra bit of safety: instead of taking a tempfile pointer, we'll take a pointer-to-pointer and set it to NULL after freeing the object. This makes it safe to double-call functions like delete_tempfile(), as the second call treats the NULL input as a noop. Several callsites follow this pattern. The resulting patch does have a fair bit of noise, as each caller needs to be converted to handle: 1. Storing a pointer instead of the struct itself. 2. Passing the pointer instead of taking the struct address. 3. Handling a "struct tempfile *" return instead of a file descriptor. We could play games to make this less noisy. For example, by defining the tempfile like this: struct tempfile { struct heap_allocated_part_of_tempfile { int fd; ...etc } *actual_data; } Callers would continue to have a "struct tempfile", and it would be "active" only when the inner pointer was non-NULL. But that just makes things more awkward in the long run. There aren't that many callers, so we can simply bite the bullet and adjust all of them. And the compiler makes it easy for us to find them all. Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-09-05 15:15:08 +03:00
pidfile = register_tempfile(pidfile_path);
free(pidfile_path);
return NULL;
}
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
/*
* Returns 0 if there was no previous error and gc can proceed, 1 if
* gc should not proceed due to an error in the last run. Prints a
* message and returns -1 if an error occurred while reading gc.log
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
*/
static int report_last_gc_error(void)
{
struct strbuf sb = STRBUF_INIT;
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
int ret = 0;
ssize_t len;
struct stat st;
char *gc_log_path = git_pathdup("gc.log");
if (stat(gc_log_path, &st)) {
if (errno == ENOENT)
goto done;
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
ret = error_errno(_("cannot stat '%s'"), gc_log_path);
goto done;
}
if (st.st_mtime < gc_log_expire_time)
goto done;
len = strbuf_read_file(&sb, gc_log_path, 0);
if (len < 0)
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
ret = error_errno(_("cannot read '%s'"), gc_log_path);
else if (len > 0) {
/*
* A previous gc failed. Report the error, and don't
* bother with an automatic gc run since it is likely
* to fail in the same way.
*/
warning(_("The last gc run reported the following. "
"Please correct the root cause\n"
"and remove %s.\n"
"Automatic cleanup will not be performed "
"until the file is removed.\n\n"
"%s"),
gc_log_path, sb.buf);
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
ret = 1;
}
strbuf_release(&sb);
done:
free(gc_log_path);
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
return ret;
}
static void gc_before_repack(void)
{
/*
* We may be called twice, as both the pre- and
* post-daemonized phases will call us, but running these
* commands more than once is pointless and wasteful.
*/
static int done = 0;
if (done++)
return;
if (pack_refs && run_command_v_opt(pack_refs_cmd.v, RUN_GIT_CMD))
die(FAILED_RUN, pack_refs_cmd.v[0]);
if (prune_reflogs && run_command_v_opt(reflog.v, RUN_GIT_CMD))
die(FAILED_RUN, reflog.v[0]);
}
int cmd_gc(int argc, const char **argv, const char *prefix)
{
int aggressive = 0;
int auto_gc = 0;
int quiet = 0;
int force = 0;
const char *name;
pid_t pid;
int daemonized = 0;
int keep_largest_pack = -1;
2018-04-21 06:13:13 +03:00
timestamp_t dummy;
struct option builtin_gc_options[] = {
OPT__QUIET(&quiet, N_("suppress progress reporting")),
{ OPTION_STRING, 0, "prune", &prune_expire, N_("date"),
N_("prune unreferenced objects"),
PARSE_OPT_OPTARG, NULL, (intptr_t)prune_expire },
OPT_BOOL(0, "aggressive", &aggressive, N_("be more thorough (increased runtime)")),
OPT_BOOL_F(0, "auto", &auto_gc, N_("enable auto-gc mode"),
PARSE_OPT_NOCOMPLETE),
OPT_BOOL_F(0, "force", &force,
N_("force running gc even if there may be another gc running"),
PARSE_OPT_NOCOMPLETE),
OPT_BOOL(0, "keep-largest-pack", &keep_largest_pack,
N_("repack all other packs except the largest pack")),
OPT_END()
};
if (argc == 2 && !strcmp(argv[1], "-h"))
usage_with_options(builtin_gc_usage, builtin_gc_options);
strvec_pushl(&pack_refs_cmd, "pack-refs", "--all", "--prune", NULL);
strvec_pushl(&reflog, "reflog", "expire", "--all", NULL);
strvec_pushl(&repack, "repack", "-d", "-l", NULL);
strvec_pushl(&prune, "prune", "--expire", NULL);
strvec_pushl(&prune_worktrees, "worktree", "prune", "--expire", NULL);
strvec_pushl(&rerere, "rerere", "gc", NULL);
/* default expiry time, overwritten in gc_config */
gc_config();
if (parse_expiry_date(gc_log_expire, &gc_log_expire_time))
die(_("failed to parse gc.logexpiry value %s"), gc_log_expire);
if (pack_refs < 0)
pack_refs = !is_bare_repository();
argc = parse_options(argc, argv, prefix, builtin_gc_options,
builtin_gc_usage, 0);
if (argc > 0)
usage_with_options(builtin_gc_usage, builtin_gc_options);
2018-04-21 06:13:13 +03:00
if (prune_expire && parse_expiry_date(prune_expire, &dummy))
die(_("failed to parse prune expiry value %s"), prune_expire);
if (aggressive) {
strvec_push(&repack, "-f");
gc --aggressive: make --depth configurable When 1c192f3 (gc --aggressive: make it really aggressive - 2007-12-06) made --depth=250 the default value, it didn't really explain the reason behind, especially the pros and cons of --depth=250. An old mail from Linus below explains it at length. Long story short, --depth=250 is a disk saver and a performance killer. Not everybody agrees on that aggressiveness. Let the user configure it. From: Linus Torvalds <torvalds@linux-foundation.org> Subject: Re: [PATCH] gc --aggressive: make it really aggressive Date: Thu, 6 Dec 2007 08:19:24 -0800 (PST) Message-ID: <alpine.LFD.0.9999.0712060803430.13796@woody.linux-foundation.org> Gmane-URL: http://article.gmane.org/gmane.comp.gcc.devel/94637 On Thu, 6 Dec 2007, Harvey Harrison wrote: > > 7:41:25elapsed 86%CPU Heh. And this is why you want to do it exactly *once*, and then just export the end result for others ;) > -r--r--r-- 1 hharrison hharrison 324094684 2007-12-06 07:26 pack-1d46...pack But yeah, especially if you allow longer delta chains, the end result can be much smaller (and what makes the one-time repack more expensive is the window size, not the delta chain - you could make the delta chains longer with no cost overhead at packing time) HOWEVER. The longer delta chains do make it potentially much more expensive to then use old history. So there's a trade-off. And quite frankly, a delta depth of 250 is likely going to cause overflows in the delta cache (which is only 256 entries in size *and* it's a hash, so it's going to start having hash conflicts long before hitting the 250 depth limit). So when I said "--depth=250 --window=250", I chose those numbers more as an example of extremely aggressive packing, and I'm not at all sure that the end result is necessarily wonderfully usable. It's going to save disk space (and network bandwidth - the delta's will be re-used for the network protocol too!), but there are definitely downsides too, and using long delta chains may simply not be worth it in practice. (And some of it might just want to have git tuning, ie if people think that long deltas are worth it, we could easily just expand on the delta hash, at the cost of some more memory used!) That said, the good news is that working with *new* history will not be affected negatively, and if you want to be _really_ sneaky, there are ways to say "create a pack that contains the history up to a version one year ago, and be very aggressive about those old versions that we still want to have around, but do a separate pack for newer stuff using less aggressive parameters" So this is something that can be tweaked, although we don't really have any really nice interfaces for stuff like that (ie the git delta cache size is hardcoded in the sources and cannot be set in the config file, and the "pack old history more aggressively" involves some manual scripting and knowing how "git pack-objects" works rather than any nice simple command line switch). So the thing to take away from this is: - git is certainly flexible as hell - .. but to get the full power you may need to tweak things - .. happily you really only need to have one person to do the tweaking, and the tweaked end results will be available to others that do not need to know/care. And whether the difference between 320MB and 500MB is worth any really involved tweaking (considering the potential downsides), I really don't know. Only testing will tell. Linus Signed-off-by: Nguyễn Thái Ngọc Duy <pclouds@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2014-03-16 17:35:03 +04:00
if (aggressive_depth > 0)
strvec_pushf(&repack, "--depth=%d", aggressive_depth);
if (aggressive_window > 0)
strvec_pushf(&repack, "--window=%d", aggressive_window);
}
if (quiet)
strvec_push(&repack, "-q");
if (auto_gc) {
/*
* Auto-gc should be least intrusive as possible.
*/
if (!need_to_gc())
return 0;
if (!quiet) {
if (detach_auto)
fprintf(stderr, _("Auto packing the repository in background for optimum performance.\n"));
else
fprintf(stderr, _("Auto packing the repository for optimum performance.\n"));
fprintf(stderr, _("See \"git help gc\" for manual housekeeping.\n"));
}
if (detach_auto) {
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
int ret = report_last_gc_error();
if (ret < 0)
/* an I/O error occurred, already reported */
gc: do not return error for prior errors in daemonized mode Some build machines started consistently failing to fetch updated source using "repo sync", with error error: The last gc run reported the following. Please correct the root cause and remove /build/.repo/projects/tools/git.git/gc.log. Automatic cleanup will not be performed until the file is removed. warning: There are too many unreachable loose objects; run 'git prune' to remove them. The cause takes some time to describe. In v2.0.0-rc0~145^2 (gc: config option for running --auto in background, 2014-02-08), "git gc --auto" learned to run in the background instead of blocking the invoking command. In this mode, it closed stderr to avoid interleaving output with any subsequent commands, causing warnings like the above to be swallowed; v2.6.3~24^2 (gc: save log from daemonized gc --auto and print it next time, 2015-09-19) addressed that by storing any diagnostic output in .git/gc.log and allowing the next "git gc --auto" run to print it. To avoid wasteful repeated fruitless gcs, when gc.log is present, the subsequent "gc --auto" would die after printing its contents. Most git commands, such as "git fetch", ignore the exit status from "git gc --auto" so all is well at this point: the user gets to see the error message, and the fetch succeeds, without a wasteful additional attempt at an automatic gc. External tools like repo[1], though, do care about the exit status from "git gc --auto". In non-daemonized mode, the exit status is straightforward: if there is an error, it is nonzero, but after a warning like the above, the status is zero. The daemonized mode, as a side effect of the other properties provided, offers a very strange exit code convention: - if no housekeeping was required, the exit status is 0 - the first real run, after forking into the background, returns exit status 0 unconditionally. The parent process has no way to know whether gc will succeed. - if there is any diagnostic output in gc.log, subsequent runs return a nonzero exit status to indicate that gc was not triggered. There's nothing for the calling program to act on on the basis of that error. Use status 0 consistently instead, to indicate that we decided not to run a gc (just like if no housekeeping was required). This way, repo and similar tools can get the benefit of the same behavior as tools like "git fetch" that ignore the exit status from gc --auto. Once the period of time described by gc.pruneExpire elapses, the unreachable loose objects will be removed by "git gc --auto" automatically. [1] https://gerrit-review.googlesource.com/c/git-repo/+/10598/ Reported-by: Andrii Dehtiarov <adehtiarov@google.com> Helped-by: Jeff King <peff@peff.net> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2018-07-17 09:57:40 +03:00
exit(128);
if (ret == 1)
/* Last gc --auto failed. Skip this one. */
return 0;
gc: run pre-detach operations under lock We normally try to avoid having two auto-gc operations run at the same time, because it wastes resources. This was done long ago in 64a99eb47 (gc: reject if another gc is running, unless --force is given, 2013-08-08). When we do a detached auto-gc, we run the ref-related commands _before_ detaching, to avoid confusing lock contention. This was done by 62aad1849 (gc --auto: do not lock refs in the background, 2014-05-25). These two features do not interact well. The pre-detach operations are run before we check the gc.pid lock, meaning that on a busy repository we may run many of them concurrently. Ideally we'd take the lock before spawning any operations, and hold it for the duration of the program. This is tricky, though, with the way the pid-file interacts with the daemonize() process. Other processes will check that the pid recorded in the pid-file still exists. But detaching causes us to fork and continue running under a new pid. So if we take the lock before detaching, the pid-file will have a bogus pid in it. We'd have to go back and update it with the new pid after detaching. We'd also have to play some tricks with the tempfile subsystem to tweak the "owner" field, so that the parent process does not clean it up on exit, but the child process does. Instead, we can do something a bit simpler: take the lock only for the duration of the pre-detach work, then detach, then take it again for the post-detach work. Technically, this means that the post-detach lock could lose to another process doing pre-detach work. But in the long run this works out. That second process would then follow-up by doing post-detach work. Unless it was in turn blocked by a third process doing pre-detach work, and so on. This could in theory go on indefinitely, as the pre-detach work does not repack, and so need_to_gc() will continue to trigger. But in each round we are racing between the pre- and post-detach locks. Eventually, one of the post-detach locks will win the race and complete the full gc. So in the worst case, we may racily repeat the pre-detach work, but we would never do so simultaneously (it would happen via a sequence of serialized race-wins). Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-07-11 12:06:35 +03:00
if (lock_repo_for_gc(force, &pid))
return 0;
gc_before_repack(); /* dies on failure */
gc: run pre-detach operations under lock We normally try to avoid having two auto-gc operations run at the same time, because it wastes resources. This was done long ago in 64a99eb47 (gc: reject if another gc is running, unless --force is given, 2013-08-08). When we do a detached auto-gc, we run the ref-related commands _before_ detaching, to avoid confusing lock contention. This was done by 62aad1849 (gc --auto: do not lock refs in the background, 2014-05-25). These two features do not interact well. The pre-detach operations are run before we check the gc.pid lock, meaning that on a busy repository we may run many of them concurrently. Ideally we'd take the lock before spawning any operations, and hold it for the duration of the program. This is tricky, though, with the way the pid-file interacts with the daemonize() process. Other processes will check that the pid recorded in the pid-file still exists. But detaching causes us to fork and continue running under a new pid. So if we take the lock before detaching, the pid-file will have a bogus pid in it. We'd have to go back and update it with the new pid after detaching. We'd also have to play some tricks with the tempfile subsystem to tweak the "owner" field, so that the parent process does not clean it up on exit, but the child process does. Instead, we can do something a bit simpler: take the lock only for the duration of the pre-detach work, then detach, then take it again for the post-detach work. Technically, this means that the post-detach lock could lose to another process doing pre-detach work. But in the long run this works out. That second process would then follow-up by doing post-detach work. Unless it was in turn blocked by a third process doing pre-detach work, and so on. This could in theory go on indefinitely, as the pre-detach work does not repack, and so need_to_gc() will continue to trigger. But in each round we are racing between the pre- and post-detach locks. Eventually, one of the post-detach locks will win the race and complete the full gc. So in the worst case, we may racily repeat the pre-detach work, but we would never do so simultaneously (it would happen via a sequence of serialized race-wins). Signed-off-by: Jeff King <peff@peff.net> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2017-07-11 12:06:35 +03:00
delete_tempfile(&pidfile);
/*
* failure to daemonize is ok, we'll continue
* in foreground
*/
daemonized = !daemonize();
}
} else {
struct string_list keep_pack = STRING_LIST_INIT_NODUP;
if (keep_largest_pack != -1) {
if (keep_largest_pack)
find_base_packs(&keep_pack, 0);
} else if (big_pack_threshold) {
find_base_packs(&keep_pack, big_pack_threshold);
}
add_repack_all_option(&keep_pack);
string_list_clear(&keep_pack, 0);
}
name = lock_repo_for_gc(force, &pid);
if (name) {
if (auto_gc)
return 0; /* be quiet on --auto */
die(_("gc is already running on machine '%s' pid %"PRIuMAX" (use --force if not)"),
name, (uintmax_t)pid);
}
if (daemonized) {
hold_lock_file_for_update(&log_lock,
git_path("gc.log"),
LOCK_DIE_ON_ERROR);
dup2(get_lock_file_fd(&log_lock), 2);
sigchain_push_common(process_log_file_on_signal);
atexit(process_log_file_at_exit);
}
gc_before_repack();
if (!repository_format_precious_objects) {
close_object_store(the_repository->objects);
if (run_command_v_opt(repack.v, RUN_GIT_CMD))
die(FAILED_RUN, repack.v[0]);
if (prune_expire) {
strvec_push(&prune, prune_expire);
if (quiet)
strvec_push(&prune, "--no-progress");
if (has_promisor_remote())
strvec_push(&prune,
"--exclude-promisor-objects");
if (run_command_v_opt(prune.v, RUN_GIT_CMD))
die(FAILED_RUN, prune.v[0]);
}
}
if (prune_worktrees_expire) {
strvec_push(&prune_worktrees, prune_worktrees_expire);
if (run_command_v_opt(prune_worktrees.v, RUN_GIT_CMD))
die(FAILED_RUN, prune_worktrees.v[0]);
}
if (run_command_v_opt(rerere.v, RUN_GIT_CMD))
die(FAILED_RUN, rerere.v[0]);
report_garbage = report_pack_garbage;
reprepare_packed_git(the_repository);
if (pack_garbage.nr > 0) {
close_object_store(the_repository->objects);
clean_pack_garbage();
}
prepare_repo_settings(the_repository);
if (the_repository->settings.gc_write_commit_graph == 1)
commit-graph.h: store an odb in 'struct write_commit_graph_context' There are lots of places in 'commit-graph.h' where a function either has (or almost has) a full 'struct object_directory *', accesses '->path', and then throws away the rest of the struct. This can cause headaches when comparing the locations of object directories across alternates (e.g., in the case of deciding if two commit-graph layers can be merged). These paths are normalized with 'normalize_path_copy()' which mitigates some comparison issues, but not all [1]. Replace usage of 'char *object_dir' with 'odb->path' by storing a 'struct object_directory *' in the 'write_commit_graph_context' structure. This is an intermediate step towards getting rid of all path normalization in 'commit-graph.c'. Resolving a user-provided '--object-dir' argument now requires that we compare it to the known alternates for equality. Prior to this patch, an unknown '--object-dir' argument would silently exit with status zero. This can clearly lead to unintended behavior, such as verifying commit-graphs that aren't in a repository's own object store (or one of its alternates), or causing a typo to mask a legitimate commit-graph verification failure. Make this error non-silent by 'die()'-ing when the given '--object-dir' does not match any known alternate object store. [1]: In my testing, for example, I can get one side of the commit-graph code to fill object_dir with "./objects" and the other with just "objects". Signed-off-by: Taylor Blau <me@ttaylorr.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-02-04 08:51:50 +03:00
write_commit_graph_reachable(the_repository->objects->odb,
!quiet && !daemonized ? COMMIT_GRAPH_WRITE_PROGRESS : 0,
NULL);
if (auto_gc && too_many_loose_objects())
warning(_("There are too many unreachable loose objects; "
"run 'git prune' to remove them."));
if (!daemonized)
unlink(git_path("gc.log"));
return 0;
}
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
static const char *const builtin_maintenance_run_usage[] = {
N_("git maintenance run [--auto] [--[no-]quiet] [--task=<task>] [--schedule]"),
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
NULL
};
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
enum schedule_priority {
SCHEDULE_NONE = 0,
SCHEDULE_WEEKLY = 1,
SCHEDULE_DAILY = 2,
SCHEDULE_HOURLY = 3,
};
static enum schedule_priority parse_schedule(const char *value)
{
if (!value)
return SCHEDULE_NONE;
if (!strcasecmp(value, "hourly"))
return SCHEDULE_HOURLY;
if (!strcasecmp(value, "daily"))
return SCHEDULE_DAILY;
if (!strcasecmp(value, "weekly"))
return SCHEDULE_WEEKLY;
return SCHEDULE_NONE;
}
static int maintenance_opt_schedule(const struct option *opt, const char *arg,
int unset)
{
enum schedule_priority *priority = opt->value;
if (unset)
die(_("--no-schedule is not allowed"));
*priority = parse_schedule(arg);
if (!*priority)
die(_("unrecognized --schedule argument '%s'"), arg);
return 0;
}
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
struct maintenance_run_opts {
int auto_flag;
int quiet;
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
enum schedule_priority schedule;
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
};
/* Remember to update object flag allocation in object.h */
#define SEEN (1u<<0)
struct cg_auto_data {
int num_not_in_graph;
int limit;
};
static int dfs_on_ref(const char *refname,
const struct object_id *oid, int flags,
void *cb_data)
{
struct cg_auto_data *data = (struct cg_auto_data *)cb_data;
int result = 0;
struct object_id peeled;
struct commit_list *stack = NULL;
struct commit *commit;
if (!peel_ref(refname, &peeled))
oid = &peeled;
if (oid_object_info(the_repository, oid, NULL) != OBJ_COMMIT)
return 0;
commit = lookup_commit(the_repository, oid);
if (!commit)
return 0;
maintenance: test commit-graph auto condition The auto condition for the commit-graph maintenance task walks refs looking for commits that are not in the commit-graph file. This was added in 4ddc79b2 (maintenance: add auto condition for commit-graph task, 2020-09-17) but was left untested. The initial goal of this change was to demonstrate the feature works properly by adding tests. However, there was an off-by-one error that caused the basic tests around maintenance.commit-graph.auto=1 to fail when it should work. The subtlety is that if a ref tip is not in the commit-graph, then we were not adding that to the total count. In the test, we see that we have only added one commit since our last commit-graph write, so the auto condition would say there is nothing to do. The fix is simple: add the check for the commit-graph position to see that the tip is not in the commit-graph file before starting our walk. Since this happens before adding to the DFS stack, we do not need to clear our (currently empty) commit list. This does add some extra complexity for the test, because we also want to verify that the walk along the parents actually does some work. This means we need to add at least two commits in a row without writing the commit-graph. However, we also need to make sure no additional refs are pointing to the middle of this list or else the for_each_ref() in should_write_commit_graph() might visit these commits as tips instead of doing a DFS walk. Hence, the last two commits are added with "git commit" instead of "test_commit". Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-08 03:50:39 +03:00
if (parse_commit(commit) ||
commit_graph_position(commit) != COMMIT_NOT_FROM_GRAPH)
return 0;
maintenance: test commit-graph auto condition The auto condition for the commit-graph maintenance task walks refs looking for commits that are not in the commit-graph file. This was added in 4ddc79b2 (maintenance: add auto condition for commit-graph task, 2020-09-17) but was left untested. The initial goal of this change was to demonstrate the feature works properly by adding tests. However, there was an off-by-one error that caused the basic tests around maintenance.commit-graph.auto=1 to fail when it should work. The subtlety is that if a ref tip is not in the commit-graph, then we were not adding that to the total count. In the test, we see that we have only added one commit since our last commit-graph write, so the auto condition would say there is nothing to do. The fix is simple: add the check for the commit-graph position to see that the tip is not in the commit-graph file before starting our walk. Since this happens before adding to the DFS stack, we do not need to clear our (currently empty) commit list. This does add some extra complexity for the test, because we also want to verify that the walk along the parents actually does some work. This means we need to add at least two commits in a row without writing the commit-graph. However, we also need to make sure no additional refs are pointing to the middle of this list or else the for_each_ref() in should_write_commit_graph() might visit these commits as tips instead of doing a DFS walk. Hence, the last two commits are added with "git commit" instead of "test_commit". Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-10-08 03:50:39 +03:00
data->num_not_in_graph++;
if (data->num_not_in_graph >= data->limit)
return 1;
commit_list_append(commit, &stack);
while (!result && stack) {
struct commit_list *parent;
commit = pop_commit(&stack);
for (parent = commit->parents; parent; parent = parent->next) {
if (parse_commit(parent->item) ||
commit_graph_position(parent->item) != COMMIT_NOT_FROM_GRAPH ||
parent->item->object.flags & SEEN)
continue;
parent->item->object.flags |= SEEN;
data->num_not_in_graph++;
if (data->num_not_in_graph >= data->limit) {
result = 1;
break;
}
commit_list_append(parent->item, &stack);
}
}
free_commit_list(stack);
return result;
}
static int should_write_commit_graph(void)
{
int result;
struct cg_auto_data data;
data.num_not_in_graph = 0;
data.limit = 100;
git_config_get_int("maintenance.commit-graph.auto",
&data.limit);
if (!data.limit)
return 0;
if (data.limit < 0)
return 1;
result = for_each_ref(dfs_on_ref, &data);
repo_clear_commit_marks(the_repository, SEEN);
return result;
}
static int run_write_commit_graph(struct maintenance_run_opts *opts)
{
struct child_process child = CHILD_PROCESS_INIT;
child.git_cmd = 1;
strvec_pushl(&child.args, "commit-graph", "write",
"--split", "--reachable", NULL);
if (opts->quiet)
strvec_push(&child.args, "--no-progress");
return !!run_command(&child);
}
static int maintenance_task_commit_graph(struct maintenance_run_opts *opts)
{
prepare_repo_settings(the_repository);
if (!the_repository->settings.core_commit_graph)
return 0;
close_object_store(the_repository->objects);
if (run_write_commit_graph(opts)) {
error(_("failed to write commit-graph"));
return 1;
}
return 0;
}
maintenance: add prefetch task When working with very large repositories, an incremental 'git fetch' command can download a large amount of data. If there are many other users pushing to a common repo, then this data can rival the initial pack-file size of a 'git clone' of a medium-size repo. Users may want to keep the data on their local repos as close as possible to the data on the remote repos by fetching periodically in the background. This can break up a large daily fetch into several smaller hourly fetches. The task is called "prefetch" because it is work done in advance of a foreground fetch to make that 'git fetch' command much faster. However, if we simply ran 'git fetch <remote>' in the background, then the user running a foreground 'git fetch <remote>' would lose some important feedback when a new branch appears or an existing branch updates. This is especially true if a remote branch is force-updated and this isn't noticed by the user because it occurred in the background. Further, the functionality of 'git push --force-with-lease' becomes suspect. When running 'git fetch <remote> <options>' in the background, use the following options for careful updating: 1. --no-tags prevents getting a new tag when a user wants to see the new tags appear in their foreground fetches. 2. --refmap= removes the configured refspec which usually updates refs/remotes/<remote>/* with the refs advertised by the remote. While this looks confusing, this was documented and tested by b40a50264ac (fetch: document and test --refmap="", 2020-01-21), including this sentence in the documentation: Providing an empty `<refspec>` to the `--refmap` option causes Git to ignore the configured refspecs and rely entirely on the refspecs supplied as command-line arguments. 3. By adding a new refspec "+refs/heads/*:refs/prefetch/<remote>/*" we can ensure that we actually load the new values somewhere in our refspace while not updating refs/heads or refs/remotes. By storing these refs here, the commit-graph job will update the commit-graph with the commits from these hidden refs. 4. --prune will delete the refs/prefetch/<remote> refs that no longer appear on the remote. 5. --no-write-fetch-head prevents updating FETCH_HEAD. We've been using this step as a critical background job in Scalar [1] (and VFS for Git). This solved a pain point that was showing up in user reports: fetching was a pain! Users do not like waiting to download the data that was created while they were away from their machines. After implementing background fetch, the foreground fetch commands sped up significantly because they mostly just update refs and download a small amount of new data. The effect is especially dramatic when paried with --no-show-forced-udpates (through fetch.showForcedUpdates=false). [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/FetchStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:31 +03:00
static int fetch_remote(const char *remote, struct maintenance_run_opts *opts)
{
struct child_process child = CHILD_PROCESS_INIT;
child.git_cmd = 1;
strvec_pushl(&child.args, "fetch", remote, "--prune", "--no-tags",
"--no-write-fetch-head", "--recurse-submodules=no",
"--refmap=", NULL);
if (opts->quiet)
strvec_push(&child.args, "--quiet");
strvec_pushf(&child.args, "+refs/heads/*:refs/prefetch/%s/*", remote);
return !!run_command(&child);
}
static int append_remote(struct remote *remote, void *cbdata)
{
struct string_list *remotes = (struct string_list *)cbdata;
string_list_append(remotes, remote->name);
return 0;
}
static int maintenance_task_prefetch(struct maintenance_run_opts *opts)
{
int result = 0;
struct string_list_item *item;
struct string_list remotes = STRING_LIST_INIT_DUP;
if (for_each_remote(append_remote, &remotes)) {
error(_("failed to fill remotes"));
result = 1;
goto cleanup;
}
for_each_string_list_item(item, &remotes)
result |= fetch_remote(item->string, opts);
cleanup:
string_list_clear(&remotes, 0);
return result;
}
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
static int maintenance_task_gc(struct maintenance_run_opts *opts)
{
struct child_process child = CHILD_PROCESS_INIT;
child.git_cmd = 1;
strvec_push(&child.args, "gc");
if (opts->auto_flag)
strvec_push(&child.args, "--auto");
if (opts->quiet)
strvec_push(&child.args, "--quiet");
else
strvec_push(&child.args, "--no-quiet");
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
close_object_store(the_repository->objects);
return run_command(&child);
}
maintenance: add loose-objects task One goal of background maintenance jobs is to allow a user to disable auto-gc (gc.auto=0) but keep their repository in a clean state. Without any cleanup, loose objects will clutter the object database and slow operations. In addition, the loose objects will take up extra space because they are not stored with deltas against similar objects. Create a 'loose-objects' task for the 'git maintenance run' command. This helps clean up loose objects without disrupting concurrent Git commands using the following sequence of events: 1. Run 'git prune-packed' to delete any loose objects that exist in a pack-file. Concurrent commands will prefer the packed version of the object to the loose version. (Of course, there are exceptions for commands that specifically care about the location of an object. These are rare for a user to run on purpose, and we hope a user that has selected background maintenance will not be trying to do foreground maintenance.) 2. Run 'git pack-objects' on a batch of loose objects. These objects are grouped by scanning the loose object directories in lexicographic order until listing all loose objects -or- reaching 50,000 objects. This is more than enough if the loose objects are created only by a user doing normal development. We noticed users with _millions_ of loose objects because VFS for Git downloads blobs on-demand when a file read operation requires populating a virtual file. This step is based on a similar step in Scalar [1] and VFS for Git. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/LooseObjectsStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:32 +03:00
static int prune_packed(struct maintenance_run_opts *opts)
{
struct child_process child = CHILD_PROCESS_INIT;
child.git_cmd = 1;
strvec_push(&child.args, "prune-packed");
if (opts->quiet)
strvec_push(&child.args, "--quiet");
return !!run_command(&child);
}
struct write_loose_object_data {
FILE *in;
int count;
int batch_size;
};
static int loose_object_auto_limit = 100;
static int loose_object_count(const struct object_id *oid,
const char *path,
void *data)
{
int *count = (int*)data;
if (++(*count) >= loose_object_auto_limit)
return 1;
return 0;
}
static int loose_object_auto_condition(void)
{
int count = 0;
git_config_get_int("maintenance.loose-objects.auto",
&loose_object_auto_limit);
if (!loose_object_auto_limit)
return 0;
if (loose_object_auto_limit < 0)
return 1;
return for_each_loose_file_in_objdir(the_repository->objects->odb->path,
loose_object_count,
NULL, NULL, &count);
}
maintenance: add loose-objects task One goal of background maintenance jobs is to allow a user to disable auto-gc (gc.auto=0) but keep their repository in a clean state. Without any cleanup, loose objects will clutter the object database and slow operations. In addition, the loose objects will take up extra space because they are not stored with deltas against similar objects. Create a 'loose-objects' task for the 'git maintenance run' command. This helps clean up loose objects without disrupting concurrent Git commands using the following sequence of events: 1. Run 'git prune-packed' to delete any loose objects that exist in a pack-file. Concurrent commands will prefer the packed version of the object to the loose version. (Of course, there are exceptions for commands that specifically care about the location of an object. These are rare for a user to run on purpose, and we hope a user that has selected background maintenance will not be trying to do foreground maintenance.) 2. Run 'git pack-objects' on a batch of loose objects. These objects are grouped by scanning the loose object directories in lexicographic order until listing all loose objects -or- reaching 50,000 objects. This is more than enough if the loose objects are created only by a user doing normal development. We noticed users with _millions_ of loose objects because VFS for Git downloads blobs on-demand when a file read operation requires populating a virtual file. This step is based on a similar step in Scalar [1] and VFS for Git. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/LooseObjectsStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:32 +03:00
static int bail_on_loose(const struct object_id *oid,
const char *path,
void *data)
{
return 1;
}
static int write_loose_object_to_stdin(const struct object_id *oid,
const char *path,
void *data)
{
struct write_loose_object_data *d = (struct write_loose_object_data *)data;
fprintf(d->in, "%s\n", oid_to_hex(oid));
return ++(d->count) > d->batch_size;
}
static int pack_loose(struct maintenance_run_opts *opts)
{
struct repository *r = the_repository;
int result = 0;
struct write_loose_object_data data;
struct child_process pack_proc = CHILD_PROCESS_INIT;
/*
* Do not start pack-objects process
* if there are no loose objects.
*/
if (!for_each_loose_file_in_objdir(r->objects->odb->path,
bail_on_loose,
NULL, NULL, NULL))
return 0;
pack_proc.git_cmd = 1;
strvec_push(&pack_proc.args, "pack-objects");
if (opts->quiet)
strvec_push(&pack_proc.args, "--quiet");
strvec_pushf(&pack_proc.args, "%s/pack/loose", r->objects->odb->path);
pack_proc.in = -1;
if (start_command(&pack_proc)) {
error(_("failed to start 'git pack-objects' process"));
return 1;
}
data.in = xfdopen(pack_proc.in, "w");
data.count = 0;
data.batch_size = 50000;
for_each_loose_file_in_objdir(r->objects->odb->path,
write_loose_object_to_stdin,
NULL,
NULL,
&data);
fclose(data.in);
if (finish_command(&pack_proc)) {
error(_("failed to finish 'git pack-objects' process"));
result = 1;
}
return result;
}
static int maintenance_task_loose_objects(struct maintenance_run_opts *opts)
{
return prune_packed(opts) || pack_loose(opts);
}
static int incremental_repack_auto_condition(void)
{
struct packed_git *p;
int enabled;
int incremental_repack_auto_limit = 10;
int count = 0;
if (git_config_get_bool("core.multiPackIndex", &enabled) ||
!enabled)
return 0;
git_config_get_int("maintenance.incremental-repack.auto",
&incremental_repack_auto_limit);
if (!incremental_repack_auto_limit)
return 0;
if (incremental_repack_auto_limit < 0)
return 1;
for (p = get_packed_git(the_repository);
count < incremental_repack_auto_limit && p;
p = p->next) {
if (!p->multi_pack_index)
count++;
}
return count >= incremental_repack_auto_limit;
}
maintenance: add incremental-repack task The previous change cleaned up loose objects using the 'loose-objects' that can be run safely in the background. Add a similar job that performs similar cleanups for pack-files. One issue with running 'git repack' is that it is designed to repack all pack-files into a single pack-file. While this is the most space-efficient way to store object data, it is not time or memory efficient. This becomes extremely important if the repo is so large that a user struggles to store two copies of the pack on their disk. Instead, perform an "incremental" repack by collecting a few small pack-files into a new pack-file. The multi-pack-index facilitates this process ever since 'git multi-pack-index expire' was added in 19575c7 (multi-pack-index: implement 'expire' subcommand, 2019-06-10) and 'git multi-pack-index repack' was added in ce1e4a1 (midx: implement midx_repack(), 2019-06-10). The 'incremental-repack' task runs the following steps: 1. 'git multi-pack-index write' creates a multi-pack-index file if one did not exist, and otherwise will update the multi-pack-index with any new pack-files that appeared since the last write. This is particularly relevant with the background fetch job. When the multi-pack-index sees two copies of the same object, it stores the offset data into the newer pack-file. This means that some old pack-files could become "unreferenced" which I will use to mean "a pack-file that is in the pack-file list of the multi-pack-index but none of the objects in the multi-pack-index reference a location inside that pack-file." 2. 'git multi-pack-index expire' deletes any unreferenced pack-files and updaes the multi-pack-index to drop those pack-files from the list. This is safe to do as concurrent Git processes will see the multi-pack-index and not open those packs when looking for object contents. (Similar to the 'loose-objects' job, there are some Git commands that open pack-files regardless of the multi-pack-index, but they are rarely used. Further, a user that self-selects to use background operations would likely refrain from using those commands.) 3. 'git multi-pack-index repack --bacth-size=<size>' collects a set of pack-files that are listed in the multi-pack-index and creates a new pack-file containing the objects whose offsets are listed by the multi-pack-index to be in those objects. The set of pack- files is selected greedily by sorting the pack-files by modified time and adding a pack-file to the set if its "expected size" is smaller than the batch size until the total expected size of the selected pack-files is at least the batch size. The "expected size" is calculated by taking the size of the pack-file divided by the number of objects in the pack-file and multiplied by the number of objects from the multi-pack-index with offset in that pack-file. The expected size approximates how much data from that pack-file will contribute to the resulting pack-file size. The intention is that the resulting pack-file will be close in size to the provided batch size. The next run of the incremental-repack task will delete these repacked pack-files during the 'expire' step. In this version, the batch size is set to "0" which ignores the size restrictions when selecting the pack-files. It instead selects all pack-files and repacks all packed objects into a single pack-file. This will be updated in the next change, but it requires doing some calculations that are better isolated to a separate change. These steps are based on a similar background maintenance step in Scalar (and VFS for Git) [1]. This was incredibly effective for users of the Windows OS repository. After using the same VFS for Git repository for over a year, some users had _thousands_ of pack-files that combined to up to 250 GB of data. We noticed a few users were running into the open file descriptor limits (due in part to a bug in the multi-pack-index fixed by af96fe3 (midx: add packs to packed_git linked list, 2019-04-29). These pack-files were mostly small since they contained the commits and trees that were pushed to the origin in a given hour. The GVFS protocol includes a "prefetch" step that asks for pre-computed pack- files containing commits and trees by timestamp. These pack-files were grouped into "daily" pack-files once a day for up to 30 days. If a user did not request prefetch packs for over 30 days, then they would get the entire history of commits and trees in a new, large pack-file. This led to a large number of pack-files that had poor delta compression. By running this pack-file maintenance step once per day, these repos with thousands of packs spanning 200+ GB dropped to dozens of pack- files spanning 30-50 GB. This was done all without removing objects from the system and using a constant batch size of two gigabytes. Once the work was done to reduce the pack-files to small sizes, the batch size of two gigabytes means that not every run triggers a repack operation, so the following run will not expire a pack-file. This has kept these repos in a "clean" state. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/PackfileMaintenanceStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:36 +03:00
static int multi_pack_index_write(struct maintenance_run_opts *opts)
{
struct child_process child = CHILD_PROCESS_INIT;
child.git_cmd = 1;
strvec_pushl(&child.args, "multi-pack-index", "write", NULL);
if (opts->quiet)
strvec_push(&child.args, "--no-progress");
if (run_command(&child))
return error(_("failed to write multi-pack-index"));
return 0;
}
static int multi_pack_index_expire(struct maintenance_run_opts *opts)
{
struct child_process child = CHILD_PROCESS_INIT;
child.git_cmd = 1;
strvec_pushl(&child.args, "multi-pack-index", "expire", NULL);
if (opts->quiet)
strvec_push(&child.args, "--no-progress");
close_object_store(the_repository->objects);
if (run_command(&child))
return error(_("'git multi-pack-index expire' failed"));
return 0;
}
maintenance: auto-size incremental-repack batch When repacking during the 'incremental-repack' task, we use the --batch-size option in 'git multi-pack-index repack'. The initial setting used --batch-size=0 to repack everything into a single pack-file. This is not sustainable for a large repository. The amount of work required is also likely to use too many system resources for a background job. Update the 'incremental-repack' task by dynamically computing a --batch-size option based on the current pack-file structure. The dynamic default size is computed with this idea in mind for a client repository that was cloned from a very large remote: there is likely one "big" pack-file that was created at clone time. Thus, do not try repacking it as it is likely packed efficiently by the server. Instead, we select the second-largest pack-file, and create a batch size that is one larger than that pack-file. If there are three or more pack-files, then this guarantees that at least two will be combined into a new pack-file. Of course, this means that the second-largest pack-file size is likely to grow over time and may eventually surpass the initially-cloned pack-file. Recall that the pack-file batch is selected in a greedy manner: the packs are considered from oldest to newest and are selected if they have size smaller than the batch size until the total selected size is larger than the batch size. Thus, that oldest "clone" pack will be first to repack after the new data creates a pack larger than that. We also want to place some limits on how large these pack-files become, in order to bound the amount of time spent repacking. A maximum batch-size of two gigabytes means that large repositories will never be packed into a single pack-file using this job, but also that repack is rather expensive. This is a trade-off that is valuable to have if the maintenance is being run automatically or in the background. Users who truly want to optimize for space and performance (and are willing to pay the upfront cost of a full repack) can use the 'gc' task to do so. Create a test for this two gigabyte limit by creating an EXPENSIVE test that generates two pack-files of roughly 2.5 gigabytes in size, then performs an incremental repack. Check that the --batch-size argument in the subcommand uses the hard-coded maximum. Helped-by: Chris Torek <chris.torek@gmail.com> Reported-by: Son Luong Ngoc <sluongng@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:37 +03:00
#define TWO_GIGABYTES (INT32_MAX)
static off_t get_auto_pack_size(void)
{
/*
* The "auto" value is special: we optimize for
* one large pack-file (i.e. from a clone) and
* expect the rest to be small and they can be
* repacked quickly.
*
* The strategy we select here is to select a
* size that is one more than the second largest
* pack-file. This ensures that we will repack
* at least two packs if there are three or more
* packs.
*/
off_t max_size = 0;
off_t second_largest_size = 0;
off_t result_size;
struct packed_git *p;
struct repository *r = the_repository;
reprepare_packed_git(r);
for (p = get_all_packs(r); p; p = p->next) {
if (p->pack_size > max_size) {
second_largest_size = max_size;
max_size = p->pack_size;
} else if (p->pack_size > second_largest_size)
second_largest_size = p->pack_size;
}
result_size = second_largest_size + 1;
/* But limit ourselves to a batch size of 2g */
if (result_size > TWO_GIGABYTES)
result_size = TWO_GIGABYTES;
return result_size;
}
maintenance: add incremental-repack task The previous change cleaned up loose objects using the 'loose-objects' that can be run safely in the background. Add a similar job that performs similar cleanups for pack-files. One issue with running 'git repack' is that it is designed to repack all pack-files into a single pack-file. While this is the most space-efficient way to store object data, it is not time or memory efficient. This becomes extremely important if the repo is so large that a user struggles to store two copies of the pack on their disk. Instead, perform an "incremental" repack by collecting a few small pack-files into a new pack-file. The multi-pack-index facilitates this process ever since 'git multi-pack-index expire' was added in 19575c7 (multi-pack-index: implement 'expire' subcommand, 2019-06-10) and 'git multi-pack-index repack' was added in ce1e4a1 (midx: implement midx_repack(), 2019-06-10). The 'incremental-repack' task runs the following steps: 1. 'git multi-pack-index write' creates a multi-pack-index file if one did not exist, and otherwise will update the multi-pack-index with any new pack-files that appeared since the last write. This is particularly relevant with the background fetch job. When the multi-pack-index sees two copies of the same object, it stores the offset data into the newer pack-file. This means that some old pack-files could become "unreferenced" which I will use to mean "a pack-file that is in the pack-file list of the multi-pack-index but none of the objects in the multi-pack-index reference a location inside that pack-file." 2. 'git multi-pack-index expire' deletes any unreferenced pack-files and updaes the multi-pack-index to drop those pack-files from the list. This is safe to do as concurrent Git processes will see the multi-pack-index and not open those packs when looking for object contents. (Similar to the 'loose-objects' job, there are some Git commands that open pack-files regardless of the multi-pack-index, but they are rarely used. Further, a user that self-selects to use background operations would likely refrain from using those commands.) 3. 'git multi-pack-index repack --bacth-size=<size>' collects a set of pack-files that are listed in the multi-pack-index and creates a new pack-file containing the objects whose offsets are listed by the multi-pack-index to be in those objects. The set of pack- files is selected greedily by sorting the pack-files by modified time and adding a pack-file to the set if its "expected size" is smaller than the batch size until the total expected size of the selected pack-files is at least the batch size. The "expected size" is calculated by taking the size of the pack-file divided by the number of objects in the pack-file and multiplied by the number of objects from the multi-pack-index with offset in that pack-file. The expected size approximates how much data from that pack-file will contribute to the resulting pack-file size. The intention is that the resulting pack-file will be close in size to the provided batch size. The next run of the incremental-repack task will delete these repacked pack-files during the 'expire' step. In this version, the batch size is set to "0" which ignores the size restrictions when selecting the pack-files. It instead selects all pack-files and repacks all packed objects into a single pack-file. This will be updated in the next change, but it requires doing some calculations that are better isolated to a separate change. These steps are based on a similar background maintenance step in Scalar (and VFS for Git) [1]. This was incredibly effective for users of the Windows OS repository. After using the same VFS for Git repository for over a year, some users had _thousands_ of pack-files that combined to up to 250 GB of data. We noticed a few users were running into the open file descriptor limits (due in part to a bug in the multi-pack-index fixed by af96fe3 (midx: add packs to packed_git linked list, 2019-04-29). These pack-files were mostly small since they contained the commits and trees that were pushed to the origin in a given hour. The GVFS protocol includes a "prefetch" step that asks for pre-computed pack- files containing commits and trees by timestamp. These pack-files were grouped into "daily" pack-files once a day for up to 30 days. If a user did not request prefetch packs for over 30 days, then they would get the entire history of commits and trees in a new, large pack-file. This led to a large number of pack-files that had poor delta compression. By running this pack-file maintenance step once per day, these repos with thousands of packs spanning 200+ GB dropped to dozens of pack- files spanning 30-50 GB. This was done all without removing objects from the system and using a constant batch size of two gigabytes. Once the work was done to reduce the pack-files to small sizes, the batch size of two gigabytes means that not every run triggers a repack operation, so the following run will not expire a pack-file. This has kept these repos in a "clean" state. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/PackfileMaintenanceStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:36 +03:00
static int multi_pack_index_repack(struct maintenance_run_opts *opts)
{
struct child_process child = CHILD_PROCESS_INIT;
child.git_cmd = 1;
strvec_pushl(&child.args, "multi-pack-index", "repack", NULL);
if (opts->quiet)
strvec_push(&child.args, "--no-progress");
maintenance: auto-size incremental-repack batch When repacking during the 'incremental-repack' task, we use the --batch-size option in 'git multi-pack-index repack'. The initial setting used --batch-size=0 to repack everything into a single pack-file. This is not sustainable for a large repository. The amount of work required is also likely to use too many system resources for a background job. Update the 'incremental-repack' task by dynamically computing a --batch-size option based on the current pack-file structure. The dynamic default size is computed with this idea in mind for a client repository that was cloned from a very large remote: there is likely one "big" pack-file that was created at clone time. Thus, do not try repacking it as it is likely packed efficiently by the server. Instead, we select the second-largest pack-file, and create a batch size that is one larger than that pack-file. If there are three or more pack-files, then this guarantees that at least two will be combined into a new pack-file. Of course, this means that the second-largest pack-file size is likely to grow over time and may eventually surpass the initially-cloned pack-file. Recall that the pack-file batch is selected in a greedy manner: the packs are considered from oldest to newest and are selected if they have size smaller than the batch size until the total selected size is larger than the batch size. Thus, that oldest "clone" pack will be first to repack after the new data creates a pack larger than that. We also want to place some limits on how large these pack-files become, in order to bound the amount of time spent repacking. A maximum batch-size of two gigabytes means that large repositories will never be packed into a single pack-file using this job, but also that repack is rather expensive. This is a trade-off that is valuable to have if the maintenance is being run automatically or in the background. Users who truly want to optimize for space and performance (and are willing to pay the upfront cost of a full repack) can use the 'gc' task to do so. Create a test for this two gigabyte limit by creating an EXPENSIVE test that generates two pack-files of roughly 2.5 gigabytes in size, then performs an incremental repack. Check that the --batch-size argument in the subcommand uses the hard-coded maximum. Helped-by: Chris Torek <chris.torek@gmail.com> Reported-by: Son Luong Ngoc <sluongng@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:37 +03:00
strvec_pushf(&child.args, "--batch-size=%"PRIuMAX,
(uintmax_t)get_auto_pack_size());
maintenance: add incremental-repack task The previous change cleaned up loose objects using the 'loose-objects' that can be run safely in the background. Add a similar job that performs similar cleanups for pack-files. One issue with running 'git repack' is that it is designed to repack all pack-files into a single pack-file. While this is the most space-efficient way to store object data, it is not time or memory efficient. This becomes extremely important if the repo is so large that a user struggles to store two copies of the pack on their disk. Instead, perform an "incremental" repack by collecting a few small pack-files into a new pack-file. The multi-pack-index facilitates this process ever since 'git multi-pack-index expire' was added in 19575c7 (multi-pack-index: implement 'expire' subcommand, 2019-06-10) and 'git multi-pack-index repack' was added in ce1e4a1 (midx: implement midx_repack(), 2019-06-10). The 'incremental-repack' task runs the following steps: 1. 'git multi-pack-index write' creates a multi-pack-index file if one did not exist, and otherwise will update the multi-pack-index with any new pack-files that appeared since the last write. This is particularly relevant with the background fetch job. When the multi-pack-index sees two copies of the same object, it stores the offset data into the newer pack-file. This means that some old pack-files could become "unreferenced" which I will use to mean "a pack-file that is in the pack-file list of the multi-pack-index but none of the objects in the multi-pack-index reference a location inside that pack-file." 2. 'git multi-pack-index expire' deletes any unreferenced pack-files and updaes the multi-pack-index to drop those pack-files from the list. This is safe to do as concurrent Git processes will see the multi-pack-index and not open those packs when looking for object contents. (Similar to the 'loose-objects' job, there are some Git commands that open pack-files regardless of the multi-pack-index, but they are rarely used. Further, a user that self-selects to use background operations would likely refrain from using those commands.) 3. 'git multi-pack-index repack --bacth-size=<size>' collects a set of pack-files that are listed in the multi-pack-index and creates a new pack-file containing the objects whose offsets are listed by the multi-pack-index to be in those objects. The set of pack- files is selected greedily by sorting the pack-files by modified time and adding a pack-file to the set if its "expected size" is smaller than the batch size until the total expected size of the selected pack-files is at least the batch size. The "expected size" is calculated by taking the size of the pack-file divided by the number of objects in the pack-file and multiplied by the number of objects from the multi-pack-index with offset in that pack-file. The expected size approximates how much data from that pack-file will contribute to the resulting pack-file size. The intention is that the resulting pack-file will be close in size to the provided batch size. The next run of the incremental-repack task will delete these repacked pack-files during the 'expire' step. In this version, the batch size is set to "0" which ignores the size restrictions when selecting the pack-files. It instead selects all pack-files and repacks all packed objects into a single pack-file. This will be updated in the next change, but it requires doing some calculations that are better isolated to a separate change. These steps are based on a similar background maintenance step in Scalar (and VFS for Git) [1]. This was incredibly effective for users of the Windows OS repository. After using the same VFS for Git repository for over a year, some users had _thousands_ of pack-files that combined to up to 250 GB of data. We noticed a few users were running into the open file descriptor limits (due in part to a bug in the multi-pack-index fixed by af96fe3 (midx: add packs to packed_git linked list, 2019-04-29). These pack-files were mostly small since they contained the commits and trees that were pushed to the origin in a given hour. The GVFS protocol includes a "prefetch" step that asks for pre-computed pack- files containing commits and trees by timestamp. These pack-files were grouped into "daily" pack-files once a day for up to 30 days. If a user did not request prefetch packs for over 30 days, then they would get the entire history of commits and trees in a new, large pack-file. This led to a large number of pack-files that had poor delta compression. By running this pack-file maintenance step once per day, these repos with thousands of packs spanning 200+ GB dropped to dozens of pack- files spanning 30-50 GB. This was done all without removing objects from the system and using a constant batch size of two gigabytes. Once the work was done to reduce the pack-files to small sizes, the batch size of two gigabytes means that not every run triggers a repack operation, so the following run will not expire a pack-file. This has kept these repos in a "clean" state. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/PackfileMaintenanceStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:36 +03:00
close_object_store(the_repository->objects);
if (run_command(&child))
return error(_("'git multi-pack-index repack' failed"));
return 0;
}
static int maintenance_task_incremental_repack(struct maintenance_run_opts *opts)
{
prepare_repo_settings(the_repository);
if (!the_repository->settings.core_multi_pack_index) {
warning(_("skipping incremental-repack task because core.multiPackIndex is disabled"));
return 0;
}
if (multi_pack_index_write(opts))
return 1;
if (multi_pack_index_expire(opts))
return 1;
if (multi_pack_index_repack(opts))
return 1;
return 0;
}
typedef int maintenance_task_fn(struct maintenance_run_opts *opts);
/*
* An auto condition function returns 1 if the task should run
* and 0 if the task should NOT run. See needs_to_gc() for an
* example.
*/
typedef int maintenance_auto_fn(void);
struct maintenance_task {
const char *name;
maintenance_task_fn *fn;
maintenance_auto_fn *auto_condition;
unsigned enabled:1;
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
enum schedule_priority schedule;
/* -1 if not selected. */
int selected_order;
};
enum maintenance_task_label {
maintenance: add prefetch task When working with very large repositories, an incremental 'git fetch' command can download a large amount of data. If there are many other users pushing to a common repo, then this data can rival the initial pack-file size of a 'git clone' of a medium-size repo. Users may want to keep the data on their local repos as close as possible to the data on the remote repos by fetching periodically in the background. This can break up a large daily fetch into several smaller hourly fetches. The task is called "prefetch" because it is work done in advance of a foreground fetch to make that 'git fetch' command much faster. However, if we simply ran 'git fetch <remote>' in the background, then the user running a foreground 'git fetch <remote>' would lose some important feedback when a new branch appears or an existing branch updates. This is especially true if a remote branch is force-updated and this isn't noticed by the user because it occurred in the background. Further, the functionality of 'git push --force-with-lease' becomes suspect. When running 'git fetch <remote> <options>' in the background, use the following options for careful updating: 1. --no-tags prevents getting a new tag when a user wants to see the new tags appear in their foreground fetches. 2. --refmap= removes the configured refspec which usually updates refs/remotes/<remote>/* with the refs advertised by the remote. While this looks confusing, this was documented and tested by b40a50264ac (fetch: document and test --refmap="", 2020-01-21), including this sentence in the documentation: Providing an empty `<refspec>` to the `--refmap` option causes Git to ignore the configured refspecs and rely entirely on the refspecs supplied as command-line arguments. 3. By adding a new refspec "+refs/heads/*:refs/prefetch/<remote>/*" we can ensure that we actually load the new values somewhere in our refspace while not updating refs/heads or refs/remotes. By storing these refs here, the commit-graph job will update the commit-graph with the commits from these hidden refs. 4. --prune will delete the refs/prefetch/<remote> refs that no longer appear on the remote. 5. --no-write-fetch-head prevents updating FETCH_HEAD. We've been using this step as a critical background job in Scalar [1] (and VFS for Git). This solved a pain point that was showing up in user reports: fetching was a pain! Users do not like waiting to download the data that was created while they were away from their machines. After implementing background fetch, the foreground fetch commands sped up significantly because they mostly just update refs and download a small amount of new data. The effect is especially dramatic when paried with --no-show-forced-udpates (through fetch.showForcedUpdates=false). [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/FetchStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:31 +03:00
TASK_PREFETCH,
maintenance: add loose-objects task One goal of background maintenance jobs is to allow a user to disable auto-gc (gc.auto=0) but keep their repository in a clean state. Without any cleanup, loose objects will clutter the object database and slow operations. In addition, the loose objects will take up extra space because they are not stored with deltas against similar objects. Create a 'loose-objects' task for the 'git maintenance run' command. This helps clean up loose objects without disrupting concurrent Git commands using the following sequence of events: 1. Run 'git prune-packed' to delete any loose objects that exist in a pack-file. Concurrent commands will prefer the packed version of the object to the loose version. (Of course, there are exceptions for commands that specifically care about the location of an object. These are rare for a user to run on purpose, and we hope a user that has selected background maintenance will not be trying to do foreground maintenance.) 2. Run 'git pack-objects' on a batch of loose objects. These objects are grouped by scanning the loose object directories in lexicographic order until listing all loose objects -or- reaching 50,000 objects. This is more than enough if the loose objects are created only by a user doing normal development. We noticed users with _millions_ of loose objects because VFS for Git downloads blobs on-demand when a file read operation requires populating a virtual file. This step is based on a similar step in Scalar [1] and VFS for Git. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/LooseObjectsStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:32 +03:00
TASK_LOOSE_OBJECTS,
maintenance: add incremental-repack task The previous change cleaned up loose objects using the 'loose-objects' that can be run safely in the background. Add a similar job that performs similar cleanups for pack-files. One issue with running 'git repack' is that it is designed to repack all pack-files into a single pack-file. While this is the most space-efficient way to store object data, it is not time or memory efficient. This becomes extremely important if the repo is so large that a user struggles to store two copies of the pack on their disk. Instead, perform an "incremental" repack by collecting a few small pack-files into a new pack-file. The multi-pack-index facilitates this process ever since 'git multi-pack-index expire' was added in 19575c7 (multi-pack-index: implement 'expire' subcommand, 2019-06-10) and 'git multi-pack-index repack' was added in ce1e4a1 (midx: implement midx_repack(), 2019-06-10). The 'incremental-repack' task runs the following steps: 1. 'git multi-pack-index write' creates a multi-pack-index file if one did not exist, and otherwise will update the multi-pack-index with any new pack-files that appeared since the last write. This is particularly relevant with the background fetch job. When the multi-pack-index sees two copies of the same object, it stores the offset data into the newer pack-file. This means that some old pack-files could become "unreferenced" which I will use to mean "a pack-file that is in the pack-file list of the multi-pack-index but none of the objects in the multi-pack-index reference a location inside that pack-file." 2. 'git multi-pack-index expire' deletes any unreferenced pack-files and updaes the multi-pack-index to drop those pack-files from the list. This is safe to do as concurrent Git processes will see the multi-pack-index and not open those packs when looking for object contents. (Similar to the 'loose-objects' job, there are some Git commands that open pack-files regardless of the multi-pack-index, but they are rarely used. Further, a user that self-selects to use background operations would likely refrain from using those commands.) 3. 'git multi-pack-index repack --bacth-size=<size>' collects a set of pack-files that are listed in the multi-pack-index and creates a new pack-file containing the objects whose offsets are listed by the multi-pack-index to be in those objects. The set of pack- files is selected greedily by sorting the pack-files by modified time and adding a pack-file to the set if its "expected size" is smaller than the batch size until the total expected size of the selected pack-files is at least the batch size. The "expected size" is calculated by taking the size of the pack-file divided by the number of objects in the pack-file and multiplied by the number of objects from the multi-pack-index with offset in that pack-file. The expected size approximates how much data from that pack-file will contribute to the resulting pack-file size. The intention is that the resulting pack-file will be close in size to the provided batch size. The next run of the incremental-repack task will delete these repacked pack-files during the 'expire' step. In this version, the batch size is set to "0" which ignores the size restrictions when selecting the pack-files. It instead selects all pack-files and repacks all packed objects into a single pack-file. This will be updated in the next change, but it requires doing some calculations that are better isolated to a separate change. These steps are based on a similar background maintenance step in Scalar (and VFS for Git) [1]. This was incredibly effective for users of the Windows OS repository. After using the same VFS for Git repository for over a year, some users had _thousands_ of pack-files that combined to up to 250 GB of data. We noticed a few users were running into the open file descriptor limits (due in part to a bug in the multi-pack-index fixed by af96fe3 (midx: add packs to packed_git linked list, 2019-04-29). These pack-files were mostly small since they contained the commits and trees that were pushed to the origin in a given hour. The GVFS protocol includes a "prefetch" step that asks for pre-computed pack- files containing commits and trees by timestamp. These pack-files were grouped into "daily" pack-files once a day for up to 30 days. If a user did not request prefetch packs for over 30 days, then they would get the entire history of commits and trees in a new, large pack-file. This led to a large number of pack-files that had poor delta compression. By running this pack-file maintenance step once per day, these repos with thousands of packs spanning 200+ GB dropped to dozens of pack- files spanning 30-50 GB. This was done all without removing objects from the system and using a constant batch size of two gigabytes. Once the work was done to reduce the pack-files to small sizes, the batch size of two gigabytes means that not every run triggers a repack operation, so the following run will not expire a pack-file. This has kept these repos in a "clean" state. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/PackfileMaintenanceStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:36 +03:00
TASK_INCREMENTAL_REPACK,
TASK_GC,
TASK_COMMIT_GRAPH,
/* Leave as final value */
TASK__COUNT
};
static struct maintenance_task tasks[] = {
maintenance: add prefetch task When working with very large repositories, an incremental 'git fetch' command can download a large amount of data. If there are many other users pushing to a common repo, then this data can rival the initial pack-file size of a 'git clone' of a medium-size repo. Users may want to keep the data on their local repos as close as possible to the data on the remote repos by fetching periodically in the background. This can break up a large daily fetch into several smaller hourly fetches. The task is called "prefetch" because it is work done in advance of a foreground fetch to make that 'git fetch' command much faster. However, if we simply ran 'git fetch <remote>' in the background, then the user running a foreground 'git fetch <remote>' would lose some important feedback when a new branch appears or an existing branch updates. This is especially true if a remote branch is force-updated and this isn't noticed by the user because it occurred in the background. Further, the functionality of 'git push --force-with-lease' becomes suspect. When running 'git fetch <remote> <options>' in the background, use the following options for careful updating: 1. --no-tags prevents getting a new tag when a user wants to see the new tags appear in their foreground fetches. 2. --refmap= removes the configured refspec which usually updates refs/remotes/<remote>/* with the refs advertised by the remote. While this looks confusing, this was documented and tested by b40a50264ac (fetch: document and test --refmap="", 2020-01-21), including this sentence in the documentation: Providing an empty `<refspec>` to the `--refmap` option causes Git to ignore the configured refspecs and rely entirely on the refspecs supplied as command-line arguments. 3. By adding a new refspec "+refs/heads/*:refs/prefetch/<remote>/*" we can ensure that we actually load the new values somewhere in our refspace while not updating refs/heads or refs/remotes. By storing these refs here, the commit-graph job will update the commit-graph with the commits from these hidden refs. 4. --prune will delete the refs/prefetch/<remote> refs that no longer appear on the remote. 5. --no-write-fetch-head prevents updating FETCH_HEAD. We've been using this step as a critical background job in Scalar [1] (and VFS for Git). This solved a pain point that was showing up in user reports: fetching was a pain! Users do not like waiting to download the data that was created while they were away from their machines. After implementing background fetch, the foreground fetch commands sped up significantly because they mostly just update refs and download a small amount of new data. The effect is especially dramatic when paried with --no-show-forced-udpates (through fetch.showForcedUpdates=false). [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/FetchStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:31 +03:00
[TASK_PREFETCH] = {
"prefetch",
maintenance_task_prefetch,
},
maintenance: add loose-objects task One goal of background maintenance jobs is to allow a user to disable auto-gc (gc.auto=0) but keep their repository in a clean state. Without any cleanup, loose objects will clutter the object database and slow operations. In addition, the loose objects will take up extra space because they are not stored with deltas against similar objects. Create a 'loose-objects' task for the 'git maintenance run' command. This helps clean up loose objects without disrupting concurrent Git commands using the following sequence of events: 1. Run 'git prune-packed' to delete any loose objects that exist in a pack-file. Concurrent commands will prefer the packed version of the object to the loose version. (Of course, there are exceptions for commands that specifically care about the location of an object. These are rare for a user to run on purpose, and we hope a user that has selected background maintenance will not be trying to do foreground maintenance.) 2. Run 'git pack-objects' on a batch of loose objects. These objects are grouped by scanning the loose object directories in lexicographic order until listing all loose objects -or- reaching 50,000 objects. This is more than enough if the loose objects are created only by a user doing normal development. We noticed users with _millions_ of loose objects because VFS for Git downloads blobs on-demand when a file read operation requires populating a virtual file. This step is based on a similar step in Scalar [1] and VFS for Git. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/LooseObjectsStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:32 +03:00
[TASK_LOOSE_OBJECTS] = {
"loose-objects",
maintenance_task_loose_objects,
loose_object_auto_condition,
maintenance: add loose-objects task One goal of background maintenance jobs is to allow a user to disable auto-gc (gc.auto=0) but keep their repository in a clean state. Without any cleanup, loose objects will clutter the object database and slow operations. In addition, the loose objects will take up extra space because they are not stored with deltas against similar objects. Create a 'loose-objects' task for the 'git maintenance run' command. This helps clean up loose objects without disrupting concurrent Git commands using the following sequence of events: 1. Run 'git prune-packed' to delete any loose objects that exist in a pack-file. Concurrent commands will prefer the packed version of the object to the loose version. (Of course, there are exceptions for commands that specifically care about the location of an object. These are rare for a user to run on purpose, and we hope a user that has selected background maintenance will not be trying to do foreground maintenance.) 2. Run 'git pack-objects' on a batch of loose objects. These objects are grouped by scanning the loose object directories in lexicographic order until listing all loose objects -or- reaching 50,000 objects. This is more than enough if the loose objects are created only by a user doing normal development. We noticed users with _millions_ of loose objects because VFS for Git downloads blobs on-demand when a file read operation requires populating a virtual file. This step is based on a similar step in Scalar [1] and VFS for Git. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/LooseObjectsStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:32 +03:00
},
maintenance: add incremental-repack task The previous change cleaned up loose objects using the 'loose-objects' that can be run safely in the background. Add a similar job that performs similar cleanups for pack-files. One issue with running 'git repack' is that it is designed to repack all pack-files into a single pack-file. While this is the most space-efficient way to store object data, it is not time or memory efficient. This becomes extremely important if the repo is so large that a user struggles to store two copies of the pack on their disk. Instead, perform an "incremental" repack by collecting a few small pack-files into a new pack-file. The multi-pack-index facilitates this process ever since 'git multi-pack-index expire' was added in 19575c7 (multi-pack-index: implement 'expire' subcommand, 2019-06-10) and 'git multi-pack-index repack' was added in ce1e4a1 (midx: implement midx_repack(), 2019-06-10). The 'incremental-repack' task runs the following steps: 1. 'git multi-pack-index write' creates a multi-pack-index file if one did not exist, and otherwise will update the multi-pack-index with any new pack-files that appeared since the last write. This is particularly relevant with the background fetch job. When the multi-pack-index sees two copies of the same object, it stores the offset data into the newer pack-file. This means that some old pack-files could become "unreferenced" which I will use to mean "a pack-file that is in the pack-file list of the multi-pack-index but none of the objects in the multi-pack-index reference a location inside that pack-file." 2. 'git multi-pack-index expire' deletes any unreferenced pack-files and updaes the multi-pack-index to drop those pack-files from the list. This is safe to do as concurrent Git processes will see the multi-pack-index and not open those packs when looking for object contents. (Similar to the 'loose-objects' job, there are some Git commands that open pack-files regardless of the multi-pack-index, but they are rarely used. Further, a user that self-selects to use background operations would likely refrain from using those commands.) 3. 'git multi-pack-index repack --bacth-size=<size>' collects a set of pack-files that are listed in the multi-pack-index and creates a new pack-file containing the objects whose offsets are listed by the multi-pack-index to be in those objects. The set of pack- files is selected greedily by sorting the pack-files by modified time and adding a pack-file to the set if its "expected size" is smaller than the batch size until the total expected size of the selected pack-files is at least the batch size. The "expected size" is calculated by taking the size of the pack-file divided by the number of objects in the pack-file and multiplied by the number of objects from the multi-pack-index with offset in that pack-file. The expected size approximates how much data from that pack-file will contribute to the resulting pack-file size. The intention is that the resulting pack-file will be close in size to the provided batch size. The next run of the incremental-repack task will delete these repacked pack-files during the 'expire' step. In this version, the batch size is set to "0" which ignores the size restrictions when selecting the pack-files. It instead selects all pack-files and repacks all packed objects into a single pack-file. This will be updated in the next change, but it requires doing some calculations that are better isolated to a separate change. These steps are based on a similar background maintenance step in Scalar (and VFS for Git) [1]. This was incredibly effective for users of the Windows OS repository. After using the same VFS for Git repository for over a year, some users had _thousands_ of pack-files that combined to up to 250 GB of data. We noticed a few users were running into the open file descriptor limits (due in part to a bug in the multi-pack-index fixed by af96fe3 (midx: add packs to packed_git linked list, 2019-04-29). These pack-files were mostly small since they contained the commits and trees that were pushed to the origin in a given hour. The GVFS protocol includes a "prefetch" step that asks for pre-computed pack- files containing commits and trees by timestamp. These pack-files were grouped into "daily" pack-files once a day for up to 30 days. If a user did not request prefetch packs for over 30 days, then they would get the entire history of commits and trees in a new, large pack-file. This led to a large number of pack-files that had poor delta compression. By running this pack-file maintenance step once per day, these repos with thousands of packs spanning 200+ GB dropped to dozens of pack- files spanning 30-50 GB. This was done all without removing objects from the system and using a constant batch size of two gigabytes. Once the work was done to reduce the pack-files to small sizes, the batch size of two gigabytes means that not every run triggers a repack operation, so the following run will not expire a pack-file. This has kept these repos in a "clean" state. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/PackfileMaintenanceStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:36 +03:00
[TASK_INCREMENTAL_REPACK] = {
"incremental-repack",
maintenance_task_incremental_repack,
incremental_repack_auto_condition,
maintenance: add incremental-repack task The previous change cleaned up loose objects using the 'loose-objects' that can be run safely in the background. Add a similar job that performs similar cleanups for pack-files. One issue with running 'git repack' is that it is designed to repack all pack-files into a single pack-file. While this is the most space-efficient way to store object data, it is not time or memory efficient. This becomes extremely important if the repo is so large that a user struggles to store two copies of the pack on their disk. Instead, perform an "incremental" repack by collecting a few small pack-files into a new pack-file. The multi-pack-index facilitates this process ever since 'git multi-pack-index expire' was added in 19575c7 (multi-pack-index: implement 'expire' subcommand, 2019-06-10) and 'git multi-pack-index repack' was added in ce1e4a1 (midx: implement midx_repack(), 2019-06-10). The 'incremental-repack' task runs the following steps: 1. 'git multi-pack-index write' creates a multi-pack-index file if one did not exist, and otherwise will update the multi-pack-index with any new pack-files that appeared since the last write. This is particularly relevant with the background fetch job. When the multi-pack-index sees two copies of the same object, it stores the offset data into the newer pack-file. This means that some old pack-files could become "unreferenced" which I will use to mean "a pack-file that is in the pack-file list of the multi-pack-index but none of the objects in the multi-pack-index reference a location inside that pack-file." 2. 'git multi-pack-index expire' deletes any unreferenced pack-files and updaes the multi-pack-index to drop those pack-files from the list. This is safe to do as concurrent Git processes will see the multi-pack-index and not open those packs when looking for object contents. (Similar to the 'loose-objects' job, there are some Git commands that open pack-files regardless of the multi-pack-index, but they are rarely used. Further, a user that self-selects to use background operations would likely refrain from using those commands.) 3. 'git multi-pack-index repack --bacth-size=<size>' collects a set of pack-files that are listed in the multi-pack-index and creates a new pack-file containing the objects whose offsets are listed by the multi-pack-index to be in those objects. The set of pack- files is selected greedily by sorting the pack-files by modified time and adding a pack-file to the set if its "expected size" is smaller than the batch size until the total expected size of the selected pack-files is at least the batch size. The "expected size" is calculated by taking the size of the pack-file divided by the number of objects in the pack-file and multiplied by the number of objects from the multi-pack-index with offset in that pack-file. The expected size approximates how much data from that pack-file will contribute to the resulting pack-file size. The intention is that the resulting pack-file will be close in size to the provided batch size. The next run of the incremental-repack task will delete these repacked pack-files during the 'expire' step. In this version, the batch size is set to "0" which ignores the size restrictions when selecting the pack-files. It instead selects all pack-files and repacks all packed objects into a single pack-file. This will be updated in the next change, but it requires doing some calculations that are better isolated to a separate change. These steps are based on a similar background maintenance step in Scalar (and VFS for Git) [1]. This was incredibly effective for users of the Windows OS repository. After using the same VFS for Git repository for over a year, some users had _thousands_ of pack-files that combined to up to 250 GB of data. We noticed a few users were running into the open file descriptor limits (due in part to a bug in the multi-pack-index fixed by af96fe3 (midx: add packs to packed_git linked list, 2019-04-29). These pack-files were mostly small since they contained the commits and trees that were pushed to the origin in a given hour. The GVFS protocol includes a "prefetch" step that asks for pre-computed pack- files containing commits and trees by timestamp. These pack-files were grouped into "daily" pack-files once a day for up to 30 days. If a user did not request prefetch packs for over 30 days, then they would get the entire history of commits and trees in a new, large pack-file. This led to a large number of pack-files that had poor delta compression. By running this pack-file maintenance step once per day, these repos with thousands of packs spanning 200+ GB dropped to dozens of pack- files spanning 30-50 GB. This was done all without removing objects from the system and using a constant batch size of two gigabytes. Once the work was done to reduce the pack-files to small sizes, the batch size of two gigabytes means that not every run triggers a repack operation, so the following run will not expire a pack-file. This has kept these repos in a "clean" state. [1] https://github.com/microsoft/scalar/blob/master/Scalar.Common/Maintenance/PackfileMaintenanceStep.cs Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-25 15:33:36 +03:00
},
[TASK_GC] = {
"gc",
maintenance_task_gc,
need_to_gc,
1,
},
[TASK_COMMIT_GRAPH] = {
"commit-graph",
maintenance_task_commit_graph,
should_write_commit_graph,
},
};
static int compare_tasks_by_selection(const void *a_, const void *b_)
{
const struct maintenance_task *a = a_;
const struct maintenance_task *b = b_;
return b->selected_order - a->selected_order;
}
static int maintenance_run_tasks(struct maintenance_run_opts *opts)
{
int i, found_selected = 0;
int result = 0;
struct lock_file lk;
struct repository *r = the_repository;
char *lock_path = xstrfmt("%s/maintenance", r->objects->odb->path);
if (hold_lock_file_for_update(&lk, lock_path, LOCK_NO_DEREF) < 0) {
/*
* Another maintenance command is running.
*
* If --auto was provided, then it is likely due to a
* recursive process stack. Do not report an error in
* that case.
*/
if (!opts->auto_flag && !opts->quiet)
warning(_("lock file '%s' exists, skipping maintenance"),
lock_path);
free(lock_path);
return 0;
}
free(lock_path);
for (i = 0; !found_selected && i < TASK__COUNT; i++)
found_selected = tasks[i].selected_order >= 0;
if (found_selected)
QSORT(tasks, TASK__COUNT, compare_tasks_by_selection);
for (i = 0; i < TASK__COUNT; i++) {
if (found_selected && tasks[i].selected_order < 0)
continue;
if (!found_selected && !tasks[i].enabled)
continue;
if (opts->auto_flag &&
(!tasks[i].auto_condition ||
!tasks[i].auto_condition()))
continue;
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
if (opts->schedule && tasks[i].schedule < opts->schedule)
continue;
trace2_region_enter("maintenance", tasks[i].name, r);
if (tasks[i].fn(opts)) {
error(_("task '%s' failed"), tasks[i].name);
result = 1;
}
trace2_region_leave("maintenance", tasks[i].name, r);
}
rollback_lock_file(&lk);
return result;
}
static void initialize_maintenance_strategy(void)
{
char *config_str;
if (git_config_get_string("maintenance.strategy", &config_str))
return;
if (!strcasecmp(config_str, "incremental")) {
tasks[TASK_GC].schedule = SCHEDULE_NONE;
tasks[TASK_COMMIT_GRAPH].enabled = 1;
tasks[TASK_COMMIT_GRAPH].schedule = SCHEDULE_HOURLY;
tasks[TASK_PREFETCH].enabled = 1;
tasks[TASK_PREFETCH].schedule = SCHEDULE_HOURLY;
tasks[TASK_INCREMENTAL_REPACK].enabled = 1;
tasks[TASK_INCREMENTAL_REPACK].schedule = SCHEDULE_DAILY;
tasks[TASK_LOOSE_OBJECTS].enabled = 1;
tasks[TASK_LOOSE_OBJECTS].schedule = SCHEDULE_DAILY;
}
}
static void initialize_task_config(int schedule)
{
int i;
struct strbuf config_name = STRBUF_INIT;
gc_config();
if (schedule)
initialize_maintenance_strategy();
for (i = 0; i < TASK__COUNT; i++) {
int config_value;
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
char *config_str;
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
strbuf_reset(&config_name);
strbuf_addf(&config_name, "maintenance.%s.enabled",
tasks[i].name);
if (!git_config_get_bool(config_name.buf, &config_value))
tasks[i].enabled = config_value;
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
strbuf_reset(&config_name);
strbuf_addf(&config_name, "maintenance.%s.schedule",
tasks[i].name);
if (!git_config_get_string(config_name.buf, &config_str)) {
tasks[i].schedule = parse_schedule(config_str);
free(config_str);
}
}
strbuf_release(&config_name);
}
static int task_option_parse(const struct option *opt,
const char *arg, int unset)
{
int i, num_selected = 0;
struct maintenance_task *task = NULL;
BUG_ON_OPT_NEG(unset);
for (i = 0; i < TASK__COUNT; i++) {
if (tasks[i].selected_order >= 0)
num_selected++;
if (!strcasecmp(tasks[i].name, arg)) {
task = &tasks[i];
}
}
if (!task) {
error(_("'%s' is not a valid task"), arg);
return 1;
}
if (task->selected_order >= 0) {
error(_("task '%s' cannot be selected multiple times"), arg);
return 1;
}
task->selected_order = num_selected + 1;
return 0;
}
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
static int maintenance_run(int argc, const char **argv, const char *prefix)
{
int i;
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
struct maintenance_run_opts opts;
struct option builtin_maintenance_run_options[] = {
OPT_BOOL(0, "auto", &opts.auto_flag,
N_("run tasks based on the state of the repository")),
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
OPT_CALLBACK(0, "schedule", &opts.schedule, N_("frequency"),
N_("run tasks based on frequency"),
maintenance_opt_schedule),
OPT_BOOL(0, "quiet", &opts.quiet,
N_("do not report progress or other information over stderr")),
OPT_CALLBACK_F(0, "task", NULL, N_("task"),
N_("run a specific task"),
PARSE_OPT_NONEG, task_option_parse),
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
OPT_END()
};
memset(&opts, 0, sizeof(opts));
opts.quiet = !isatty(2);
for (i = 0; i < TASK__COUNT; i++)
tasks[i].selected_order = -1;
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
argc = parse_options(argc, argv, prefix,
builtin_maintenance_run_options,
builtin_maintenance_run_usage,
PARSE_OPT_STOP_AT_NON_OPTION);
maintenance: add --schedule option and config Maintenance currently triggers when certain data-size thresholds are met, such as number of pack-files or loose objects. Users may want to run certain maintenance tasks based on frequency instead. For example, a user may want to perform a 'prefetch' task every hour, or 'gc' task every day. To help these users, update the 'git maintenance run' command to include a '--schedule=<frequency>' option. The allowed frequencies are 'hourly', 'daily', and 'weekly'. These values are also allowed in a new config value 'maintenance.<task>.schedule'. The 'git maintenance run --schedule=<frequency>' checks the '*.schedule' config value for each enabled task to see if the configured frequency is at least as frequent as the frequency from the '--schedule' argument. We use the following order, for full clarity: 'hourly' > 'daily' > 'weekly' Use new 'enum schedule_priority' to track these values numerically. The following cron table would run the scheduled tasks with the correct frequencies: 0 1-23 * * * git -C <repo> maintenance run --schedule=hourly 0 0 * * 1-6 git -C <repo> maintenance run --schedule=daily 0 0 * * 0 git -C <repo> maintenance run --schedule=weekly This cron schedule will run --schedule=hourly every hour except at midnight. This avoids a concurrent run with the --schedule=daily that runs at midnight every day except the first day of the week. This avoids a concurrent run with the --schedule=weekly that runs at midnight on the first day of the week. Since --schedule=daily also runs the 'hourly' tasks and --schedule=weekly runs the 'hourly' and 'daily' tasks, we will still see all tasks run with the proper frequencies. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:15 +03:00
if (opts.auto_flag && opts.schedule)
die(_("use at most one of --auto and --schedule=<frequency>"));
initialize_task_config(opts.schedule);
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
if (argc != 0)
usage_with_options(builtin_maintenance_run_usage,
builtin_maintenance_run_options);
return maintenance_run_tasks(&opts);
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
}
static int maintenance_register(void)
{
char *config_value;
struct child_process config_set = CHILD_PROCESS_INIT;
struct child_process config_get = CHILD_PROCESS_INIT;
/* Disable foreground maintenance */
git_config_set("maintenance.auto", "false");
/* Set maintenance strategy, if unset */
if (!git_config_get_string("maintenance.strategy", &config_value))
free(config_value);
else
git_config_set("maintenance.strategy", "incremental");
config_get.git_cmd = 1;
strvec_pushl(&config_get.args, "config", "--global", "--get",
"--fixed-value", "maintenance.repo",
the_repository->worktree ? the_repository->worktree
: the_repository->gitdir,
NULL);
config_get.out = -1;
if (start_command(&config_get))
return error(_("failed to run 'git config'"));
/* We already have this value in our config! */
if (!finish_command(&config_get))
return 0;
config_set.git_cmd = 1;
strvec_pushl(&config_set.args, "config", "--add", "--global", "maintenance.repo",
the_repository->worktree ? the_repository->worktree
: the_repository->gitdir,
NULL);
return run_command(&config_set);
}
static int maintenance_unregister(void)
{
struct child_process config_unset = CHILD_PROCESS_INIT;
config_unset.git_cmd = 1;
strvec_pushl(&config_unset.args, "config", "--global", "--unset",
"--fixed-value", "maintenance.repo",
the_repository->worktree ? the_repository->worktree
: the_repository->gitdir,
NULL);
return run_command(&config_unset);
}
maintenance: add start/stop subcommands Add new subcommands to 'git maintenance' that start or stop background maintenance using 'cron', when available. This integration is as simple as I could make it, barring some implementation complications. The schedule is laid out as follows: 0 1-23 * * * $cmd maintenance run --schedule=hourly 0 0 * * 1-6 $cmd maintenance run --schedule=daily 0 0 * * 0 $cmd maintenance run --schedule=weekly where $cmd is a properly-qualified 'git for-each-repo' execution: $cmd=$path/git --exec-path=$path for-each-repo --config=maintenance.repo where $path points to the location of the Git executable running 'git maintenance start'. This is critical for systems with multiple versions of Git. Specifically, macOS has a system version at '/usr/bin/git' while the version that users can install resides at '/usr/local/bin/git' (symlinked to '/usr/local/libexec/git-core/git'). This will also use your locally-built version if you build and run this in your development environment without installing first. This conditional schedule avoids having cron launch multiple 'git for-each-repo' commands in parallel. Such parallel commands would likely lead to the 'hourly' and 'daily' tasks competing over the object database lock. This could lead to to some tasks never being run! Since the --schedule=<frequency> argument will run all tasks with _at least_ the given frequency, the daily runs will also run the hourly tasks. Similarly, the weekly runs will also run the daily and hourly tasks. The GIT_TEST_CRONTAB environment variable is not intended for users to edit, but instead as a way to mock the 'crontab [-l]' command. This variable is set in test-lib.sh to avoid a future test from accidentally running anything with the cron integration from modifying the user's schedule. We use GIT_TEST_CRONTAB='test-tool crontab <file>' in our tests to check how the schedule is modified in 'git maintenance (start|stop)' commands. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:18 +03:00
#define BEGIN_LINE "# BEGIN GIT MAINTENANCE SCHEDULE"
#define END_LINE "# END GIT MAINTENANCE SCHEDULE"
static int update_background_schedule(int run_maintenance)
{
int result = 0;
int in_old_region = 0;
struct child_process crontab_list = CHILD_PROCESS_INIT;
struct child_process crontab_edit = CHILD_PROCESS_INIT;
FILE *cron_list, *cron_in;
const char *crontab_name;
struct strbuf line = STRBUF_INIT;
struct lock_file lk;
char *lock_path = xstrfmt("%s/schedule", the_repository->objects->odb->path);
if (hold_lock_file_for_update(&lk, lock_path, LOCK_NO_DEREF) < 0)
return error(_("another process is scheduling background maintenance"));
crontab_name = getenv("GIT_TEST_CRONTAB");
if (!crontab_name)
crontab_name = "crontab";
strvec_split(&crontab_list.args, crontab_name);
strvec_push(&crontab_list.args, "-l");
crontab_list.in = -1;
crontab_list.out = dup(lk.tempfile->fd);
crontab_list.git_cmd = 0;
if (start_command(&crontab_list)) {
result = error(_("failed to run 'crontab -l'; your system might not support 'cron'"));
goto cleanup;
}
/* Ignore exit code, as an empty crontab will return error. */
finish_command(&crontab_list);
/*
* Read from the .lock file, filtering out the old
* schedule while appending the new schedule.
*/
cron_list = fdopen(lk.tempfile->fd, "r");
rewind(cron_list);
strvec_split(&crontab_edit.args, crontab_name);
crontab_edit.in = -1;
crontab_edit.git_cmd = 0;
if (start_command(&crontab_edit)) {
result = error(_("failed to run 'crontab'; your system might not support 'cron'"));
goto cleanup;
}
cron_in = fdopen(crontab_edit.in, "w");
if (!cron_in) {
result = error(_("failed to open stdin of 'crontab'"));
goto done_editing;
}
while (!strbuf_getline_lf(&line, cron_list)) {
if (!in_old_region && !strcmp(line.buf, BEGIN_LINE))
in_old_region = 1;
else if (in_old_region && !strcmp(line.buf, END_LINE))
maintenance: add start/stop subcommands Add new subcommands to 'git maintenance' that start or stop background maintenance using 'cron', when available. This integration is as simple as I could make it, barring some implementation complications. The schedule is laid out as follows: 0 1-23 * * * $cmd maintenance run --schedule=hourly 0 0 * * 1-6 $cmd maintenance run --schedule=daily 0 0 * * 0 $cmd maintenance run --schedule=weekly where $cmd is a properly-qualified 'git for-each-repo' execution: $cmd=$path/git --exec-path=$path for-each-repo --config=maintenance.repo where $path points to the location of the Git executable running 'git maintenance start'. This is critical for systems with multiple versions of Git. Specifically, macOS has a system version at '/usr/bin/git' while the version that users can install resides at '/usr/local/bin/git' (symlinked to '/usr/local/libexec/git-core/git'). This will also use your locally-built version if you build and run this in your development environment without installing first. This conditional schedule avoids having cron launch multiple 'git for-each-repo' commands in parallel. Such parallel commands would likely lead to the 'hourly' and 'daily' tasks competing over the object database lock. This could lead to to some tasks never being run! Since the --schedule=<frequency> argument will run all tasks with _at least_ the given frequency, the daily runs will also run the hourly tasks. Similarly, the weekly runs will also run the daily and hourly tasks. The GIT_TEST_CRONTAB environment variable is not intended for users to edit, but instead as a way to mock the 'crontab [-l]' command. This variable is set in test-lib.sh to avoid a future test from accidentally running anything with the cron integration from modifying the user's schedule. We use GIT_TEST_CRONTAB='test-tool crontab <file>' in our tests to check how the schedule is modified in 'git maintenance (start|stop)' commands. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:18 +03:00
in_old_region = 0;
else if (!in_old_region)
fprintf(cron_in, "%s\n", line.buf);
maintenance: add start/stop subcommands Add new subcommands to 'git maintenance' that start or stop background maintenance using 'cron', when available. This integration is as simple as I could make it, barring some implementation complications. The schedule is laid out as follows: 0 1-23 * * * $cmd maintenance run --schedule=hourly 0 0 * * 1-6 $cmd maintenance run --schedule=daily 0 0 * * 0 $cmd maintenance run --schedule=weekly where $cmd is a properly-qualified 'git for-each-repo' execution: $cmd=$path/git --exec-path=$path for-each-repo --config=maintenance.repo where $path points to the location of the Git executable running 'git maintenance start'. This is critical for systems with multiple versions of Git. Specifically, macOS has a system version at '/usr/bin/git' while the version that users can install resides at '/usr/local/bin/git' (symlinked to '/usr/local/libexec/git-core/git'). This will also use your locally-built version if you build and run this in your development environment without installing first. This conditional schedule avoids having cron launch multiple 'git for-each-repo' commands in parallel. Such parallel commands would likely lead to the 'hourly' and 'daily' tasks competing over the object database lock. This could lead to to some tasks never being run! Since the --schedule=<frequency> argument will run all tasks with _at least_ the given frequency, the daily runs will also run the hourly tasks. Similarly, the weekly runs will also run the daily and hourly tasks. The GIT_TEST_CRONTAB environment variable is not intended for users to edit, but instead as a way to mock the 'crontab [-l]' command. This variable is set in test-lib.sh to avoid a future test from accidentally running anything with the cron integration from modifying the user's schedule. We use GIT_TEST_CRONTAB='test-tool crontab <file>' in our tests to check how the schedule is modified in 'git maintenance (start|stop)' commands. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:18 +03:00
}
if (run_maintenance) {
struct strbuf line_format = STRBUF_INIT;
const char *exec_path = git_exec_path();
fprintf(cron_in, "%s\n", BEGIN_LINE);
fprintf(cron_in,
"# The following schedule was created by Git\n");
fprintf(cron_in, "# Any edits made in this region might be\n");
fprintf(cron_in,
"# replaced in the future by a Git command.\n\n");
strbuf_addf(&line_format,
"%%s %%s * * %%s \"%s/git\" --exec-path=\"%s\" for-each-repo --config=maintenance.repo maintenance run --schedule=%%s\n",
exec_path, exec_path);
fprintf(cron_in, line_format.buf, "0", "1-23", "*", "hourly");
fprintf(cron_in, line_format.buf, "0", "0", "1-6", "daily");
fprintf(cron_in, line_format.buf, "0", "0", "0", "weekly");
strbuf_release(&line_format);
fprintf(cron_in, "\n%s\n", END_LINE);
}
fflush(cron_in);
fclose(cron_in);
close(crontab_edit.in);
done_editing:
if (finish_command(&crontab_edit)) {
result = error(_("'crontab' died"));
goto cleanup;
}
fclose(cron_list);
cleanup:
rollback_lock_file(&lk);
return result;
}
static int maintenance_start(void)
{
if (maintenance_register())
warning(_("failed to add repo to global config"));
return update_background_schedule(1);
}
static int maintenance_stop(void)
{
return update_background_schedule(0);
}
static const char builtin_maintenance_usage[] = N_("git maintenance <subcommand> [<options>]");
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
int cmd_maintenance(int argc, const char **argv, const char *prefix)
{
if (argc < 2 ||
(argc == 2 && !strcmp(argv[1], "-h")))
usage(builtin_maintenance_usage);
if (!strcmp(argv[1], "run"))
return maintenance_run(argc - 1, argv + 1, prefix);
maintenance: add start/stop subcommands Add new subcommands to 'git maintenance' that start or stop background maintenance using 'cron', when available. This integration is as simple as I could make it, barring some implementation complications. The schedule is laid out as follows: 0 1-23 * * * $cmd maintenance run --schedule=hourly 0 0 * * 1-6 $cmd maintenance run --schedule=daily 0 0 * * 0 $cmd maintenance run --schedule=weekly where $cmd is a properly-qualified 'git for-each-repo' execution: $cmd=$path/git --exec-path=$path for-each-repo --config=maintenance.repo where $path points to the location of the Git executable running 'git maintenance start'. This is critical for systems with multiple versions of Git. Specifically, macOS has a system version at '/usr/bin/git' while the version that users can install resides at '/usr/local/bin/git' (symlinked to '/usr/local/libexec/git-core/git'). This will also use your locally-built version if you build and run this in your development environment without installing first. This conditional schedule avoids having cron launch multiple 'git for-each-repo' commands in parallel. Such parallel commands would likely lead to the 'hourly' and 'daily' tasks competing over the object database lock. This could lead to to some tasks never being run! Since the --schedule=<frequency> argument will run all tasks with _at least_ the given frequency, the daily runs will also run the hourly tasks. Similarly, the weekly runs will also run the daily and hourly tasks. The GIT_TEST_CRONTAB environment variable is not intended for users to edit, but instead as a way to mock the 'crontab [-l]' command. This variable is set in test-lib.sh to avoid a future test from accidentally running anything with the cron integration from modifying the user's schedule. We use GIT_TEST_CRONTAB='test-tool crontab <file>' in our tests to check how the schedule is modified in 'git maintenance (start|stop)' commands. Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-11 20:49:18 +03:00
if (!strcmp(argv[1], "start"))
return maintenance_start();
if (!strcmp(argv[1], "stop"))
return maintenance_stop();
if (!strcmp(argv[1], "register"))
return maintenance_register();
if (!strcmp(argv[1], "unregister"))
return maintenance_unregister();
maintenance: create basic maintenance runner The 'gc' builtin is our current entrypoint for automatically maintaining a repository. This one tool does many operations, such as repacking the repository, packing refs, and rewriting the commit-graph file. The name implies it performs "garbage collection" which means several different things, and some users may not want to use this operation that rewrites the entire object database. Create a new 'maintenance' builtin that will become a more general- purpose command. To start, it will only support the 'run' subcommand, but will later expand to add subcommands for scheduling maintenance in the background. For now, the 'maintenance' builtin is a thin shim over the 'gc' builtin. In fact, the only option is the '--auto' toggle, which is handed directly to the 'gc' builtin. The current change is isolated to this simple operation to prevent more interesting logic from being lost in all of the boilerplate of adding a new builtin. Use existing builtin/gc.c file because we want to share code between the two builtins. It is possible that we will have 'maintenance' replace the 'gc' builtin entirely at some point, leaving 'git gc' as an alias for some specific arguments to 'git maintenance run'. Create a new test_subcommand helper that allows us to test if a certain subcommand was run. It requires storing the GIT_TRACE2_EVENT logs in a file. A negation mode is available that will be used in later tests. Helped-by: Jonathan Nieder <jrnieder@gmail.com> Signed-off-by: Derrick Stolee <dstolee@microsoft.com> Signed-off-by: Junio C Hamano <gitster@pobox.com>
2020-09-17 21:11:42 +03:00
die(_("invalid subcommand: %s"), argv[1]);
}