We found a dlm-blocked situation caused by continuous breakdown of
recovery masters described below. To solve this problem, we should
purge recovery lock once detecting recovery master goes down.
N3 N2 N1(reco master)
go down
pick up recovery lock and
begin recoverying for N2
go down
pick up recovery
lock failed, then
purge it:
dlm_purge_lockres
->DROPPING_REF is set
send deref to N1 failed,
recovery lock is not purged
find N1 go down, begin
recoverying for N1, but
blocked in dlm_do_recovery
as DROPPING_REF is set:
dlm_do_recovery
->dlm_pick_recovery_master
->dlmlock
->dlm_get_lock_resource
->__dlm_wait_on_lockres_flags(tmpres,
DLM_LOCK_RES_DROPPING_REF);
Fixes: 8c03439681 ("ocfs2/dlm: clear DROPPING_REF flag when the master goes down")
Link: http://lkml.kernel.org/r/578453AF.8030404@huawei.com
Signed-off-by: Jun Piao <piaojun@huawei.com>
Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
Reviewed-by: Jiufei Xue <xuejiufei@huawei.com>
Reviewed-by: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We always mixed in the parent pointer into the dentry name hash, but we
did it late at lookup time. It turns out that we can simplify that
lookup-time action by salting the hash with the parent pointer early
instead of late.
A few other users of our string hashes also wanted to mix in their own
pointers into the hash, and those are updated to use the same mechanism.
Hash users that don't have any particular initial salt can just use the
NULL pointer as a no-salt.
Cc: Vegard Nossum <vegard.nossum@oracle.com>
Cc: George Spelvin <linux@sciencehorizons.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
when o2hb detect a node down, it first set the dead node to recovery map
and create ocfs2rec which will replay journal for dead node. o2hb
thread then call dlm_do_local_recovery_cleanup() to delete the lock for
dead node. After the lock of dead node is gone, locks for other nodes
can be granted and may modify the meta data without replaying journal of
the dead node. The detail is described as follows.
N1 N2 N3(master)
modify the extent tree of
inode, and commit
dirty metadata to journal,
then goes down.
o2hb thread detects
N1 goes down, set
recovery map and
delete the lock of N1.
dlm_thread flush ast
for the lock of N2.
do not detect the death
of N1, so recovery map is
empty.
read inode from disk
without replaying
the journal of N1 and
modify the extent tree
of the inode that N1
had modified.
ocfs2rec recover the
journal of N1.
The modification of N2
is lost.
The modification of N1 and N2 are not serial, and it will lead to
read-only file system. We can set recovery_waiting flag to the lock
resource after delete the lock for dead node to prevent other node from
getting the lock before dlm recovery. After dlm recovery, the recovery
map on N2 is not empty, ocfs2_inode_lock_full_nested() will wait for ocfs2
recovery.
Signed-off-by: Jiufei Xue <xuejiufei@huawei.com>
Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Master returns in-progress to non-master node when it can not clear the
refmap bit right now. And non-master node will not purge the lock
resource until receiving deref done message.
Signed-off-by: xuejiufei <xuejiufei@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This series of patches is to fix the dis-order issue of setting/clearing
refmap bit described below.
Node 1 Node 2(master)
dlmlock
dlm_do_master_request
dlm_master_request_handler
-> dlm_lockres_set_refmap_bit
dlmlock succeed
dlmunlock succeed
dlm_purge_lockres
dlm_deref_handler
-> find lock resource is in
DLM_LOCK_RES_SETREF_INPROG state,
so dispatch a deref work
dlm_purge_lockres succeed.
call dlmlock again
dlm_do_master_request
dlm_master_request_handler
-> dlm_lockres_set_refmap_bit
deref work trigger, call
dlm_lockres_clear_refmap_bit
to clear Node 1 from refmap
dlm_purge_lockres succeed
dlm_send_remote_lock_request
return DLM_IVLOCKID because
the lockres is not exist
BUG if the lockres is $RECOVERY
This series of patches add a new message to keep the order of set and
clear. Other nodes can purge the lock resource only after the refmap bit
on master is cleared.
This patch is to add DEREF_DONE message and corresponding handler. Node
can purge the lock resource after receiving this message. As a new
message is added, so increase the minor number of dlm protocol version.
Signed-off-by: xuejiufei <xuejiufei@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Reviewed-by: Joseph Qi <joseph.qi@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Refer to cluster/tcp.h, NET_MAX_PAYLOAD_BYTES is a typo for
O2NET_MAX_PAYLOAD_BYTES.
Since currently DLM_MIG_LOCKRES_RESERVED is not actually used, it won't
cause any problem. But we'd better correct it for further use.
Signed-off-by: Joseph Qi <joseph.qi@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Joseph Qi <joseph.qi@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
lksb flags are defined both in dlmapi.h and dlmcommon.h. So clean them
up from dlmcommon.h.
Signed-off-by: Joseph Qi <joseph.qi@huawei.com>
Reviewed-by: Jiufei Xue <xuejiufei@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
__dlm_wait_on_lockres_flags_set() is declared but not implemented and
used. So remove it.
Signed-off-by: Joseph Qi <joseph.qi@huawei.com>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When workqueue is delayed, it may occur that a lockres is purged while it
is still queued for master assert. it may trigger BUG() as follows.
N1 N2
dlm_get_lockres()
->dlm_do_master_requery
is the master of lockres,
so queue assert_master work
dlm_thread() start running
and purge the lockres
dlm_assert_master_worker()
send assert master message
to other nodes
receiving the assert_master
message, set master to N2
dlmlock_remote() send create_lock message to N2, but receive DLM_IVLOCKID,
if it is RECOVERY lockres, it triggers the BUG().
Another BUG() is triggered when N3 become the new master and send
assert_master to N1, N1 will trigger the BUG() because owner doesn't
match. So we should not purge lockres when it is queued for assert
master.
Signed-off-by: joyce.xue <xuejiufei@huawei.com>
Reviewed-by: Mark Fasheh <mfasheh@suse.de>
Cc: Joel Becker <jlbec@evilplan.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
dlm_recovery_ctxt.received is unused.
ocfs2_should_refresh_lock_res() can only return 0 or 1, so the error
handling code in ocfs2_super_lock() is unneeded.
Signed-off-by: joyce.xue <xuejiufei@huawei.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
[dan.carpenter@oracle.com: fix up some NULL dereference bugs]
Signed-off-by: Dong Fang <yp.fangdong@gmail.com>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Jeff Liu <jeff.liu@oracle.com>
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch adds a new dlm message DLM_BEGIN_EXIT_DOMAIN_MSG and ups the dlm
protocol to 1.2.
o2dlm sends this new message in dlm_unregister_domain() to mark the beginning
of the exit domain. This message is sent to all nodes in the domain.
Currently o2dlm has no way of informing other nodes of its impending exit.
This information is useful as the other nodes could disregard the exiting
node in certain operations. For example, in resource migration. If two or
more nodes were umounting in parallel, it would be more efficient if o2dlm
were to choose a non-exiting node to be the new master node rather than an
exiting one.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Reviewed-by: Mark Fasheh <mfasheh@suse.com>
Signed-off-by: Joel Becker <jlbec@evilplan.org>
Patch cleans up the gunk added by commit 388c4bcb4e.
dlm_is_lockres_migrateable() now returns 1 if lockresource is deemed
migrateable and 0 if not.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <jlbec@evilplan.org>
In o2dlm, the enumerated message values are part of the protocol.
The patch hard codes each value so as to reduce the chance of an editing
error causing a protocol mismatch.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Adds new dlm message DLM_QUERY_NODEINFO that sends the attributes of all
registered nodes. This message is sent if the negotiated dlm protocol is
1.1 or higher. If the information of the joining node does not match
that of any existing nodes, the join domain request is rejected.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Adds new dlm message DLM_QUERY_REGION that sends the names of all active
heartbeat regions. This message is only sent in the global heartbeat
mode. If the regions in the joining node do not fully match the ones in
the active nodes, the join domain request is rejected.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
While umounting, a block mle doesn't get freed if dlm is shutdown after
master request is received but before assert master. This results in unclean
shutdown of dlm domain.
This patch frees all mles that lie around after other nodes were notified about
exiting the dlm and marking dlm state as leaving. Only block mles are expected
to be around, so we log ERROR for other mles but still free them.
Signed-off-by: Srinivas Eeda <srinivas.eeda@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Currently we process a dirty lockres with the lockres->spinlock taken. While
during the process, we may need to lock on dlm->ast_lock. This breaks the
dependency of dlm->ast_lock(lock first) and lockres->spinlock(lock second).
This patch fixes the problem.
Since we can't release lockres->spinlock, we have to take dlm->ast_lock
just before taking the lockres->spinlock and release it after lockres->spinlock
is released. And use __dlm_queue_bast()/__dlm_queue_ast(), the nolock version,
in dlm_shuffle_lists(). There are no too many locks on a lockres, so there is no
performance harm.
Signed-off-by: Wengang Wang <wen.gang.wang@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Lockres hash size of 16KB is far too small for large filesystems (where we
have hundreds of thousands of lock resources stored in the table).
This patch increases it to 128KB.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
This patch removes struct dlm_lock_name and adds the entries directly
to struct dlm_master_list_entry. Under the new scheme, both mles that
are backed by a lockres or not, will have the name populated in mle->mname.
This allows us to get rid of code that was figuring out the location of
the mle name.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch inlines dlm_set_lockres_owner() and dlm_change_lockres_owner().
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch replaces the lockres counts that tracked the number number of
locally and remotely mastered lockres' with a current and total count. The
total count is the number of lockres' that have been created since the dlm
domain was created.
The number of locally and remotely mastered counts can be computed using
the locking_state output.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
The lifetime of a mle is limited to the duration of the lockres mastery
process. While typically this lifetime is fairly short, we have noticed
the number of mles explode under certain circumstances. This patch tracks
the number of each different types of mles and should help us determine
how best to speed up the mastery process.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
With this patch, the mles are stored in a hash and not a simple list.
This should improve the mle lookup time when the number of outstanding
masteries is large.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch adds code to create and destroy the dlm->master_hash.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
For master mle, the name it stored in the attached lockres in struct qstr.
For block and migration mle, the name is stored inline in struct dlm_lock_name.
This patch attempts to make struct dlm_lock_name look like a struct qstr. While
we could use struct qstr, we don't because we want to avoid having to malloc
and free the lockname string as the mle's lifetime is fairly short.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch encapsulates adding and removing of the mle from the
dlm->master_list. This patch is part of the series of patches that
converts the mle list to a mle hash.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch adds a new lock, dlm->tracking_lock, to protect adding/removing
lockres' to/from the dlm->tracking_list. We were previously using dlm->spinlock
for the same, but that proved inadequate as we could be freeing a lockres from
a context that did not hold that lock. As the new lock only protects this list,
we can explicitly take it when removing the lockres from the tracking list.
This bug was exposed when testing multiple processes concurrently flock() the
same file.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch moves some mle related definitions from dlmmaster.c
to dlmcommon.h. Future patches need these definitions to dump mle
debugging information.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.beckeroracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch dumps the dlm state (dlm_ctxt) into a debugfs file.
Useful for debugging.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch creates the debugfs directories that will hold the
files to be used to dump the dlm state.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch links all the lockres' to a tracking list in dlm_ctxt.
We will use this in an upcoming patch that will walk the entire
list and to dump the lockres states to a debugfs file.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
This patch makes the o2dlm allocate memory for lockres, lockname and lock
structures from slabcaches rather than kmalloc. This allows us to not only
make these allocs more efficient but also allows us to track the memory being
consumed by these structures.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mfasheh@suse.com>
During migration, the recovery master node may be asked to master a lockres
it may not know about. In that case, it would not only have to create a
lockres and add it to the hash, but also remember to to do the _put_
corresponding to the kref_init in dlm_init_lockres(), as soon as the migration
is completed. Yes, we don't wait for the dlm_purge_lockres() to do that
matching put. Note the ref added for it being in the hash protects the lockres
from being freed prematurely.
This patch adds that missing put, as described above, to plug a memleak.
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
struct dlm_query_join_packet is made up of four one-byte fields. They
are effectively in big-endian order already. However, little-endian
machines swap them before putting the packet on the wire (because
query_join's response is a status, and that status is treated as a u32
on the wire). Thus, a big-endian and little-endian machines will
treat this structure differently.
The solution is to have little-endian machines swap the structure when
converting from the structure to the u32 representation.
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
Currently, when ocfs2 nodes connect via TCP, they advertise their
compatibility level. If the versions do not match, two nodes cannot speak
to each other and they disconnect. As a result, this provides no forward or
backwards compatibility.
This patch implements a simple protocol negotiation at the dlm level by
introducing a major/minor version number scheme for entities that
communicate. Specifically, o2dlm has a major/minor version for interaction
with o2dlm on other nodes, and ocfs2 itself has a major/minor version for
interacting with the filesystem on other nodes.
This will allow rolling upgrades of ocfs2 clusters when changes to the
locking or network protocols can be done in a backwards compatible manner.
In those cases, only the minor number is changed and the negotatied protocol
minor is returned from dlm join. In the far less likely event that a
required protocol change makes backwards compatibility impossible, we simply
bump the major number.
Signed-off-by: Joel Becker <joel.becker@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
There is a small window where a joining node may not see the node(s) that
just died but are still part of the domain. To fix this, we must disallow
join requests if the joining node has a different node map.
A new field node_map is added to dlm_query_join_request to send the current
nodes nodemap along with join request. On the receiving end the nodes that
are part of the cluster verifies if this new node sees all the nodes that
are still part of the cluster. They disallow the join if the maps mismatch.
Signed-off-by: Srinivas Eeda <srinivas.eeda@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
Eventhough the set refmap bit message is sent before the clear refmap
message, currently there is no guarentee that the set message will be
handled before the clear. This patch prevents the clear refmap to be
processed while the node is sending assert master messages to other
nodes. (The set refmap message is sent as a response to the assert
master request).
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
This patch prevents the dlm from sending the clear refmap message
before the set refmap. We use the newly created post function handler
routine to accomplish the task.
Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
Currently o2net allows one handler function per message type. This
patch adds the ability to call another function to be called after
the handler has returned the message to the other node.
Handlers are now given the option of returning a context (in the form of a
void **) which will be passed back into the post message handler function.
Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Sunil Mushran <sunil.mushran@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
dlmthread was removing lockres' from the dirty list
and resetting the dirty flag before shuffling the list.
This patch retains the dirty state flag until the lists
are shuffled.
Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Sunil Mushran <Sunil.Mushran@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
This patch makes some needlessly global functions static.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
This was previously broken and migration of some locks had to be temporarily
disabled. We use a new (and backward-incompatible) set of network messages
to account for all references to a lock resources held across the cluster.
once these are all freed, the master node may then free the lock resource
memory once its local references are dropped.
Signed-off-by: Kurt Hackel <kurt.hackel@oracle.com>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
The OCFS2 DLM uses strlen() to determine lock name length, which excludes
the possibility of putting binary values in the name string. Fix this by
requiring that string length be passed in as a parameter.
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
dlm_lockres_master_requery() became global without any external usage.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
piaojun
Linus Torvalds
Jiufei Xue
Joseph Qi
Dong Fang
Al Viro
Sunil Mushran
Srinivas Eeda
Wengang Wang
Joel Becker
Kurt Hackel
Adrian Bunk
David Howells
Mark Fasheh