Currently we have the concept of a "suite" and a "flavour" in our task
configuration. Typically, the "suite" refers to the high-level test harness
like "mochitest" or "reftest", whereas the flavour is more specific, e.g
"browser-chrome-instrumentation" or "crashtest". However the line between suite
and flavour is not applied with any semblance of consistency which results in
inconsistent naming throughout the tree.
This patch gets rid of the concept of "flavours" entirely (at least when it
comes to task configuration). A suite is a type of test run, for example:
- mochitest-plain
- mochitest-devtools-chrome
- mochitest-browser-chrome-instrumentation
- jsreftest
- reftest
- firefox-ui-functional-remote
etc
There is no confusion here between suites and flavours because flavours don't
exist. However, there are a couple of places where we *do* need to know what
"test harness" is used to run a suite. These cases are:
1. For SCHEDULES moz.build rules
2. For the desktop_unittest.py mozharness script which takes arguments like
--mochitest-suite=browser (this is not a compelling use of this information
and should be refactored to work more like the android_emulator_unittest.py
script)
So to get this information, this patch introduces a new concept of a "category"
which is the overall "test harness" that runs the suite. For many suites, the
"category" is identical to the suite name. Unlike flavours, "categories" have
no bearing on how we call or refer to the suite.
Differential Revision: https://phabricator.services.mozilla.com/D27554
--HG--
extra : moz-landing-system : lando
We are starting to spin off more and more "variants" of test suites. These are
usually just duplicates of our pre-existing tasks, except with an additional
pref set.
Currently there are two variants (serviceworker-e10s and socketprocess-e10s),
but a third will be added soon (fission). This change ensures we handle these
types of requests in a consistent and well defined manner. It also splits tasks
in a loop, so we don't accidentally risk combinatorial explosion.
Variants should typically be reserved for very large changes that will impact
the entire codebase (think e10s).
Differential Revision: https://phabricator.services.mozilla.com/D28061
--HG--
extra : moz-landing-system : lando
Many tasks (release tasks and cached tasks, in particular) should be re-run rather
than retriggered. Disable retrigger action for those tasks by default.
Differential Revision: https://phabricator.services.mozilla.com/D27206
--HG--
extra : moz-landing-system : lando
When we set the nightly attribute the tasks don't run on-push, so we use a new attribute.
Differential Revision: https://phabricator.services.mozilla.com/D22709
--HG--
extra : moz-landing-system : lando
There are several kinds that cache tasks based on the inputs that go into the task. Historically,
these inputs included the name of upstream tasks. This change these tasks to include the digest
of the upstream tasks.
This also bumps the version of the docker and toolchain as every digest is changed for them.
Differential Revision: https://phabricator.services.mozilla.com/D11949
--HG--
extra : moz-landing-system : lando
If a task has `required_signoffs` that match a release's `required_signoffs`, we should defer running that task until we have a matching `signoff_url`.
- add filter_out_missing_signoffs
- add transforms changes to inherit upstream `required_signoffs`
- add `mar-signing` `required_signoffs` to the `partials-signing`, `mar-signing`, and `mar-signing-l10n` kinds
Differential Revision: https://phabricator.services.mozilla.com/D11734
--HG--
extra : moz-landing-system : lando
If a task has `required_signoffs` that match a release's `required_signoffs`, we should defer running that task until we have a matching `signoff_url`.
- add filter_out_missing_signoffs
- add transforms changes to inherit upstream `required_signoffs`
- add `mar-signing` `required_signoffs` to the `partials-signing`, `mar-signing`, and `mar-signing-l10n` kinds
Differential Revision: https://phabricator.services.mozilla.com/D11734
--HG--
extra : moz-landing-system : lando
The digest for a docker image task did not include the digest for the parent
image in it, and so in particular did not depend on the versions of packages
included in a parent image.
If two branches have a docker image with identical docker files, but different
parents, this would lead to them both getting the same digest, leading to
unexpected interference between the branches.
This fixes things by including the digest of the parent image as input into the
digest of child images.
Differential Revision: https://phabricator.services.mozilla.com/D11807
--HG--
extra : moz-landing-system : lando
This adds a transform which will evaluate the 'serviceworker-e10s' key. If
False, the task is returned as is. If True, the task is run with the
'dom.serviceWorkers.parent_intercept' pref enabled. If 'both' the task is split
into two tasks, one with the pref, and one without.
Differential Revision: https://phabricator.services.mozilla.com/D7480
--HG--
extra : moz-landing-system : lando
Currently, many tasks fetch content from the Internets. A problem with
that is fetching from the Internets is unreliable: servers may have
outages or be slow; content may disappear or change out from under us.
The unreliability of 3rd party services poses a risk to Firefox CI.
If services aren't available, we could potentially not run some CI tasks.
In the worst case, we might not be able to release Firefox. That would
be bad. In fact, as I write this, gmplib.org has been unavailable for
~24 hours and Firefox CI is unable to retrieve the GMP source code.
As a result, building GCC toolchains is failing.
A solution to this is to make tasks more hermetic by depending on
fewer network services (which by definition aren't reliable over time
and therefore introduce instability).
This commit attempts to mitigate some external service dependencies
by introducing the *fetch* task kind.
The primary goal of the *fetch* kind is to obtain remote content and
re-expose it as a task artifact. By making external content available
as a cached task artifact, we allow dependent tasks to consume this
content without touching the service originally providing that
content, thus eliminating a run-time dependency and making tasks more
hermetic and reproducible over time.
We introduce a single "fetch-url" "using" flavor to define tasks that
fetch single URLs and then re-expose that URL as an artifact. Powering
this is a new, minimal "fetch" Docker image that contains a
"fetch-content" Python script that does the work for us.
We have added tasks to fetch source archives used to build the GCC
toolchains.
Fetching remote content and re-exposing it as an artifact is not
very useful by itself: the value is in having tasks use those
artifacts.
We introduce a taskgraph transform that allows tasks to define an
array of "fetches." Each entry corresponds to the name of a "fetch"
task kind. When present, the corresponding "fetch" task is added as a
dependency. And the task ID and artifact path from that "fetch" task
is added to the MOZ_FETCHES environment variable of the task depending
on it. Our "fetch-content" script has a "task-artifacts"
sub-command that tasks can execute to perform retrieval of all
artifacts listed in MOZ_FETCHES.
To prove all of this works, the code for fetching dependencies when
building GCC toolchains has been updated to use `fetch-content`. The
now-unused legacy code has been deleted.
This commit improves the reliability and efficiency of GCC toolchain
tasks. Dependencies now all come from task artifacts and should always
be available in the common case. In addition, `fetch-content` downloads
and extracts files concurrently. This makes it faster than the serial
application which we were previously using.
There are some things I don't like about this commit.
First, a new Docker image and Python script for downloading URLs feels
a bit heavyweight. The Docker image is definitely overkill as things
stand. I can eventually justify it because I want to implement support
for fetching and repackaging VCS repositories and for caching Debian
packages. These will require more packages than what I'm comfortable
installing on the base Debian image, therefore justifying a dedicated
image.
The `fetch-content static-url` sub-command could definitely be
implemented as a shell script. But Python is readily available and
is more pleasant to maintain than shell, so I wrote it in Python.
`fetch-content task-artifacts` is more advanced and writing it in
Python is more justified, IMO. FWIW, the script is Python 3 only,
which conveniently gives us access to `concurrent.futures`, which
facilitates concurrent download.
`fetch-content` also duplicates functionality found elsewhere.
generic-worker's task payload supports a "mounts" feature which
facilitates downloading remote content, including from a task
artifact. However, this feature doesn't exist on docker-worker.
So we have to implement downloading inside the task rather than
at the worker level. I concede that if all workers had generic-worker's
"mounts" feature and supported concurrent download, `fetch-content`
wouldn't need to exist.
`fetch-content` also duplicates functionality of
`mach artifact toolchain`. I probably could have used
`mach artifact toolchain` instead of writing
`fetch-content task-artifacts`. However, I didn't want to introduce
the requirement of a VCS checkout. `mach artifact toolchain` has its
origins in providing a feature to the build system. And "fetching
artifacts from tasks" is a more generic feature than that. I think
it should be implemented as a generic feature and not something that is
"toolchain" specific.
I think the best place for a generic "fetch content" feature is in
the worker, where content can be defined in the task payload. But as
explained above, that feature isn't universally available. The next
best place is probably run-task. run-task already performs generic,
very-early task preparation steps, such as performing a VCS checkout.
I would like to fold `fetch-content` into run-task and make it all
driven by environment variables. But run-task is currently Python 2
and achieving concurrency would involve a bit of programming (or
adding package dependencies). I may very well port run-task to Python
3 and then fold fetch-content into it. Or maybe we leave
`fetch-content` as a standalone script.
MozReview-Commit-ID: AGuTcwNcNJR
--HG--
extra : source : 0b941cbdca76fb2fbb98dc5bbc1a0237c69954d0
extra : histedit_source : a3e43bdd8a9a58550bef02fec3be832ca304ea93
Currently, many tasks fetch content from the Internets. A problem with
that is fetching from the Internets is unreliable: servers may have
outages or be slow; content may disappear or change out from under us.
The unreliability of 3rd party services poses a risk to Firefox CI.
If services aren't available, we could potentially not run some CI tasks.
In the worst case, we might not be able to release Firefox. That would
be bad. In fact, as I write this, gmplib.org has been unavailable for
~24 hours and Firefox CI is unable to retrieve the GMP source code.
As a result, building GCC toolchains is failing.
A solution to this is to make tasks more hermetic by depending on
fewer network services (which by definition aren't reliable over time
and therefore introduce instability).
This commit attempts to mitigate some external service dependencies
by introducing the *fetch* task kind.
The primary goal of the *fetch* kind is to obtain remote content and
re-expose it as a task artifact. By making external content available
as a cached task artifact, we allow dependent tasks to consume this
content without touching the service originally providing that
content, thus eliminating a run-time dependency and making tasks more
hermetic and reproducible over time.
We introduce a single "fetch-url" "using" flavor to define tasks that
fetch single URLs and then re-expose that URL as an artifact. Powering
this is a new, minimal "fetch" Docker image that contains a
"fetch-content" Python script that does the work for us.
We have added tasks to fetch source archives used to build the GCC
toolchains.
Fetching remote content and re-exposing it as an artifact is not
very useful by itself: the value is in having tasks use those
artifacts.
We introduce a taskgraph transform that allows tasks to define an
array of "fetches." Each entry corresponds to the name of a "fetch"
task kind. When present, the corresponding "fetch" task is added as a
dependency. And the task ID and artifact path from that "fetch" task
is added to the MOZ_FETCHES environment variable of the task depending
on it. Our "fetch-content" script has a "task-artifacts"
sub-command that tasks can execute to perform retrieval of all
artifacts listed in MOZ_FETCHES.
To prove all of this works, the code for fetching dependencies when
building GCC toolchains has been updated to use `fetch-content`. The
now-unused legacy code has been deleted.
This commit improves the reliability and efficiency of GCC toolchain
tasks. Dependencies now all come from task artifacts and should always
be available in the common case. In addition, `fetch-content` downloads
and extracts files concurrently. This makes it faster than the serial
application which we were previously using.
There are some things I don't like about this commit.
First, a new Docker image and Python script for downloading URLs feels
a bit heavyweight. The Docker image is definitely overkill as things
stand. I can eventually justify it because I want to implement support
for fetching and repackaging VCS repositories and for caching Debian
packages. These will require more packages than what I'm comfortable
installing on the base Debian image, therefore justifying a dedicated
image.
The `fetch-content static-url` sub-command could definitely be
implemented as a shell script. But Python is readily available and
is more pleasant to maintain than shell, so I wrote it in Python.
`fetch-content task-artifacts` is more advanced and writing it in
Python is more justified, IMO. FWIW, the script is Python 3 only,
which conveniently gives us access to `concurrent.futures`, which
facilitates concurrent download.
`fetch-content` also duplicates functionality found elsewhere.
generic-worker's task payload supports a "mounts" feature which
facilitates downloading remote content, including from a task
artifact. However, this feature doesn't exist on docker-worker.
So we have to implement downloading inside the task rather than
at the worker level. I concede that if all workers had generic-worker's
"mounts" feature and supported concurrent download, `fetch-content`
wouldn't need to exist.
`fetch-content` also duplicates functionality of
`mach artifact toolchain`. I probably could have used
`mach artifact toolchain` instead of writing
`fetch-content task-artifacts`. However, I didn't want to introduce
the requirement of a VCS checkout. `mach artifact toolchain` has its
origins in providing a feature to the build system. And "fetching
artifacts from tasks" is a more generic feature than that. I think
it should be implemented as a generic feature and not something that is
"toolchain" specific.
I think the best place for a generic "fetch content" feature is in
the worker, where content can be defined in the task payload. But as
explained above, that feature isn't universally available. The next
best place is probably run-task. run-task already performs generic,
very-early task preparation steps, such as performing a VCS checkout.
I would like to fold `fetch-content` into run-task and make it all
driven by environment variables. But run-task is currently Python 2
and achieving concurrency would involve a bit of programming (or
adding package dependencies). I may very well port run-task to Python
3 and then fold fetch-content into it. Or maybe we leave
`fetch-content` as a standalone script.
MozReview-Commit-ID: AGuTcwNcNJR
--HG--
extra : rebase_source : 4918b8c3bac53d63665006802054038bfbca0314
for L10n jobs should run per-push based on the corresponding builds
Differential Revision: https://phabricator.services.mozilla.com/D1406
--HG--
extra : rebase_source : 207d1c25e37ab2619a09fb209282ffe55025de26
for L10n jobs should run per-push based on the corresponding builds
Differential Revision: https://phabricator.services.mozilla.com/D1406
--HG--
extra : rebase_source : 902aaff764db6df06b96e7ba8c7d717afd6869e3
These two new attributes will help us determine which tasks belong in which release promotion graphs. In the future, we'll specify these for all shippable builds as well, and we can reduce the usage of the `product` keys. However, sometimes we need `product` to differ from `shipping-product` (e.g., `mobile` vs `fennec`; in this case we mean `stage_product` vs `shipping_product`), so I haven't yet touched those key/value pairs.
MozReview-Commit-ID: LEuf4CS277Q
--HG--
extra : rebase_source : 4971883500658ace209bec97a639a54c94292b50
Tasks that have the 'always_target' attribute set will be always be included
in the target_task_graph, regardless of target task filtering.
Furthermore, if they were only added because of this attribute (i.e, the
filters would have excluded the task), then the task will be a candidate for
optimization even if the 'optimize_target_tasks' parameter is False.
MozReview-Commit-ID: 9eoVJ5qpAMO
--HG--
extra : rebase_source : 9635002720d088ca9870649f3143d6293c666610
It is desirable for e.g. smooth toolchain transitions, to be able to
refer them with generic name from toolchain jobs, while they have more
specific names, including version numbers. For example, in a near
future, there could be a linux64-gcc-4.9 toolchain and a linux64-gcc-6.
The default would be former, but at some point we'd want to switch to
the latter, without having to change all the toolchain definitions.
Moreover, when the switch happens, it would be desirable to have some
jobs stick with the old version, which is hard to keep track of when
all the toolchain definitions for build jobs use the same versioned
toolchain. With an alias, jobs that want the default use the alias, and
jobs that want to use a specific version use the versioned toolchain
name.
--HG--
extra : rebase_source : 467d713edd00dbe358483f5ee749fa56900714dd
The toolchain jobs produce artifacts that are going to be used by other
jobs, but there is no reliable way for the decision task to know the
name of those artifacts. So we make their definition required in the
toolchain job definitions.
--HG--
extra : rebase_source : eb668cf2bf7793b44dc3d91e3ddb8d7c69e3de52
{central, autoland, inbound} are logically treated as a single unit for
many tasks and policies. Let's formalize that collection via a "trunk"
alias.
MozReview-Commit-ID: H4JPTyu2J2F
--HG--
extra : rebase_source : f4cabfc48e24b9f55b833bd95bdbf81b036cee6d
This is a more robust approach than using substring matching on task labels.
As an optimization, this simply avoids creating balrog tasks for unsigned beets
using only-for-attributes, rather than omitting them in a transform.
MozReview-Commit-ID: 8MNOxu0WgXo
--HG--
extra : source : 1aeb99ce3e6c2576b7b9b71c1cdf97a1d1889a49
This is a more robust approach than using substring matching on task labels.
As an optimization, this simply avoids creating balrog tasks for unsigned beets
using only-for-attributes, rather than omitting them in a transform.
MozReview-Commit-ID: 8MNOxu0WgXo
--HG--
extra : rebase_source : 9e93a996241bcb0d345c18813919a41320e95651
This fixes the ability to run mozbase via `-j mozbase`, with the
added advantage that it will obey `-p` too.
MozReview-Commit-ID: 1zkitUephXk
--HG--
extra : rebase_source : 0ebb65363d5f5813bc7ccb379768df54310d39c1
Nick Alexander
Andrew Halberstadt
Tom Prince
Justin Wood
Brindusan Cristian
Bastien Abadie
Dorel Luca
Byron Campen [:bwc]
Gurzau Raul
Cosmin Sabou
Simon Fraser
Aki Sasaki
Andreea Pavel
Johan Lorenzo
Narcis Beleuzu
Chris AtLee
Gregory Szorc
Bogdan Tara
Rob Wood
Aki Sasaki
Mike Hommey
Rail Aliiev
Dustin J. Mitchell
Carsten "Tomcat" Book