Summary:
For Harbormaster tasks which want to poll or wait, this lets them say "try again a little later" without having to sleep and hold a queue slot.
This is basically the same as failing, except that we don't increment the failure counter. Instead, we just set the current lease to the correct length and then exit. The task will be retried after the lease expires.
Test Plan: Using both `bin/harbormaster` and `phd debug taskmaster`, ran a lot of waiting tasks through the queue, faking them to either yield or not yield in a controlled manner. The queue responded as expected, yielding tasks appropraitely and retrying them later.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D8792
Summary:
Without this, build steps that have no options (like "wait for previous commits") don't actually save, since the transaction array is empty.
This also generally nice and consistent.
Test Plan: Created a new "wait" step, viewed transaction log.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D8791
Summary:
This hooks up all the pieces of the build pipeline so `harbormaster.sendmessage` actually works. Particularly:
- Candidate build steps (i.e., those which interact with external systems) can now "Wait for Message". This pauses them indefinitely when they complete, until something calls `harbormaster.sendmessage`.
- After processing a target, we check if we should move it to PASSED or WAITING.
- Before updating a build, we move WAITING targets with pending messages to either PASSED or FAILED.
- I added an explicit "Building" state, which doesn't affect workflows but communicates more information to human users.
A big part of this is avoiding races. I believe we get the correct behavior no matter which order events occur in:
- We update builds after targets complete and after we receive messages, so we're guaranteed to update once both these conditions are true. This means messages can't be lost (even if they arrive before a build completes).
- The minor changes to the build engine logic mean that firing additional build updates is always safe, no matter what the current state of the build is.
- The build itself is protected by a lock in the build engine.
- The target is not covered by an explicit lock, but for all states only the engine (waiting) //or// the worker (all other states) can interact with it. All of the interactions also move the target state forward to the same destination and have no other side effects.
- Messages are only consumed inside the engine lock, so they don't need an explicit lock.
Test Plan:
- Made an HTTP request wait after completion, then ran a pile of builds through it using `bin/harbormaster build` and the web UI.
- Passed and failed message-awaiting builds with `harbormaster.sendmessage`.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley, zeeg
Differential Revision: https://secure.phabricator.com/D8788
Summary: Fixes T4590. Use the credentials custom field to allow Harbormaster HTTP requests to include usernames/passwords.
Test Plan: Ran a build plan with credentials, verified they were sent to the remote server.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4590
Differential Revision: https://secure.phabricator.com/D8786
Summary:
See discussion in D8773. Three small adjustments which should help prevent this kind of issue:
- When queueing followup tasks, hold them on the worker until we finish the task, then queue them only if the work was successful.
- Increase the default lease time from 60 seconds to 2 hours. Although most tasks finish in far fewer than 60 seconds, the daemons are generally stable nowadays and these short leases don't serve much of a purpose. I think they also date from an era where lease expiry and failure were less clearly distinguished.
- Increase the default wait-after-failure from 60 seconds to 5 minutes. This largely dates from the MetaMTA era, where Facebook ran services with high failure rates and it was appropriate to repeatedly hammer them until things went through. In modern infrastructure, such failures are rare.
Test Plan:
- Verified that tasks queued properly after the main task was updated.
- Verified that leases default to 7200 seconds.
- Intentionally failed a task and verified default 300 second wait before retry.
- Removed all default leases shorter than 7200 seconds (there was only one).
- Checked all the wait before retry implementations for anything much shorter than 5 minutes (they all seem reasonable).
Reviewers: btrahan, sowedance
Reviewed By: sowedance
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D8774
Summary: Ref T1049. I'm fair sure this is just a case of bad data in my local install, but we probably don't want the default page for Harbormaster to break when there's invalid / missing container or buildable handles on any of the builds.
Test Plan: Loaded the page, didn't get a crash due to null reference.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: demo, epriestley, Korvin
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8608
Summary:
Ref T1049. Currently, the "add" dialog lets you select a build step type, but then immediately creates one. If you "cancel" from the edit screen, you end up with an empty (and almost certainly invalid) build step.
Instead, don't create the step until it's valid.
Test Plan: Add Step -> Pick Type -> Add Step -> Cancel no longer creates empty step.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8605
Summary:
Ref T1049. Allows external systems to send a message to a build target. The primary intended use case is:
- You make an HTTP request to Jenkins.
- The build goes into a "waiting" state.
- Later, Jenkins calls `harbormaster.sendmessage` to report that the target passed or failed.
- The build continues as appropriate.
This is deceptively complicated because:
- There are a lot of race concerns. We might get a message back from an external system before it even responds to the request we made. We want to make sure we process these messages no matter when we receive them.
- These messages need to be sent to a build target (vs a build or buildable) because we'll get into trouble with parallelization later on otherwise (Jenkins is told to do 3 builds; we can't tell which ones failed or what overall state is unless the message are sent to targets).
- I initially thought about implementing this as a separate "Wait for a response from an external system" build step. This gets a lot more complicated for users once we do parallelization, though. Particularly, in the case where you've told Jenkins to do 3 builds, the three "wait" steps need to know which target they're waiting for (and jenkins needs to know some unique identifier for each target). So this pretty much boils down to a more complicated, more error-prone version of using target PHIDs.
This makes the already-muddy Build UI a bit worse, but it needs a general clarity pass anyway (it's showing way too much uninteresting data, and should show a better summary of results instead).
Test Plan:
- This doesn't really do anything interesting yet.
- Used Conduit to send messages to build plans.
- Viewed the messages on the build screen.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8604
Summary: Ref T1049. Tweaks some of the UI and code to improve / clean it up a bit.
Test Plan: Ran build plans, browsed UI.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8603
Summary: Ref T1049. For consistency, rename these to "Harbormaster...".
Test Plan: Ran migration, ran builds, everything still works fine.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8602
Summary: Ref T1049. D8588 already required custom code to change what it extends, so this is as good a time as we're going to get to move to more standard class name.
Test Plan: `arc liberate`; `arc lint`
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8601
Summary:
Ref T1049. Fixes T4602. Moves all the funky field stuff to CustomField. Uses ApplicationTransactions to apply and record edits.
This makes "artifact" fields a little less nice (but still perfectly usable). With D8599, I think they're reasonable overall. We can improve this in the future.
All other field types are better (e.g., fixes weird bugs with "bool", fixes lots of weird behavior around required fields), and this gives us access to many new field types.
Test Plan:
Made a bunch of step edits. Here's an example:
{F133694}
Note that:
- "Required" fields work correctly.
- the transaction record is shown at the bottom of the page.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T4602, T1049
Differential Revision: https://secure.phabricator.com/D8600
Summary:
Ref T1049. In Harbormaster, build steps may have various inputs (like a host they should run on) and outputs (like a reference to an uploaded file).
- Currently, inputs aren't defined anywhere (except implicitly at runtime).
- Instead, define inputs explicitly.
- Currently, outputs are defined in a way that loses information when misconfigured (the keys will collide).
- Instead, define inputs and outputs so they work whether a step is configured correctly or not.
- Currently, there's no simple way to see a step's inputs and outputs.
- Add some UI for this.
- Currently, reordering steps has some surprising side effects.
- Instead of invalidating steps after reordering them, validate them at display time and warn the user.
Test Plan:
{F133679}
{F133680}
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley, chad
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8599
Summary: Ref T1049. This generally simplifies things. The steps which don't support variables generally don't make sense to support varaibles anyway.
Test Plan: Edited some steps.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D8588
Summary: Precedence here was mucked up.
Test Plan: Plan with no explicit "method" now defaults to POST correctly.
Reviewers: dctrwatson, btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D8559
Summary:
This revision adds a 'method' field to the HTTP request harbormaster build step. This allows the user to specify GET, POST, DELETE, and PUT (limited by the underlying wrapper phabricator uses for HTTP requests). I'm not sure how much sense PUT makes, but oh well.
Existing plans shouldn't break, as if this field is an empty string, we default to POST, which is the old behavior.
Fixes T4604
Test Plan: 1) Verified that the empty string does, in fact, issue a POST request. Changed the method to be GET and observed that the problem described in T4604 is resolved.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: aran, epriestley
Maniphest Tasks: T4604
Differential Revision: https://secure.phabricator.com/D8520
Summary: This can be a command, which might be arbitrarily long, but the column is VARCHAR(255).
Test Plan: `grep`
Reviewers: dctrwatson, btrahan
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D8446
Summary:
Ref T1191. Test that MySQL's rules match those of `phutil_is_utf8_with_only_bmp_characters()`:
- Build a string with //every// character that we consider to be a BMP character.
- Write it into MySQL.
- Read it back out.
- Make sure MySQL didn't truncate it.
Test Plan: Ran unit test. This test runs pretty quickly (50ms), the string with every character isn't all that enormous.
Reviewers: btrahan, arice
Reviewed By: arice
CC: chad, arice, aran
Maniphest Tasks: T1191
Differential Revision: https://secure.phabricator.com/D8314
Summary:
Ref T4379. I want project subscriptions to work like this (yell if this seems whacky, since it makes subscriptions mean somethign a little different for projects than they do for other objects):
- You can only subscribe to a project if you're a project member.
- When you're added as a member, you're added as a subscriber.
- When you're removed as a member, you're removed as a subscriber.
- While you're a member, you can optionally unsubscribe.
From a UI perspective:
- We don't show the subscriber list, since it's going to be some uninteresting subset of the member list.
- We don't show CC transactions in history, since they're an uninteresting near-approximation of the membership transactions.
- You only see the subscription controls if you're a member.
To do this, I've augmented `PhabricatorSubscribableInterface` with two new methods. It would be nice if we were on PHP 5.4+ and could just use traits for this, but we should get data about version usage before we think about this. For now, copy/paste the default implementations into every implementing class.
Then, I implemented the interface in `PhabricatorProject` but with alternate defaults.
Test Plan:
- Used the normal interaction on existing objects.
- This has no actual effect on projects, verified no subscription stuff mysteriously appeared.
- Hit the new error case by fiddling with the UI.
Reviewers: btrahan
Reviewed By: btrahan
CC: chad, aran
Maniphest Tasks: T4379
Differential Revision: https://secure.phabricator.com/D8165
Summary:
Hosted repositories have muddied this distinction somewhat. In some cases, we only want to use the real remote URI, and the call is only relevant for imported repositories.
In other cases, we want the URI we'd plug into `git clone`.
Move this logic into `PhabricatorRepository` and make the distinction more clear.
Test Plan: Viewed SVN, Git, and Mercurial hosted and remote repositories, all the URIs looked reasonable.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran, dctrwatson
Differential Revision: https://secure.phabricator.com/D8096
Summary: Fixes T4336. This updates the build engine to delete all artifacts when targets are being deleted. This prevents conflicts when builds are restarted.
Test Plan: Restarted a build that had a lease host step and it didn't crash.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T4336
Differential Revision: https://secure.phabricator.com/D8092
Summary:
Ref T2015. Several fixes:
- `checkForCancellation()` no longer exists, and isn't relevant for resumable stops. Throw it away for now.
- Fix an issue where a build could pass even if the final step failed.
- `phlog()` exceptions so they show up in `bin/harbormaster` and the daemon logs.
- Write an exception log if a step fails.
- Add a "throw an exception" step to debug this stuff more easily.
Test Plan:
- Grepped for `checkForCancellation()`.
- Ran a failing build where the final step caused the failure.
- Observed `phlog()` in `bin/harbormaster` output.
- Observed log in web UI:
{F101168}
Reviewers: btrahan, hach-que
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D7935
Summary: Ref T2015. This workflow is a little weird (runs in a dialog, no edit-before-create step, lots of internal classnames). Make it a little more standard.
Test Plan: See screenshots.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D7908
Summary: Ref T1049. Creates convenience actions at the Buildable level to stop, resume, or restart all builds.
Test Plan:
- Stopped all builds.
- Resumed all builds.
- Restarted all builds.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7899
Summary:
Ref T1049. Improves the UI:
- Pending commands, like "stopping", are shown separately from the current status.
- Pending commands are shown on the list view.
- Builds can be restarted, stopped and resumed from the list view.
- Add a missing crumb.
Test Plan:
{F99022}
{F99023}
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7898
Summary: Ref T1049. The logic in the BuildEngine is a little different from the logic on the Build itself. Make these more consistent, and make queued commands more private.
Test Plan: Restarted, stopped, and resumed a build.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7897
Summary:
Ref T1049. Currently you can cancel a build, but now that we're tracking a lot more state we can stop, resume, and restart builds.
When the user issues a command against a build, I'm writing it into an auxiliary queue (`HarbormasterBuildCommand`) and then reading them out in the worker. This is mostly to avoid race messes where we try to `save()` the object in multiple places: basically, the BuildEngine is the //only// thing that writes to Build objects, and it holds a lock while it does it.
Test Plan:
- Created a plan which runs "sleep 2" a bunch of times in a row.
- Stopped, resumed, and restarted it.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran, chad
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7892
Summary:
Ref T1049. Currently, the Harbormaster worker looks like this:
foreach (step) {
run_step(step);
}
This means steps can't ever be run in parallel. Instead, split it into two workers. The "Build" worker starts things off, and basically does:
update_build();
(We could theoretically do this in the original process because it should never take very long, but since there's a lock and it's a little bit complex it seemed cleaner to separate it.)
The "Target" worker runs an individual target (like a command, or an HTTP request, or whatever), then updates the build:
run_one_step(step);
update_build();
The new `update_build()` mechanism in `HarbormasterBuildEngine` does this, roughly:
figure_out_overall_status_of_all_steps();
if (build is done) { done(); }
if (build is fail) { fail(); }
foreach (step that is ready to run) {
queue_target_worker_for_step(step);
}
So, overall:
- The part of the code that updates Builds is completely separated from the part of the code that updates Targets.
- Targets can run in parallel.
Test Plan:
- Ran a bunch of builds via `bin/harbormaster build`.
- Ran a bunch of builds via web UI.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7890
Summary:
Ref T2015. Not directly related to Drydock, but I've wanted to do this for a bit.
Introduce a common base class for all the workflows in the scripts in `bin/*`. This slightly reduces code duplication by moving `isExecutable()` to the base, but also provides `getViewer()`. This is a little nicer than `PhabricatorUser::getOmnipotentUser()` and gives us a layer of indirection if we ever want to introduce more general viewer mechanisms in scripts.
Test Plan: Lint; ran some of the scripts.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D7838
Summary:
Ref T2015. Moves a bunch of raw object loads into modern policy-aware queries.
Also straightens out the Log and Lease policies a little bit: there are legitimate states where these objects are not attached to a resource (particularly, while a lease is being acquired). Handle these more gracefully.
Test Plan: Lint / browsed stuff.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D7836
Summary: See thread; fixes fatal. The actual name of this method is `getHarbormaster...`.
NOTE: This fixes a fatal in Differential which impedes review, so I'm pushing it as-is.
Test Plan: Browsed a revision.
Reviewers: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D7834
Summary: Ref T2015. DrydockLease predates widespread adoption of policies. Make it -- and its query -- policy aware.
Test Plan: Browsed leases from the web UI. Grepped for callsites.
Reviewers: btrahan
Reviewed By: btrahan
CC: hach-que, aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D7826
Summary:
Ref T1049. Adds `bin/harbormaster` and `bin/harbormaster build` for applying plans from the console. Since this gets `--trace`, it's much easier to debug what's going on.
This doesn't work properly with some of the Drydock steps yet, I need to look at those. I think `setRunAllTasksInProcess` probably obsoletes some of the mechanisms. It might also not work with "Wait for Builds" but I didn't check.
Test Plan: Used `bin/harbormaster` to run a bunch of builds. Ran builds from web UI.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7825
Summary:
Ref T1049. Generally, it's useful to separate test/trial/manual runs from production/automatic runs.
For example, you don't want to email a bunch of people that the build is broken just because you messed something up when writing a new build plan. You'd rather try it first, then promote it into production once you have some good runs.
Similarly, test runs generally should not affect the outside world, etc. Finally, some build steps (like "wait for other buildables") may want to behave differently when run in production/automation than when run in a testing environment (where they should probably continue immediately).
So, formalize the distinction between automatic buildables (those created passively by the system in response to events) and manual buildables (those created explicitly by users). Add filtering, and stop the automated parts of the system from interacting with the manual parts (for example, we won't show manual results on revisions).
This also moves the "Apply Build Plan" to a third, new home: instead of the sidebar or Buildables, it's now on plans. I think this generally makes more sense given how things have developed. Broadly, this improves isolation of test environments.
Test Plan: Created some builds, browsed around, used filters, etc.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7824
Summary: Ref T1049. Adds "Repository", "Revision", "Diff" and "Commit" as searchable fields.
Test Plan: Used all the fields to filter things.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7823
Summary: Ref T4195. Ref T2783. We have an old-school implementation of this; move it into a LowLevel query and make callers all run through Conduit. I need the LowLevel query for hooks, to implement an "is merge commit" Herald rule.
Test Plan:
- Ran query via Conduit for SVN, Mercurial, Git.
- Parsed a commit which closed a revision, attach/closed worked correctly.
- Browsed Diffusion.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T4195, T2783
Differential Revision: https://secure.phabricator.com/D7808
Summary: We currently have a lot of calls to `addCrumb(id(new PhabricatorCrumbView())->...)` which can be expressed much more simply with a convenience method. Nearly all crumbs are only textual.
Test Plan:
- This was mostly automated, then I cleaned up a few unusual sites manually.
- Bunch of grep / randomly clicking around.
Reviewers: btrahan, chad
Reviewed By: btrahan
CC: hach-que, aran
Differential Revision: https://secure.phabricator.com/D7787
Summary: Not every revision belongs to a repository, so we might end up here with `$repo` still equal to `null`. Don't fatal if we do.
Test Plan: iiam
Reviewers: btrahan, hach-que, zeeg
Reviewed By: hach-que
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7771
Summary: This adds a build step which will block a build from continuing if there are previous builds of the build plan still running.
Test Plan: Configured a build plan with a wait of 60 seconds and a "wait for previous builds", then started a build. While that was still building, reconfigured the plan to have a wait time of 3 seconds, started it, and saw it move into the "Waiting" status. When the 60 second build finished, both builds passed.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7745
Summary: This adds a build step in Harbormaster for publishing file artifacts as fragments in Phragment.
Test Plan:
Created a build plan with the following steps:
* Lease Host
* Upload Artifact
* Publish Fragment
Ran the build plan against a buildable and saw the fragment get created in Phragment. Ran the plan again and saw the fragment get updated with a new version. Modified the file that got uploaded and ran the plan again, checked the history of the fragment, and saw the differences represented as a Diff-Match-Patch patch.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T4205
Differential Revision: https://secure.phabricator.com/D7742
Summary: This implements support for explicitly marking the sequence of build steps. Users can now drag and re-order build steps in plans, and artifact dependencies are re-calculated so that if you move "Run Command" before "Lease Host", the "Run Command" step has it's artifact setting cleared and thus the step becomes invalid.
Test Plan: Re-ordered build steps and observed dependencies being correctly recalculated.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7715
Summary: This implements a build step for uploading an artifact from a build machine to Phabricator. It uses SFTP so that it will work on both UNIX and Windows build machines.
Test Plan: Ran an "Upload Artifact" build against a Windows machine (with FreeSSHD installed). The artifact uploaded to Phabricator, appeared on the build view and the file contents could be viewed from Phabricator.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7582
Summary: Currently the "Edit Build Plan" page crashes if there are any build steps with invalid implementations (because the implementation class has been removed or renamed). This updates the Edit Build Plan page so that steps with invalid implementations can be deleted.
Test Plan: Looked at a build plan with invalid configurations and deleted it's steps.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T4111, T1049
Differential Revision: https://secure.phabricator.com/D7708
Summary: This migrates the "Run Remote Command" build step over to use Drydock hosts and Harbormaster artifacts.
Test Plan:
Created a build plan with a "Lease Host" step and a "Run Command" step. Configured the "Run Command" step to use the artifact from the "Lease Host" step.
Saw the results:
{F87377}
{F87378}
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049, T4111
Differential Revision: https://secure.phabricator.com/D7707
Summary:
This adds LeaseHostBuildStepImplementation for getting leases on hosts in Drydock via Harbormaster. It stores the resulting lease in an artifact.
There is also a few bug fixes as well.
Test Plan: Created a build plan with a "Lease Host" build step. Ran the build plan and saw the build pass and the artifact in the database.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049, T4111
Differential Revision: https://secure.phabricator.com/D7706
Summary: This implements build targets as outlined in D7582. Build targets represent an instance of a build step particular to the build. Logs and artifacts have been adjusted to attach to build targets instead of build / build step pairs.
Test Plan: Ran builds and clicked around the interface. Everything seemed to work.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T4111, T1049
Differential Revision: https://secure.phabricator.com/D7703
Summary:
Ref T4122. Implements a credential management application for the uses described in T4122.
@chad, this needs an icon, HA HA HAHA HA BWW HA HA HA
bwahaha
Test Plan: See screenshots.
Reviewers: btrahan, chad
Reviewed By: btrahan
CC: chad, aran
Maniphest Tasks: T4122
Differential Revision: https://secure.phabricator.com/D7608
Summary:
Now that diffs have PHIDs we can create buildables for them.
This also adds `buildable.diff` in the variables list so the diff ID is available, and it also fixes the Cancel button on "Edit Plan" page so it redirects to the right place.
Test Plan: Created a buildable from a diff, ran a build plan against it that had `echo ${buildable.diff}` and got the right ID. Also tested the "Edit Plan" cancel redirect.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7546
Summary:
This uses an event listener to render the status of builds on their buildables. The revision and commit view now renders out the status of each of the builds.
Currently the revision controller has the results for the latest diff rendered out. We might want to show the status of previous diffs in the future, but for now I think the latest diff should do fine.
There's also a number of bug fixes in this diff, including a particularly nasty one where builds would have a build plan PHID generated for them, which resulted in handle lookups always returning invalid objects.
Test Plan: Ran builds against diffs and commits, saw them appear on the revision and commit view controllers.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7544
Summary:
Ref T1049. This is very minimal, but does what it says.
I merged the variable replacement code so Remote + HTTP can share more stuff.
Test Plan:
Ran "HTTP" and "Remote" build plans.
{F79886}
{F79887}
Reviewers: hach-que, btrahan
Reviewed By: hach-que
CC: zeeg, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7541
Summary: This prevents a crash in applying build plans when more than one buildable exists for the same object. It also adds a check into the "New Manual Build" page to ensure that users can't create a buildable for an object that already has one.
Test Plan: Tried to create a buildable for an object that already has one and a nice friendly error appeared. Applied a build plan to a buildable whose object has two buildables and didn't get a crash any more.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D7543
Summary: This adds a `build.id` variable, cleans up the naming convention of other variables and also fixes an issue in the remote command to read the buffers after the command finishes.
Test Plan: Ran a build with `/bin/echo ${build.id}` and saw the build ID come through.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7540
Summary: This puts back the stronger variable replacement that was missed the last update to D7519.
Test Plan: Re-ran a remote build that had variables in the command and everything worked as expected.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7539
Summary: This fixes an issue where content would be discarded when the content to append is larger than the chunk size limit.
Test Plan: Tested running a remote command that does `I=0; while true; do echo "$I"; I=$[$I+1]; done` and all of the outputted numbers matched the line numbers in the logs.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Differential Revision: https://secure.phabricator.com/D7537
Summary: This adds a build step implementation for running a command on a remote machine over SSH. It supports merging in various variables about the build (such as the commit hash / revision ID, repository call sign, version control type and clone URI).
Test Plan: Configured a build plan to run `/bin/true` on localhost and the build passed. Configured a build plan to run `/bin/false` on localhost and the build failed.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7519
Summary:
Depends on D7519.
This implements support for build logs in Harbormaster. This includes support for appending to a log from the "Run Remote Command" build step.
It also adds the ability to cancel builds.
Currently the build view page doesn't update the logs live; I'm sure this can be achieved with Javelin, but I don't have enough experience with Javelin to actually make it poll from updates to content in the background.
{F79151}
{F79153}
{F79150}
{F79152}
Test Plan:
Tested this by setting up SSH on a Windows machine and using a Remote Command configured with:
```
C:\Windows\system32\cmd.exe /C cd C:\Build && mkdir Build_${timestamp} && cd Build_${timestamp} && git clone --recursive https://github.com/hach-que/Tychaia.git && cd Tychaia && Protobuild.exe && C:\Windows\Microsoft.NET\Framework\v4.0.30319\MSBuild.exe Tychaia.Windows.sln
```
and observed the output of the build stream from the Windows machine into Phabricator.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7521
Summary:
Depends on D7500.
This seemed like a pretty good idea once I thought of it. Instead of having some custom triggering logic instead Harbormaster, I figured it best to leverage all of Herald's power so that users can create rules to apply builds to commits and differential revisions. This gives the added advantage that they can trigger off builds for particular types of revisions and commits, which seems like it could be really useful (e.g. run extra tests against revisions that touch sensitive areas of the code).
Test Plan: Ran the usual daemons + the Harbormaster daemon. Pushed a commit to the repository and saw both the buildable and build get created when the commit worked picked it up. Submitted a diff and saw both the buildable and build get created when the Herald rules were evaluated for the diff.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran, hwinkel
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7501
Summary: Ref T1049. Nothing fancy, but shows red for fail/error and green for pass. See discussion in D7502.
Test Plan: {F78839}
Reviewers: hach-que, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7512
Summary:
Depends on D7501.
This just renders the buildable's actual object name onto the list, so you can see at a glance what the buildable represents. I'd like to also pull across a list of builds of this buildable and change the bar color, but I'm not quite sure how to do that in the search architecture without N+1 querying.
Test Plan:
Looked at the buildable list and it looked like this:
{F78555}
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7502
Summary: This implements an interface for adding new build steps, editing existing build steps and deleting build steps from build plans. It uses the settings definitions on the build implementation to work out what fields should be displayed on the edit page.
Test Plan:
See screenshots:
{F78529}
{F78532}
{F78528}
{F78531}
{F78527}
{F78530}
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7500
Summary:
Depends on D7498.
This implements support for a "build step implementation". Build steps have an associated class name (which makes the class in PHP) and a details field, which is serialized JSON (same as PhabricatorRepository).
This also implements a SleepBuildStepImplementation which just pauses the build for a specified period of seconds.
Test Plan:
Inserted a build step with `insert into harbormaster_buildstep (phid, buildPlanPHID, className, details, dateCreated, dateModified) values ('', 'PHID-HMCP-zkh5w6czfbfpk2gxwdeo', 'SleepBuildStepImplementation', '{"seconds":5}', NOW(), NOW());` (adjusting the build plan PHID as appropriate).
Started the daemon and applied the build plan to a buildable, and saw the daemon take a 5 second delay after creating `SleepBuildStepImplementation`.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran, chad
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7499
Summary: This implements a basic Harbormaster daemon that takes pending builds and builds them (currently just sleeps 15 seconds before moving to passed state). It also implements an interface to apply a build plan to a buildable, so that users can kick off builds for a buildable.
Test Plan: Ran `bin/phd debug PhabricatorHarbormasterBuildDaemon` and used the interface to start some builds by applying a build plan. Observed them move from 'pending' to 'building' to 'passed'.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley
CC: Korvin, epriestley, aran
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7498
Summary:
Ref T1049. I don't really want to sink too much time into this right now, but a seemingly reasonable architecture came to me in a dream. Here's a high-level overview of how things fit together:
- **"Build"**: In Harbormaster, "build" means any process we want to run against a working copy. It might actually be building an executable, but it might also be running lint, running unit tests, generating documentation, generating symbols, running a deploy, setting up a sandcastle, etc.
- `HarbormasterBuildable`: A "buildable" is some piece of code which build operations can run on. Generally, this is either a Differential diff or a Diffusion commit. The Buildable class just wraps those objects and provides a layer of abstraction. Currently, you can manually create a buildable from a commit. In the future, this will be done automatically.
- `HarbormasterBuildStep`: A "build step" is an individual build operation, like "run lint", "run unit", "build docs", etc. The step defines how to perform the operation (for example, "run unit tests by executing 'arc unit'"). In this diff, this barely exists.
- `HarbormasterBuildPlan`: This glues together build steps into groups or sequences. For example, you might want to "run unit", and then "deploy" if the tests pass. You can create a build plan which says "run step "unit tests", then run step "deploy" on success" or whatever. In the future, these will also contain triggers/conditions ("Automatically run this build plan against every commit") and probably be able to define failure actions ("If this plan fails, send someone an email"). Because build plans will run commands, only administrators can manage them.
- `HarbormasterBuild`: This is the concrete result of running a `BuildPlan` against a `Buildable`. It tracks the build status and collects results, so you can see if the build is running/successful/failed. A `Buildable` may have several `Build`s, because you can execute more than one `BuildPlan` against it. For example, you might have a "documentation" build plan which you run continuously against HEAD, but a "unit" build plan which you want to run against every commit.
- `HarbormasterBuildTarget`: This is the concrete result of running a `BuildStep` against a `Buildable`. These are children of `Build`. A step might be able to produce multiple targets, but generally this is something like "Unit Tests" or "Lint" and has an overall status, so you can see at a glance that unit tests were fine but lint had some issues.
- `HarbormasterBuildItem`: An optional subitem for a target. For lint, this might be an individual file. For unit tests, an individual test. For normal builds, an executable. For deploys, a server. For documentation generation, there might just not be subitems.
- `HarbormasterBuildLog`: Provides extra information, like command/execution transcripts. This is where stdout/stderr will get dumped, and general details and other messages.
- `HarbormasterBuildArtifact`: Stores side effects or results from build steps. For example, something which builds a binary might put the binary in "Files" and then put its PHID here. Unit tests might put coverage information here. Generally, any build step which produces some high-level output object can use this table to record its existence.
This diff implements almost nothing and does nothing useful, but puts most of these object relationships in place. The two major things you can't easily do with these objects are:
1) Run arbitrary cron jobs. Jenkins does this, but it feels tacked on and I don't know of anyone using it for that. We could create fake Buildables to get a similar effect, but if we need to do this I'd rather do it elsewhere in general. Build and cron/service/monitoring feel like pretty different problems to me.
2) Run parameterized/matrix steps (maybe?). Bamboo has this plan/stage/task/job breakdown where a build step can generate a zillion actual jobs, like "build client on x86", "build server on x86", "build client on ARM", "build server on ARM", etc. We can sort of do this by having a Step map to multiple Targets, but I haven't really thought about it too much and it may end up being not-great. I'd guess we have like an 80% chance of getting a clean implementation if/when we get there. I suspect no one actually needs this, or when they do they'll just implement a custom Step and it can be parameterized at that level. I'm not too worried about this overall.
The major difference between this and Jenkins/Bamboo/TravisCI is that all three of those are **plan-centric**: the primary object in the system is a build plan, and the dashboard shows you all your build plans and the current status. I don't think this is the right model. One disadvantage is that you basically end up with top-level messaging that says "Trunk is broken", not "Trunk was broken by commit af32f392f". Harbormaster is **buildable-centric**: the primary object in the system is stuff you can run build operations against (commits/branches/revisions), and actual build plans are secondary. The main view will be "recent commits on this branch, and whether they're good or not" -- which I think is what's most important in a larger/more complex product -- not the pass/fail status of all jobs. This also makes it easier and more natural to integrate with Differential and Diffusion, which both care about the overall status of the commit/revision, not the current status of jobs.
Test Plan: Poked around, but this doesn't really do anything yet.
Reviewers: btrahan
Reviewed By: btrahan
CC: zeeg, chad, aran, seporaitis
Maniphest Tasks: T1049
Differential Revision: https://secure.phabricator.com/D7368
Summary: Ref T603. This swaps almost all queries against the repository table over to be policy aware.
Test Plan:
- Made an audit comment on a commit.
- Ran `save_lint.php`.
- Looked up a commit with `diffusion.getcommits`.
- Looked up lint messages with `diffusion.getlintmessages`.
- Clicked an external/submodule in Diffusion.
- Viewed main lint and repository lint in Diffusion.
- Completed and validated Owners paths in Owners.
- Executed dry runs via Herald.
- Queried for package owners with `owners.query`.
- Viewed Owners package.
- Edited Owners package.
- Viewed Owners package list.
- Executed `repository.query`.
- Viewed "Repository" tool repository list.
- Edited Arcanist project.
- Hit "Delete" on repository (this just tells you to use the CLI).
- Created a repository.
- Edited a repository.
- Ran `bin/repository list`.
- Ran `bin/search index rGTESTff45d13dffcfb3ea85b03aac8cc36251cacdf01c`
- Pushed and parsed a commit.
- Skipped all the Drydock stuff, as it it's hard to test and isn't normally reachable.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T603
Differential Revision: https://secure.phabricator.com/D7132
Summary:
This is very preliminary and doesn't actually do anything useful. In theory, it uses Drydock to check out a working copy and run tests. In practice, it's not actually capable of running any of our tests (because of complicated interdependency stuff), but does check out a working copy and //try// to run tests there.
Adds various sorts of utility methods to various things as well.
Test Plan: Ran `reparse.php --harbormaster --trace <commit>`, observed attempt to run tests via Drydock.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015, T1049
Differential Revision: https://secure.phabricator.com/D4215
Summary:
This commit doesn't change license of any file. It just makes the license implicit (inherited from LICENSE file in the root directory).
We are removing the headers for these reasons:
- It wastes space in editors, less code is visible in editor upon opening a file.
- It brings noise to diff of the first change of any file every year.
- It confuses Git file copy detection when creating small files.
- We don't have an explicit license header in other files (JS, CSS, images, documentation).
- Using license header in every file is not obligatory: http://www.apache.org/dev/apply-license.html#new.
This change is approved by Alma Chao (Lead Open Source and IP Counsel at Facebook).
Test Plan: Verified that the license survived only in LICENSE file and that it didn't modify externals.
Reviewers: epriestley, davidrecordon
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2035
Differential Revision: https://secure.phabricator.com/D3886
Summary:
A later diff adds unit tests against edges, but we need real objects to connect with edges. Add some trivial objects to the Harbormaster database to compliment the similar HarbormasterScratchTable.
On its own, this does nothing interesting.
Test Plan: Built unit tests on this in a followup.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1162
Differential Revision: https://secure.phabricator.com/D2937
Summary:
- `kill_init.php` said "Moving 1000 files" - I hope that this is not some limit in `FileFinder`.
- [src/infrastructure/celerity] `git mv utils.php map.php; git mv api/utils.php api.php`
- Comment `phutil_libraries` in `.arcconfig` and run `arc liberate`.
NOTE: `arc diff` timed out so I'm pushing it without review.
Test Plan:
/D1234
Browsed around, especially in `applications/repository/worker/commitchangeparser` and `applications/` in general.
Auditors: epriestley
Maniphest Tasks: T1103
Summary:
- We currently write every PHID we generate to a table. This was motivated by two concerns:
- **Understanding Data**: At Facebook, the data was sometimes kind of a mess. You could look at a random user in the ID tool and see 9000 assocs with random binary data attached to them, pointing at a zillion other objects with no idea how any of it got there. I originally created this table to have a canonical source of truth about PHID basics, at least. In practice, our data model has been really tidy and consistent, and we don't use any of the auxiliary data in this table (or even write it). The handle abstraction is powerful and covers essentially all of the useful data in the app, and we have human-readable types in the keys. So I don't think we have a real need here, and this table isn't serving it if we do.
- **Uniqueness**: With a unique key, we can be sure they're unique, even if we get astronomically unlucky and get a collision. But every table we use them in has a unique key anyway. So we actually get pretty much nothing here, except maybe some vague guarantee that we won't reallocate a key later if the original object is deleted. But it's hard to imagine any install will ever have a collision, given that the key space is 36^20 per object type.
- We also currently use PHIDs and Users in tests sometimes. This is silly and can break (see D2461).
- Drop the PHID database.
- Introduce a "Harbormaster" database (the eventual CI tool, after Drydock).
- Add a scratch table to the Harbormaster database for doing unit test meta-tests.
- Now, PHID generation does no writes, and unit tests are isolated from the application.
- @csilvers: This should slightly improve the performance of the large query-bound tail in D2457.
Test Plan: Ran unit tests. Ran storage upgrade.
Reviewers: btrahan, vrana, jungejason
Reviewed By: btrahan
CC: csilvers, aran, nh, edward
Differential Revision: https://secure.phabricator.com/D2466
