1
0
Fork 0
mirror of https://we.phorge.it/source/phorge.git synced 2024-11-25 00:02:41 +01:00
Commit graph

255 commits

Author SHA1 Message Date
Rabah Meradi
0bf106eeea Fix 4 typos in code
Summary: Fixes T12516

Test Plan: grep for those typos

Reviewers: #blessed_reviewers, epriestley

Reviewed By: #blessed_reviewers, epriestley

Subscribers: epriestley, PHID-OPKG-gm6ozazyms6q6i22gyam

Maniphest Tasks: T12516

Differential Revision: https://secure.phabricator.com/D17638
2017-04-07 04:09:56 -07:00
epriestley
706c21375e Remove empty implementations of describeAutomaticCapabilities()
Summary:
This has been replaced by `PolicyCodex` after D16830. Also:

  - Rebuild Celerity map to fix grumpy unit test.
  - Fix one issue on the policy exception workflow to accommodate the new code.

Test Plan:
  - `arc unit --everything`
  - Viewed policy explanations.
  - Viewed policy errors.

Reviewers: chad

Reviewed By: chad

Subscribers: hach-que, PHID-OPKG-gm6ozazyms6q6i22gyam

Differential Revision: https://secure.phabricator.com/D16831
2016-11-09 15:24:22 -08:00
Mike Riley
8759f7e6ec Expose Drydock blueprints via Conduit
Summary:
This search engine ports cleanly to Conduit out of the box.

Ref T11694

Test Plan: called the API method from the console, browsed blueprints in the ui

Reviewers: epriestley, #blessed_reviewers

Reviewed By: epriestley, #blessed_reviewers

Subscribers: Korvin, epriestley

Maniphest Tasks: T11694

Differential Revision: https://secure.phabricator.com/D16593
2016-10-13 20:48:24 +00:00
Mike Riley
fa90f8bef4 Expose Drydock authorizations via Conduit
Summary:
`DrydockAuthorizationSearchEngine` was being used solely to display authorizations for a specific blueprint from the web UI and consequently expected that callers set a specific blueprint before performing a query. Here we check to see if a blueprint has been set in cases where the engine could be operating from either Conduit or the web.

Ref T11694

Test Plan:
 - called the API method from the console
 - approved an authorization
 - followed the "view all" link from a blueprint page

Reviewers: #blessed_reviewers, epriestley

Reviewed By: #blessed_reviewers, epriestley

Subscribers: epriestley

Maniphest Tasks: T11694

Differential Revision: https://secure.phabricator.com/D16592
2016-10-11 19:55:43 +00:00
Chad Little
e7aa874f5e Fix getIcon calls in PHUIObjectListItem
Summary: Fixes T11501. Let's you pass in a full PHUIIconView or just the icon name to give ObjectListItem a large icon.

Test Plan: Alamanac, Applications, Drydock, Settings, Search Typeahead, Config page...

Reviewers: epriestley

Reviewed By: epriestley

Subscribers: Korvin, PHID-OPKG-gm6ozazyms6q6i22gyam

Maniphest Tasks: T11501

Differential Revision: https://secure.phabricator.com/D16421
2016-08-19 09:35:09 -07:00
epriestley
65980ac683 Convert all remaining old tabs to new PHUITabGroupViews
Summary: Ref T10628. This moves everything else over. I'll clean up the cruft in the next diff.

Test Plan:
- Viewed Conduit API page, toggled tabs.
- Viewed Harbormaster build, toggled tabs.
- Viewed a Drydock lease, swapped tabs.
- Viewed a Drydock resource, swapped tabs.
- Viewed mail, swapped tabs.
- Grepped for `addPropertyList(...)`, looked for any remaining calls with a second argument.
- Also checked rSAAS for any calls, but we don't have anything there that uses tabs.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10628

Differential Revision: https://secure.phabricator.com/D16207
2016-06-30 18:50:09 -07:00
epriestley
6f275ba144 Render browse results with global result style
Summary:
Ref T11034. This seems a little more promising. Two problems at the moment:

  - This doesn't actually provide any useful information at all right now.
  - Many object types have no profile images.

Test Plan:
{F1695254}

{F1695255}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T11034

Differential Revision: https://secure.phabricator.com/D16155
2016-06-20 16:49:02 -07:00
epriestley
c3afddec9c Add icons to the new repository edit nav
Summary:
Ref T10748. These:

  - Look nice.
  - Hint at panel contents / effects.
  - Hint which panels have been customized.
  - Allow panels with issues or errors to be highlighted with an alert/attention icon.

Test Plan: {F1256156}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15836
2016-05-03 08:01:18 -07:00
Chad Little
f2a38f52d7 Update Drydock remaining pages to new UI
Summary: Updates Console and Operations page.

Test Plan: Pull up Console, pull up status page

Reviewers: epriestley

Reviewed By: epriestley

Subscribers: Korvin

Differential Revision: https://secure.phabricator.com/D15615
2016-04-05 07:25:43 -07:00
Chad Little
a939bbc4fa Update EditEngine for two column
Summary: Cleans up EditEngine, adds new layout to EditEngine and descendents

Test Plan: Test creating a new form, reordering, marking and unmarking defaults. View new forms.

Reviewers: epriestley

Reviewed By: epriestley

Subscribers: Korvin

Differential Revision: https://secure.phabricator.com/D15531
2016-03-28 09:18:55 -07:00
Chad Little
301ecdef18 Convert Drydock to two column layout
Summary: Updates Drydock to use two column + curtain layouts.

Test Plan: Tested what I could get to, need @epriestley to run this locally for edge cases.

Reviewers: epriestley

Reviewed By: epriestley

Subscribers: Korvin, epriestley

Differential Revision: https://secure.phabricator.com/D15467
2016-03-13 08:10:36 -07:00
epriestley
2a3c3b2b98 Provide bin/nuance import and ngram indexes for sources
Summary:
Ref T10537. More infrastructure:

  - Put a `bin/nuance` in place with `bin/nuance import`. This has no useful behavior yet.
  - Allow sources to be searched by substring. This supports `bin/nuance import --source whatever` so you don't have to dig up PHIDs.

Test Plan:
  - Applied migrations.
  - Ran `bin/nuance import --source ...` (no meaningful effect, but works fine).
  - Searched for sources by substring in the UI.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10537

Differential Revision: https://secure.phabricator.com/D15436
2016-03-08 10:30:24 -08:00
epriestley
2ddd78647b Improve "Land Revision" errors for issues related to staging areas
Summary: Ref T10093. Changes must be pushed to staging before they can be landed from the web.

Test Plan:
Changes must be pushed to staging before they can be landed from the web.

{F1161909}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10093

Differential Revision: https://secure.phabricator.com/D15427
2016-03-07 12:46:07 -08:00
epriestley
19eee427ad Improve Drydock errors for empty commits and missing changes
Summary: Ref T10093. Show better errors when a commit fails because it has already been merged and when a fetch fails because the ref isn't present in the remote.

Test Plan:
{F1160794}

{F1160795}

Reviewers: chad

Reviewed By: chad

Subscribers: michaeljs1990, yelirekim

Maniphest Tasks: T10093

Differential Revision: https://secure.phabricator.com/D15420
2016-03-07 05:29:25 -08:00
epriestley
2f14a59b1d Unprototype Drydock
Summary:
Ref T10246. Drydock still has a few outstanding issues, but it's mostly minor UI stuff and the documentation has reasonable caveats about it.

Broadly, I don't expect to make any major changes to Drydock in the future (i.e., all the fundamentals seem sound at this point) and it doesn't have any major technical debt or, like, obsolete APIs or anything.

Test Plan: Saw "Drydock" as not-a-prototype in Applications.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10246

Differential Revision: https://secure.phabricator.com/D15401
2016-03-04 17:23:07 -08:00
epriestley
e6158391d2 Make Drydock repository operations a little more modern and consistent
Summary: Ref T10457. Use modern controller and UI tech to build the list view and actions.

Test Plan:
  - Viewed operation list.
  - Viewed operation detail.
  - Checked menus on mobile.

{F1139757}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10457

Differential Revision: https://secure.phabricator.com/D15393
2016-03-03 15:22:15 -08:00
epriestley
fc0dc02bb9 Allow Drydock blueprints to be tagged and searched, and give types some little icons
Summary:
Ref T10457.

  - Let blueprints be tagged so you can search and annotate them a little more easily.
  - Give each blueprint type an optional icon to make things a little easier to parse visually.

Test Plan:
  - Tagged blueprints.
  - Searched by tags.
  - Looked at nice little icons.

{F1139712}

Reviewers: chad

Reviewed By: chad

Subscribers: yelirekim

Maniphest Tasks: T10457

Differential Revision: https://secure.phabricator.com/D15392
2016-03-03 15:21:58 -08:00
epriestley
1bdf988556 Convert DrydockBlueprints to EditEngine
Summary:
Ref T10457. Fixes T10024. This primarily just modernizes blueprints to use EditEngine.

This also fixes T10024, which was an issue with stored properties not being flagged correctly.

Also slightly improves typeaheads for blueprints (more information, disabled state).

Test Plan:
  - Created and edited various types of blueprints.
  - Set and removed limits.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10024, T10457

Differential Revision: https://secure.phabricator.com/D15390
2016-03-03 15:21:25 -08:00
epriestley
01379958fa Allow Drydock blueprints to be searched by name
Summary:
Ref T10457. The ngram indexing seems to be working well; extend it into Drydock.

Also clean up the list controller a little bit.

Test Plan:
  - Ran migrations.
  - Searched for blueprints by name.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10457

Differential Revision: https://secure.phabricator.com/D15389
2016-03-03 15:21:12 -08:00
Nick Zheng
6861af0cbb When the push phase of "Land Revision" fails, show the error in the UI
Summary: T10447

Test Plan: tested on my dev instance

Reviewers: #blessed_reviewers, epriestley

Reviewed By: #blessed_reviewers, epriestley

Subscribers: Korvin, yelirekim

Differential Revision: https://secure.phabricator.com/D15350
2016-02-26 09:38:39 -08:00
epriestley
7d4b323da2 Store Almanac "service types" instead of "service classes"
Summary:
Ref T10449. Currently, we store classes (like "AlmanacClusterRepositoryServiceType") in the database.

Instead, store types (like "cluster.repository").

This is a small change, but types are a little more flexible (they let us freely reanme classes), a little cleaner (fewer magic strings in the codebase), and a little better for API usage (they're more human readable).

Make this minor usability change now, before we unprototype.

Also make services searchable by type.

Also remove old Almanac API endpoints.

Test Plan:
  - Ran migration, verified all data migrated properly.
  - Created, edited, rebound, and changed properties of services.
  - Searched for services by service type.
  - Reviewed available Conduit methods.

Reviewers: chad

Reviewed By: chad

Subscribers: yelirekim

Maniphest Tasks: T10449

Differential Revision: https://secure.phabricator.com/D15346
2016-02-26 06:21:50 -08:00
epriestley
4c97d88aa4 Allow Almanac bindings to be disabled and unused interfaces to be removed
Summary:
Fixes T9762. Ref T10246.

**Disabling Bindings**: Previously, there was no formal way to disable bindings. The internal callers sometimes check some informal property on the binding, but this is a common need and deserves first-class support in the UI. Allow bindings to be disabled.

**Deleting Interfaces**: Previously, you could not delete interfaces. Now, you can delete unused interfaces.

Also some minor cleanup and slightly less mysterious documentation.

Test Plan: Disabled bindings and deleted interfaces.

Reviewers: chad

Reviewed By: chad

Subscribers: yelirekim

Maniphest Tasks: T9762, T10246

Differential Revision: https://secure.phabricator.com/D15345
2016-02-26 06:21:23 -08:00
Chad Little
fe5cd4ca2c Move FontIcon calls to Icon
Summary: Normalizes all `setFontIcon` calls to `setIcon`.

Test Plan: UIExamples, Almanac, Apps list, etc.

Reviewers: epriestley

Reviewed By: epriestley

Subscribers: Korvin, hach-que, yelirekim

Differential Revision: https://secure.phabricator.com/D15129
2016-01-28 08:48:45 -08:00
Chad Little
36158dbdc0 Convert all calls to 'IconFont' to just 'Icon'
Summary: Mostly for consistency, we're not using other forms of icons and this makes all classes that use an icon call it in the same way.

Test Plan: tested uiexamples, lots of other random pages.

Reviewers: epriestley

Reviewed By: epriestley

Subscribers: Korvin

Differential Revision: https://secure.phabricator.com/D15125
2016-01-27 20:59:27 -08:00
epriestley
367955f3fd Improve UX and messaging for certain errors when landing revisions
Summary:
Ref T9994.

  - Allow errors to be dismissed.
  - Tailor messaging for closed/abandoned revisions.
  - Reduce scare messaging on land dialog, since it's not really that scary anymore.

Test Plan:
  - Dismissed errors.
  - Hit new warnings.
  - Wasn't as scared when landing.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T9994

Differential Revision: https://secure.phabricator.com/D14886
2015-12-27 08:01:20 -08:00
epriestley
426d648681 In Drydock, don't reset current branch to point at unrelated commit
Summary:
Fixes T10037. When we're building commit `aabbccdd`, we currently do this to check it out:

  git reset --hard aabbccdd

However, this has an undesirable side effect of moving the current branch pointer to point at `aabbccdd`. The current branch pointer may be some totally different branch which `aabbccdd` is not part of, so this is confusing and misleading.

Instead, use `git reset --hard HEAD` to get the primary effect we want (destroying staged changes) and then `git checkout aabbccdd` to checkout the commit in a detached HEAD state.

Test Plan:
  - Ran a build (a commit-focused operation) successfully.
  - Verified working copy was pointed at a detached HEAD afterward:

```
builder@sbuild001:/var/drydock/workingcopy-167/repo/git-test-ii$ git status
HEAD detached at ffc7635
nothing to commit, working directory clean
```

  - Ran a land (a branch-foused operation) successfully.
  - Verified working copy was pointed at a branch afterward:

```
builder@sbuild001:/core/data/drydock/workingcopy-168/repo/git-test$ git status
On branch master
Your branch is up-to-date with 'origin/master'.

nothing to commit, working directory clean
```

Reviewers: chad

Reviewed By: chad

Subscribers: yelirekim

Maniphest Tasks: T10037

Differential Revision: https://secure.phabricator.com/D14850
2015-12-22 06:47:47 -08:00
epriestley
7168d8edd9 Make Drydock reclaim unused resources when it reaches a resource limit
Summary:
Fixes T9994. Currently, when Drydock can't allocate a new resource because some limit has been reached, it waits patiently for a resource to become available.

It is possible that no resource will ever become available. Particularly with "Working Copy" resources, the new lease may want a copy of `rB`, but the resource may already be maxed out on `rA`.

Right now, no process exists to automatically reclaim the unused `rA`.

When we encounter this situation, try to reclaim one of the other resources if it is just sitting there unused.

Specifically:

  - Add a "reclaim" command which means "release this resource //if// it is completely unused".
  - Add a `bin/drydock reclaim` to send this command to every active resource.
  - When we try to acquire a resource and can't, but only because of some kind of limit / utilization problem, try to release an unused resource to free up some room.

Test Plan:
  - Set "Working Copy" resource limit to 1.
  - Ran "Test Configuration" in `rA`, which worked.
  - Ran "Test Configuration" in `rB`, which hung forever.
  - Applied patch.
  - Ran "Test Configuration" in `rB`, saw it reclaim the `rA` resource, use the slot, then succeed.
  - Ran "Test Configuration" in `rA` again, saw it grab the slot back.
  - Ran `bin/drydock reclaim` and saw it reclaim a bunch of old orphaned resources.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T9994

Differential Revision: https://secure.phabricator.com/D14819
2015-12-18 11:55:51 -08:00
epriestley
e9af4f8970 Fix an issue where Drydock followup tasks would not queue if the main task failed
Summary:
Ref T9994. This fixes the first issue discussed on that task, which is that when a merge fails after "arc land", we would not clean up all the leases properly.

Specifically, when a merge fails, we use `queueTask()` to schedule a followup task. This followup destroys the lease and frees the underlying resource.

However, the default behavior of `queueTask()` is to //not queue tasks// if the parent task fails. This is a reasonable, safe behavior that was originally introduced in D8774, where it kept us from sending too much mail if a task did "send some mail" and then failed a little later on and got retried.

Since I think the default behavior is correct, I just special cased the behavior for Drydock to make it queue even on failure. These are the only types of followup tasks we currently want to queue on main task failure.

(It's possible that future Blueprints might want some kind of more specialized behavior, where some tasks queue only on success, but we can cross that bridge when we come to it.)

Test Plan:
  - See T9994#149878 for test case setup.
  - I ran that test case again with this patch, and saw the followup task queue properly in the `--trace` log, a correspoinding update task show up in `/daemon/`, and the lease get destroyed when I ran it a moment later.

{F1029915}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T9994

Differential Revision: https://secure.phabricator.com/D14818
2015-12-18 08:17:04 -08:00
epriestley
2868a69f65 Remove all setObjectURI() from ActionListViews
Summary:
Ref T10004. After D14804, we get this behavior by default and no longer need to set it explicitly.

(If some endpoint did eventually need to set it explicitly, it could just change what it passes to `setHref()`, but I believe we currently have no such endpoints and do not foresee ever having any.)

Test Plan:
  - As a logged out user, clicked various links in Differential, Maniphest, Files, etc., always got redirected to a sensible place after login.
  - Grepped for `setObjectURI()`, `getObjectURI()` (there are a few remaining callsites, but to a different method with the same name in Doorkeeper).

Reviewers: chad

Reviewed By: chad

Subscribers: hach-que

Maniphest Tasks: T10004

Differential Revision: https://secure.phabricator.com/D14805
2015-12-17 08:30:22 -08:00
Joshua Spence
495cb7a2e0 Mark PhabricatorPHIDType::getPHIDTypeApplicationClass() as abstract
Summary: Fixes T9625. As explained in a `TODO` comment, seems reasonable enough.

Test Plan: Unit tests.

Reviewers: #blessed_reviewers, epriestley

Reviewed By: #blessed_reviewers, epriestley

Subscribers: Korvin, hach-que

Maniphest Tasks: T9625

Differential Revision: https://secure.phabricator.com/D14068
2015-11-03 06:47:12 +11:00
epriestley
f48a833704 Fix an issue with incorrect authorization handling in Working Copy build steps
Summary:
Fixes T9669. Two issues:

  - We were using `repositoryPHIDs` instead of `blueprintPHIDs` for the list of allowed blueprints. Use the correct value.
  - We weren't enforcing `allowedBlueprintPHIDs` fully correctly. We //did// require an authorization, so the net effect was correct in nearly all cases, but we could have selected from too large a pool in the case where the application itself was doing the authorization (e.g., from the command line).

Test Plan: Ran a build through Drydock/Harbormaster locally.

Reviewers: chad, tycho.tatitscheff

Reviewed By: chad, tycho.tatitscheff

Subscribers: tycho.tatitscheff

Maniphest Tasks: T9669

Differential Revision: https://secure.phabricator.com/D14368
2015-10-30 16:02:35 +00:00
Chad Little
4b5de5135c Fix landing icon
Summary: This is //hilarious//.

Test Plan: Test icon on local install.

Reviewers: epriestley

Reviewed By: epriestley

Subscribers: Korvin

Differential Revision: https://secure.phabricator.com/D14351
2015-10-27 12:07:28 -07:00
epriestley
198bf1198d Fix "Accepted" status constant in landing
Summary: I didn't test the positive version of this -- the constant has value `2` but when we read it from the database it's `"2"` or whatever. Just do this for now and maybe someday we'll use strings.

Test Plan: will do production things

Reviewers: chad

Reviewed By: chad

Differential Revision: https://secure.phabricator.com/D14352
2015-10-27 12:06:48 -07:00
epriestley
1c7443f8f2 Make "Land Revision" button state consistent, prevent non-accepted lands
Summary:
Ref T182. Make the disabled state of the button more accurately reflect whether clicking it will work.

Don't allow "land" to proceed unless the revision is accepted.

Test Plan: Saw button in disabled state, clicked it, got "only accepted revisions" message.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14350
2015-10-27 18:51:59 +00:00
epriestley
a763f9510e Add some Drydock documentation plus "Test Configuration" for repository automation
Summary:
Ref T182. Ref T9252.

  - Adds a "Test" repository operation that just runs `git status` to see if things work.
  - Adds a button for it in Edit Repository.
  - Shows operation status on the operation detail view to make this workflow work a little better.
  - Adds a lot of words. Words words words words.

Test Plan:
  - Tested repository operation.
  - Read words.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182, T9252

Differential Revision: https://secure.phabricator.com/D14349
2015-10-27 18:04:02 +00:00
epriestley
cea633f698 Don't show error operations after a successful land operation
Summary:
Ref T182. When viewing a revision, if there are several error operations and then a success operation, we currently show the last error. This is misleading.

Instead, don't show anything if there's a success (this may require tuning eventually if you can land multiple times onto different branches or whatever, but should be reasonable for now).

Also make the table a little nicer, particularly for merge failure output.

Test Plan: {F910385}

Reviewers: chad, Mnkras

Reviewed By: Mnkras

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14348
2015-10-26 21:27:04 +00:00
epriestley
bbf4ce79e3 Provide a username and email when running git merge --squash
Summary:
Ref T182. This command should never actually generate a commit because `--squash` prevents that, but `git` seems to sometimes hit a check for username/email configuration (maybe when merging a non-fastforward?).

Give it some dummy values to placate it. This command shouldn't commit anything so these values should never actually be used.

Test Plan: Landed rGITTESTd8c8643cb02bbe60048c6c206afc2940c760a77e.

Reviewers: chad, Mnkras

Reviewed By: Mnkras

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14347
2015-10-26 21:12:16 +00:00
epriestley
5a35dd233b Don't use --ff-only inside "Land Revision"
Summary:
Ref T182. I lifted this logic out of `arc`, but the context is a little different there, and this option is too strict in "Land Revision".

Specifically, it prevents `git` from merging unless the merge is //strictly// a fast-foward, even with `--squash`. That means revisions can't merge unless they're rebased on the current `master`, even if they have no conflicts.

(This whole process will probably need additional refinement, but the behavior without this flag is more reasonable overall than the behavior with it for now.)

Test Plan: Will land stuff in production~~

Reviewers: chad, Mnkras

Reviewed By: Mnkras

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14346
2015-10-26 20:26:56 +00:00
epriestley
0b24a6e200 Make "Land Revision" show merge conflicts more clearly
Summary:
Ref T182. We just show "an error happened" right now. Improve this behavior.

This error handling chain is a bit ad-hoc for now but we can formalize it as we hit other cases.

Test Plan:
{F910247}

{F910248}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14343
2015-10-26 20:11:21 +00:00
epriestley
2326d5f8d0 Show lease on Repository Operation detail view and awaken on failures
Summary:
Ref T182. Couple of minor improvements here:

  - Show the Drydock lease when viewing a Repository Operation detail screen. This just makes it easier to jump around between relevant objects.
  - When tasks are waiting for a lease, awaken them when it breaks or is released, not just when it is acquired. This makes the queue move forward faster when errors occur.

Test Plan:
  - Viewed a repository operation and saw a link to the lease.
  - Did a bad land (intentional merge problem) and got an error in about ~3 seconds instead of ~17.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14341
2015-10-26 20:00:49 +00:00
epriestley
a0fba642b3 Show the oldest non-failing revision land operation, or the newest failure
Summary:
Ref T182.

  - We just show the oldest operation right now, but we usually care about the oldest non-failure.
  - Only query for actual land operations when rendering the revision operations dialog (maybe eventually we'll show more stuff?).
  - For now, prevent multiple lands / repeated lands or queueing up lands while other lands are happening.

Test Plan: Landed a revision. Tried to land it more / again.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14338
2015-10-26 19:58:37 +00:00
epriestley
9c39493796 Make WorkingCopyBlueprint responsible for performing merges
Summary:
Ref T182. Currently, the "RepositoryLand" operation is responsible for performing merges when landing a revision.

However, we'd like to be able to perform these merges in a larger set of cases in the future. For example:

  - After Releeph is revamped, when someone says "I want to merge bug fix X into stable branch Y", it would probably be nice to make that a Buildable and let tests run against it without requring that it actually be pushed anywhere.
  - Same deal if we want a merge-from-Diffusion or cherry-pick-from-Diffusion operation.
  - Similar deal if we want a "random web UI edits from Diffusion".

Move the merging part into WorkingCopy so more applications can share/use it in the future.

A big chunk of this is me making stuff up for now (the ol' undocumented dictionary full of arbitrary magic keys), but I anticipate formalizing it as we move along.

Test Plan: Pushed rGITTEST0d58eef3ce0fa5a10732d2efefc56aec126bc219 up from my local install via "Land Revision".

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14337
2015-10-26 12:40:16 -07:00
epriestley
c059149eb9 Remove Drydock host resource limits and give working copies simple limits
Summary:
Ref T9252. Right now, we have very strict limits on Drydock: one lease per host, and one working copy per working copy blueprint.

These are silly and getting in the way of using "Land Revision" more widely, since we need at least one working copy for each landable repository.

For now, just remove the host limit and put a simple limit on working copies. This might need to be fancier some day (e.g., limit working copies per-host) but it is generally reasonable for the use cases of today.

Also add a `--background` flag to make testing a little easier.

(Limits are also less important nowadays than they were in the past, because pools expand slowly now and we seem to have stamped out all the "runaway train" bugs where allocators go crazy and allocate a million things.)

Test Plan:
  - With a limit of 5, ran 10 concurrent builds and saw them finish after allocating 5 total resources.
  - Removed limit, raised taskmaster concurrency to 128, ran thousands of builds in blocks of 128 or 256.
    - Saw Drydock gradually expand the pool, allocating a few more working copies at first and a lot of working copies later.
    - Got ~256 builds in ~140 seconds, which isn't a breakneck pace or anything but isn't too bad.
    - This stuff seems to be mostly bottlenecked on `sbuild` throttling inbound SSH connections. I haven't tweaked it.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T9252

Differential Revision: https://secure.phabricator.com/D14334
2015-10-26 12:39:47 -07:00
epriestley
3f193cb9e0 Give Harbormaster build steps a "View" page
Summary:
Fixes T9519. Right now, build steps go straight from the build to the edit screen.

This means that there's no way to see their edit history or review details without edit permission. In particular, this makes it a bit harder to catch the Drydock Blueprint authorization warnings from T9519.

  - Add a standard view screen.
  - Add a little warning callout to blueprint authorizations.

This also does a bit of a touchup on the weird dropshadow element from T9586. Maybe not totally design-approved now but it's less ugly, at least.

Test Plan:
{F906695}

{F906696}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T9519

Differential Revision: https://secure.phabricator.com/D14330
2015-10-26 12:38:32 -07:00
epriestley
5b619862cb Show a more reasonable status element for pull requests
Summary:
Ref T182. Replace the total mess we had before with a sort-of-reasonable element.

This automatically updates using "javascript".

Test Plan:
{F901983}

{F901984}

Used "Land Revision", saw the land status go from "Waiting" -> "Working" -> "Landed" without having to mash reload over and over again.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14314
2015-10-21 11:28:26 -07:00
epriestley
92a626fc1c Add a basic list view for repository operations
Summary: Ref T182. Nothing fancy, just make these slightly easier to work with.

Test Plan: {F884754}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14295
2015-10-16 18:47:05 -07:00
epriestley
f3f3d95702 When landing revisions via repository automation, use better metadata
Summary: Ref T182. Make a reasonable attempt to get the commit message, author, and committer data correct.

Test Plan: BEHOLD: rGITTEST810b7f17cd0c909256a45d29a5062fcf417d0489

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14280
2015-10-14 10:50:53 -07:00
epriestley
ac7edf54af Fix bad counting in SQL when enforcing Drydock allocator soft limits
Summary:
Ref T9252. This fixes a bug from D14236. D14272 discusses the observable effects of the bug, primarily that the window for racing is widened from ~a few milliseconds to several minutes under our configuration.

This SQL query is missing a `GROUP BY` clause, so all of the resources get counted as having the same status (specifically, the alphabetically earliest status any resource had, I think). For test cases this often gets the right result since the number of resources may be small and they may all have the same status, but in production this isn't true. In particular, the allocator would sometimes see "35 destroyed resources" (or whatever), when the real counts were "32 destroyed resources + 3 pending resources".

Since this allocator behavior is soft/advisory this didn't cause any actual problems, per se (we do expect races here occasionally), it just made the race very very easy to hit. For example, Drydock in production currently has three pending working copy resources. Although we do expect this to be //possible//, getting 4 resources when the configured limit is 1 should be hard (not lightning strike / cosmic radiaion hard, but "happens once a year" hard).

Also exclude destroyed resources since we never care about them.

Test Plan:
Followed the plan from D14272 and restarted two Harbormaster workers at the same time.

After this patch was applied, they no longer created two different resources (we expect it to be possible for this to happen, just very hard).

We should still be able to force this race by putting something like `sleep(10)` right before the query, then `sleep(10)` right after it. That would prevent the allocators from seeing one another (so they would both think there were no other resources) and push us down the pathway where we exceed the soft limit.

Reviewers: chad, hach-que

Reviewed By: hach-que

Maniphest Tasks: T9252

Differential Revision: https://secure.phabricator.com/D14274
2015-10-14 06:18:10 -07:00
epriestley
083a321dad Fix an issue where newly created Drydock resources could be improperly acquired
Summary:
Ref T9252. This is mostly a fix for an edge case from D14236. Here's the setup:

  - There are no resources.
  - A request for a new resource arrives.
  - We build a new resource.

Now, if we were leasing an existing resource, we'd call `canAcquireLeaseOnResource()` before acquiring a lease on the new resource.

However, for new resources we don't do that: we just acquire a lease immediately. This is wrong, because we now allow and expect some resources to be unleasable when created.

In a more complex workflow, this can also produce the wrong result and leave the lease acquired sub-optimally (and, today, deadlocked).

Make the "can we acquire?" pathway consistent for new and existing resources, so we always do the same set of checks.

Test Plan:
  - Started daemons.
  - Deleted all working copy resources.
  - Ran two working-copy-using build plans at the same time.
  - Before this change, one would often [1] acquire a lease on a pending resource which never allocated, then deadlock.
  - After this change, the same thing happens except that the lease remains pending and the work completes.

[1] Although the race this implies is allowed (resource pool limits are soft/advisory, and it is expected that we may occasionally run over them), it's MUCH easier to hit right now than I would expect it to be, so I think there's probably at least one more small bug here somewhere. I'll see if I can root it out after this change.

Reviewers: chad, hach-que

Reviewed By: hach-que

Maniphest Tasks: T9252

Differential Revision: https://secure.phabricator.com/D14272
2015-10-14 06:16:21 -07:00
epriestley
43bee4562c If the stars align, make "Land Revision" kind of work
Summary:
Ref T182. If 35 other things are configured completely correctly, make it remotely possible that this button may do something approximating the thing that the user wanted.

This primarily fleshes out the idea that "operations" (like landing, merging or cherry-picking) can have some beahavior, and when we run an operation we do whatever that behavior is instead of just running `git show`.

Broadly, this isn't too terrible because Drydock seems like it actually works properly for the most part (???!?!).

Test Plan: {F876431}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T182

Differential Revision: https://secure.phabricator.com/D14270
2015-10-13 15:46:30 -07:00