Summary:
See <rPc40bc0c8bf75#4050>. Repro steps:
- Scroll partway down the page.
- Click and drag the scroll handle.
Prior to this diff, the handle incorrectly jumps back to the top of the page. This is because we didn't store the handle's original position. (In testing, I always dragged from near the top of the page, and I don't normally drag scrollbars, so I didn't notice this.)
Test Plan: Clicking and dragging a partially scrolled handle now works correctly.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11483
Summary: This seems to improve behavior on iOS, good call.
Test Plan: Hard to be totally sure since my local install isn't set up with a real phone, but behavior seems better on iOS simulator.
Reviewers: chad
Reviewed By: chad
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11481
Summary:
See discussion on rPc40bc0c8bf75. Fixes a couple of glitchy things:
- Things were generally not nice on iOS.
- On OSX, with no mouse, the OS scrollbar and our fake scrollbar would both draw.
- Bar z-index was not set quite correctly.
Specifically, check if we need these bars. If we don't, just exit immediately and use the OS bars.
Test Plan:
- Tested Safari, Firefox, Chrome with and without a mouse.
- Tested iOS Simualtor.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11480
Summary:
Ref T2783. I think this served two purposes:
- Improving performance in cases where we "know" a repository is local.
- Preventing loops.
It is now obsolete:
- After D11476, refs can almost always resolve on a fast path.
- As T2783 moves forward, we can usually no longer know when a repository is local without actually looking it up -- almost everything is allowed to run anywhere.
- The cluster behavior in D11475 now prevents loops.
Test Plan: `grep`, browsed around. This didn't really do much of anything anymore.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2783
Differential Revision: https://secure.phabricator.com/D11477
Summary:
Ref T2783. With service-oriented calls, we take a larger performacne hit than necessary resolving refs.
Instead of resolving refs over the wire, try to resolve them from the database first. This can resolve almost all refs (commit hashes, branch and tag names).
This can't resolve weird refs like `master~50`, and obviously can't resolve invalid refs. In those cases we'll go back to the old logic, call `diffusion.resolverefs`, and end up with the right result.
Test Plan:
- Browsed repositories in Diffusion.
- Verified that service repositories no longer make unnecessary `diffusion.resolverefs` calls for common refs (branch names, commit hashes).
- Resolved refs like `master~50`, saw call to underlying VCS and correct result.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T2783
Differential Revision: https://secure.phabricator.com/D11476
Summary:
Fixes T7020. When an external user makes a Conduit request to Diffusion but the repository isn't hosted locally, we need to proxy it.
This also adds a guard layer to prevent requests from getting infinitely proxied inside the cluster.
In "trivial" configurations (where the repository is a service repository, but the service is on the local device) I'm making us always proxy anyway. This basically makes it reasonable to test this stuff (otherwise you'd have to set up two different installs) and this configuration doesn't make much sense in real life (if you're using multiple machines, making one a dedicating daemons+repo box is almost certainly the most reasonable configuration, even for a cluster size of 2).
Test Plan:
- With a service-hosted repository, made Diffusion conduit calls and browsed the UI. Verified requests got proxied once, then resovled.
- With a non-service repository, made Diffusion conduit calls and browsed UI. Verified requests were handled in-process immediately.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7020
Differential Revision: https://secure.phabricator.com/D11475
Summary: Ref T7020. I need this elsewhere, and it's relatively internal anyway.
Test Plan: Browsed around my local, cluster-configured install and saw everything working fine.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7020
Differential Revision: https://secure.phabricator.com/D11474
Summary:
Ref T7014. With a mouse plugged in, multi-panel UIs are pretty hideous on OSX. This is somewhat offputting for me in Conpherence, and really jumps out at me with the new column mocks in T7014.
Sites like Twitch and Facebook approach this by emulating the touchpad scrollbar to achieve a more aesthetic UI. Use a similar approach.
This:
- Replaces the main scrollbar with a prettier fake one.
- This prepares the standard page frame for a persistent chat column.
Test Plan:
- Seems to work properly on OSX, Chrome and Firefox. Haven't tested on IE; my Windows setup is pretty iffy at the moment.
- Tried Conpherence.
- Tried Workboards.
Reviewers: btrahan, chad
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T7014
Differential Revision: https://secure.phabricator.com/D11472
Summary: Fixes T7021. When I moved around all the timeline stuff I guess I didn't find this "corner" case, which is wildly common in the post-commit review workflow that we don't use.
Test Plan: pre-patch I could reproduce the issue and post patch I could not. The reproduction case is to have a commit with inline comments and then enough subsequent comments to have a "show older" UI. clicking "show older" now works!
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T7021
Differential Revision: https://secure.phabricator.com/D11479
Summary: Fixes T7011. Recent refactoring here caused us to begin ignoring URI parameters like `commit`. Most controllers take parameters as a `dblob`, which was still parsed properly.
Test Plan:
- Editing different commits actually edits the desired commits.
- Browsed around some `dblob` pages and verified they still work properly.
Reviewers: btrahan, chad
Reviewed By: chad
Subscribers: epriestley
Maniphest Tasks: T7011
Differential Revision: https://secure.phabricator.com/D11473
Summary:
One advantage I wanted to get out of T1191 is automated rebuilds of `quickstart.sql`. If they don't actually work, I'd like to know sooner rather than later. We haven't rebuilt in a couple months, so give it a shot.
Ran into two issues:
- Some very old patches specify overlong keys which don't work if your default charsets are utf8mb4. Shorten these. No real users have applied these in a very long time.
- Some gymnastics around `corpus` for the new Conpherence search index.
Test Plan:
- Ran `arc unit --everything`, got clean results.
- Cost to do a storage upgrade on an empty namespace dropped from ~4s to ~3s.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11454
Summary: I got these wrong and the test didn't trigger for some reason that I haven't looked into.
Test Plan: `arc unit --everything`
Reviewers: hach-que, btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11453
Summary:
Ref T5833. In some cases, we need to know if an Almanac device is the localhost or not, so we can either handle or forward the request.
To accomplish this, write a device ID when running `bin/almanac register`.
Using `--allow-key-reuse` and `--identify-as`, multiple devices are permitted to //authenticate// as one device but //identify// as different devices. In the Phacility cluster, this allows all the `repoXXX` machines to have one keypair (making key management much easier) but still work as separate devices. This is an advanced feature; normal installs with 1-3 hosts would just generate a key + device per host and identify/authenticate as the same device.
Test Plan: Ran commands with lots of flags like `PHACILITY_INSTANCE=local sudo -E ./bin/almanac register --device daemon.phacility.net --private-key ~/dev/core/conf/keys/daemon.key --force --allow-key-reuse --identify-as local001.phacility.net`. Got a good result from `AlmanacKeys::getDeviceID()` afterward.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T5833
Differential Revision: https://secure.phabricator.com/D11452
Summary: Fixes T6890. This doesn't feel like a perfect solution, but I can't think of any cases in which this will produce the wrong result either.
Test Plan: Ran `./bin/diviner generate` and checked the generated documentation for `PhabricatorCommonPasswords::loadWordlist()`. The return type was corrected shown as `map<string, bool>`.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T6890
Differential Revision: https://secure.phabricator.com/D11469
Summary: As suggested in T6950, add the method description to the response from `conduit.query`.
Test Plan: Called `echo '{}' | arc call-conduit conduit.query` and verified that the response contained the method description.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11467
Summary: Fixes T6950. Adds the return type of Conduit API methods to the `conduit.query` call.
Test Plan: Called `echo '{}' | arc call-conduit conduit.query` and verified that the return types were present in the response.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Maniphest Tasks: T6950
Differential Revision: https://secure.phabricator.com/D11466
Summary:
Fixes T6858. We shouldn't create mentions for dependent diffs.
NOTE: This won't fix the issue for existing revisions (which have the mentions edge), but I think that this is harmless.
Test Plan: Added `Depends on Dxxx` to a differential summary. Saw a `josh added a dependent revision` transaction, but no explicit mention.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T6858
Differential Revision: https://secure.phabricator.com/D11460
Summary: Add the logger as soon as possible so that the log file will contain errors if the `ws` module cannot be loaded.
Test Plan: Ran `./bin/aphlict debug` without having the `ws` module installed. Saw errors in the logs.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11457
Summary:
We have to do some garbage nonsense to write database backups right now, see T6996.
When storage isn't initialized, we previously ended up with this message gzipped in a file and an empty error. Make the behavior slightly more tolerable.
Test Plan: Saw a meaningful error after trying to back up an uninitialized database.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11449
Summary: Ref T5833. This was using the wrong constant, so we weren't validating property.
Test Plan: Tried to create a nameless network and correctly got an error.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T5833
Differential Revision: https://secure.phabricator.com/D11447
Summary: Fixes T6989. Basically return a nice dialogue like we do for "NoEffect" transactions. This is a little prettier than the other dialogue was. Also, stop adding TYPE_EDGE as a transaction type as we end up having it 2x, which then makes the error get validated 2x.
Test Plan: tried to add myself as a reviewer and got a nice error message.
Reviewers: chad, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T6989
Differential Revision: https://secure.phabricator.com/D11448
Summary:
Ref T6881. I tried to cheat here by not implementing this, but we need it for destroying triggers directly with `bin/remove destroy`, since that needs to load them by PHID.
So, cheat slightly less. Implement PolicyAware but not CursorPagedPolicyAware.
Test Plan:
- Used `bin/remove destroy` to destroy a trigger by PHID.
- Browsed daemon console.
- Ran trigger daemon.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11445
Summary:
Ref T6881. This makes it easier to fire a trigger and make sure it works properly. You can use the `--now` flag to travel through time, and test scheduling conditions with `--last` and `--next`. It will tell you when the trigger would reschedule.
Better than waiting 24 hours to see if things work.
Test Plan: Fired some backups, got useful output which made me think my code probably works correctly.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11438
Summary:
This option prohibits overwriting prototypes of native objects such as `Array`, `Date` and so on.
```lang=js
// jshint freeze:true
Array.prototype.count = function (value) { return 4; };
// -> Warning: Extending prototype of native object: 'Array'.
```
Test Plan: Linted existing JavaScript files, found no violations.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11439
Summary: Ref T6881. This is useful to show a "Next backup: 2:30 AM" sort of thing without requring callers to know how triggers work internally.
Test Plan: Showed that kind of thing in Instances.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11437
Summary:
Ref T6881. By design, the EXECUTION order only selects tasks which have been scheduled (since it performs a JOIN). This is inconsistent with other queries and problematic for withID/withPHID queries which may want to select an unscheduled task.
Switch to standard ID ordering by default.
Test Plan:
- Instances console now finds unscheduled triggers.
- Verified that all existing queries specify an explicit order.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11436
Summary: Ref T6881. When stuff with triggers is destroyed, it should destroy the triggers.
Test Plan: Will test in Instances.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11435
Summary: Ref T6881. Add a standard "just queue a task" trigger action; I expect almost all application code to use this.
Test Plan: Will test in Instances.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11429
Summary: Ref T6881. I just want to show trigger info in the instance management console.
Test Plan: Will test in Instances.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11428
Summary: Ref T6881. Before implementing subscriptions, I'm going to vet triggers by using them to do backups. Each instance will get a daily trigger for backups, and that should give us a smaller-scale test to catch issues and limitations, with more opportunities for something to go wrong since it fires more often.
Test Plan: Added unit tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11427
Summary: Ref T5952, T3404. This lays the basic plumbing for how this will work, all the way to deploying on Maniphest. Aside from what is mentioned on T5952, I think page(s) on editing application emails could use a little more helpful text about what's going on, similar to how the config page that's getting deprecated works.
Test Plan: ran migration and noted my create email address migrated successfully. used bin/mail to make a task. added another email and used bin/mail to make a task. deleted an email. edited an email. invoked various error states and they all looked good.
Reviewers: epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T3404, T5952
Differential Revision: https://secure.phabricator.com/D11418
Summary: This variable is used before it is defined.
Test Plan: `arc lint`
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11434
Summary: Enable strict mode for Javelin when running in NodeJS.
Test Plan: Made sure Aphlict still worked.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11432
Summary:
In particular, this changes the behavior of NodeJS in the following ways:
- Any attempt to get or modify the global object will result in an error.
- `null` values of `this` will no longer be evaluated to the global object and primitive values of this will not be converted to wrapper objects.
- Writing or deleting properties which have there writeable or configurable attributes set to false will now throw an error instead of failing silently.
- Adding a property to an object whose extensible attribute is false will also throw an error now.
- A functions arguments are not writeable so attempting to change them will now throw an error `arguments = [...]`.
- `with(){}` statements are gone.
- Use of `eval` is effectively banned.
- `eval` and `arguments` are not allowed as variable or function identifiers in any scope.
- The identifiers `implements`, `interface`, `let`, `package`, `private`, `protected`, `public`, `static` and `yield` are all now reserved for future use (roll on ES6).
Test Plan: Verified that Aphlict was still functional.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11430
Summary: Fixes T6963. Long term will likely make this more like other document views, but not worth the time right now since this is only location.
Test Plan: Review Phriction document at desktop and mobile breakpoints. Click menu and see menu.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Maniphest Tasks: T6963
Differential Revision: https://secure.phabricator.com/D11420
Summary:
This was broken in D11383. Basically, I had the `ws` module installed globally whilst testing, but the changes made do not work if the `ws` module is installed locally (i.e. in the `./support/aphlict/server/node_modules` directory). After poking around, it seems that this is due to the sandboxing that is done by `JX.require`.
A quick fix is to just //not// use `JX.require`, although you may have a better idea?
The error that is occurring is as follows:
```
<<< UNCAUGHT EXCEPTION! >>>
Error: Cannot find module 'ws'
at Function.Module._resolveFilename (module.js:338:15)
at Function.Module._load (module.js:280:25)
at Module.require (module.js:364:17)
at require (module.js:380:17)
at extra.require (/usr/src/phabricator/webroot/rsrc/externals/javelin/core/init_node.js:48:16)
at /usr/src/phabricator/support/aphlict/server/lib/AphlictClientServer.js:10:17
at Script.(anonymous function) [as runInNewContext] (vm.js:41:22)
at Object.JX.require (/usr/src/phabricator/webroot/rsrc/externals/javelin/core/init_node.js:58:6)
at Object.<anonymous> (/usr/src/phabricator/support/aphlict/server/aphlict_server.js:102:4)
at Module._compile (module.js:456:26)
>>> Server exited!
```
Test Plan: Now able to start the Aphlict server.
Reviewers: joshuaspence
Reviewed By: joshuaspence
Subscribers: Korvin, epriestley
Maniphest Tasks: T6987
Differential Revision: https://secure.phabricator.com/D11425
Summary: By calling `process.exit(1)` we are forcing the Node.js process to exit prematurely. Specifically, the process is terminated before the error is written to the log file. This can be verified by inspecting the value of `debug._logs[0]._writableState.writing` immediately before the process is terminated.
Test Plan: Ran `./bin/aphlict debug` without the `ws` module being installed. Verified that `<<< UNCAUGHT EXCEPTION! >>>` was echoed to the console as well as the log file. Also verified that the exit status was non-zero.
Reviewers: #blessed_reviewers, epriestley
Reviewed By: #blessed_reviewers, epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11426
Summary:
Taking a pass at revamping the edit pages in Projects. Specifically:
- Remove EditMainController
- Move actions from EditMain to Profile
- Move properties from EditMain to Profile
- Move timeline from EditMain to Profile
- Move Open Tasks from Profile to sidenavicon
- Add custom icons and colors to timeline
Feel free to bang on this a bit and give feedback, feels generally correct to me.
Test Plan: Edit everything I could on various projects. Check links, timelines, actions.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11421
Summary: Tidy the Aphlict server by splitting the functionality into two main modules, `AphlictClientServer` and `AphlictAdminServer. There is still further tidying that could be done here, but I feel that this puts us in a much better place.
Test Plan: Sent notifications via `/notification/status/`.
Reviewers: epriestley, #blessed_reviewers
Reviewed By: epriestley, #blessed_reviewers
Subscribers: Korvin, epriestley
Differential Revision: https://secure.phabricator.com/D11383
Summary:
Ref T6881. Hopefully, this is the hard part.
This adds a new daemon (the "trigger" daemon) which processes triggers, schedules them, and then executes them at the scheduled time. The design is a little complicated, but has these goals:
- High resistance to race conditions: only the application writes to the trigger table; only the daemon writes to the event table. We won't lose events if someone saves a meeting at the same time as we're sending a reminder out for it.
- Execution guarantees: scheduled events are guaranteed to execute exactly once.
- Support for arbitrarily large queues: the daemon will make progress even if there are millions of triggers in queue. The cost to update the queue is proportional to the number of changes in it; the cost to process the queue is proportional to the number of events to execute.
- Relatively good observability: you can monitor the state of the trigger queue reasonably well from the web UI.
- Modular Infrastructure: this is a very low-level construct that Calendar, Phortune, etc., should be able to build on top of.
It doesn't have this stuff yet:
- Not very robust to bad actions: a misbehaving trigger can stop the queue fairly easily. This is OK for now since we aren't planning to make it part of any other applications for a while. We do still get execute-exaclty-once, but it might not happen for a long time (until someone goes and fixes the queue), when we could theoretically continue executing other events.
- Doesn't start automatically: normal users don't need to run this thing yet so I'm not starting it by default.
- Not super well tested: I've vetted the basics but haven't run real workloads through this yet.
- No sophisticated tooling: I added some basic stuff but it's missing some pieces we'll have to build sooner or later, e.g. `bin/trigger cancel` or whatever.
- Intentionally not realtime: This design puts execution guarantees far above realtime concerns, and will not give you precise event execution at 1-second resolution. I think this is the correct goal to pursue architecturally, and certainly correct for subscriptions and meeting reminders. Events which execute after they have become irrelevant can simply decline to do anything (like a meeting reminder which executes after the meeting is over).
In general, the expectation for applications is:
- When creating an object (like a calendar event) that needs to trigger a scheduled action, write a trigger (and save the PHID if you plan to update it later).
- The daemon will process the event and schedule the action efficiently, in a race-free way.
- If you want to move the action, update the trigger and the daemon will take care of it.
- Your action will eventually dump a task into the task queue, and the task daemons will actually perform it.
Test Plan:
Using a test script like this:
```
<?php
require_once 'scripts/__init_script__.php';
$trigger = id(new PhabricatorWorkerTrigger())
->setAction(
new PhabricatorLogTriggerAction(
array(
'message' => 'test',
)))
->setClock(
new PhabricatorMetronomicTriggerClock(
array(
'period' => 33,
)))
->save();
var_dump($trigger);
```
...I queued triggers and ran the daemon:
- Verified triggers fire;
- verified triggers reschedule;
- verified trigger events show up in the web UI;
- tried different periods;
- added some triggers while the daemon was running;
- examined `phd debug` output for anything suspicious.
It seems to work in trivial use case, at least.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11419
Summary:
Ref T6881. This will probably make more sense in a couple of diffs, but this is a class that implements scheduling/recurrence rules. Two rules are provided:
- Trigger an event at a specific time (e.g., a meeting reminder notification).
- Trigger an event on the Nth day of every month (e.g., a subscription bill).
At some point, we'll presumably add a rule for T2896 (maybe using the "RRULE" spec) so you can do stuff like "the second to last thursday of every month", etc., but we don't need that for now.
(The "Nth day of every month, or move it back if no such day exists" rule doesn't seem to be expressible with the "RRULE" format, so implementing that wouldn't give us a superset of this. I think this rule is correct and desirable for this purpose, though.)
Test Plan: Added and executed unit tests.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Maniphest Tasks: T6881
Differential Revision: https://secure.phabricator.com/D11403
Summary:
This is unusual, but if `getWorkerInstance()` throws we end up with an undefined `$worker` when recovering from the exception.
Instead, handle this case slightly more gracefully.
The easiest way to hit this is to schedule a task for a worker that doesn't exist (or remove an existing worker, which is what I did to hit it).
Test Plan: Saw a more graceful error recovery; ran some normal successful tasks out of the queue.
Reviewers: btrahan
Reviewed By: btrahan
Subscribers: epriestley
Differential Revision: https://secure.phabricator.com/D11413
