Summary:
See discussion in D4204. Facebook currently has a 314MB remarkup cache with a 55MB index, which is slow to access. Under the theory that this is an index size/quality problem (the current index is on a potentially-384-byte field, with many keys sharing prefixes), provide a more general index with fancy new features:
- It implements PhutilKeyValueCache, so it can be a component in cache stacks and supports TTL.
- It has a 12-byte hash-based key.
- It automatically compresses large blocks of data (most of what we store is highly-compressible HTML).
Test Plan:
- Basics:
- Loaded /paste/, saw caches generate and save.
- Reloaded /paste/, saw the page hit cache.
- GC:
- Ran GC daemon, saw nothing.
- Set maximum lifetime to 1 second, ran GC daemon, saw it collect the entire cache.
- Deflate:
- Selected row formats from the database, saw a mixture of 'raw' and 'deflate' storage.
- Used profiler to verify that 'deflate' is fast (12 calls @ 220us on my paste list).
- Ran unit tests
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D4259
Summary:
Split Pholio's transaction implementation into generic and application-specific parts. Moves us toward generic transactions, with support for:
- Editing and deleting comments.
- Setting visibility of individual comments (I'm not a fan of this feature but we'll see).
I want to move everything to a more generic piece of infrastructure but there's very little they can share right now so adding transactions to, e.g., Paste or Macros (T2157) means massive amounts of similar code.
Tons of work left to do here, but I think it basically works. Here's a screenshot:
{F26820}
Test Plan: Made transactions in Pholio.
Reviewers: btrahan, vrana, chad
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2104
Differential Revision: https://secure.phabricator.com/D4136
Summary:
I'm not going to land this until it's a bit more fleshed out since it would just confuse users, but this is probably more reviewable as a few diffs adding a couple features than one ULTRA-diff adding everything. Implement application basics for Pholio. This does more or less nothing, but adds storage, subscribe, flag, markup, indexing, query basics, PHIDs, handle loads, a couple of realy really basic controllers, etc.
Basic hierarchy is:
- **Moleskine**: Top-level object like a Differential Revision, like "Ponder Feed Ideas".
- **Image**: Each Moleskine has one or more images, like the unexpanded / expanded / mobile / empty states of feed.
- **Transaction**: Comment or edit, like Maniphest. I generally want to move most apps to a transaction model so we can log edits.
- **PixelComment**: Equivalent of an inline comment.
Test Plan: Created a fake object and viewed it.
Reviewers: btrahan, chad
Reviewed By: btrahan
CC: aran, davidreuss
Maniphest Tasks: T2097
Differential Revision: https://secure.phabricator.com/D3817
Summary:
People hit three issues with D3914:
- As per T2059, we applied a schema change from a `.php` patch, which currently does not work if you use a different user to make schema changes than for normal use.
- Since the change in question is idempotent, just move it to a `.sql` patch. We'll follow up in T2059 and fix it properly.
- Rogue daemons at several installs used old code (expecting autoincrement) to insert into the new table (no autoincrement), thereby creating tasks with ID 0.
- Rename the table so they'll fail.
- This also makes the code a little more consistent.
- Some installs now have tasks with ID 0.
- Use checks against null rather than against 0 so we can process these tasks.
The major issues this fixes are the schema upgrade failure in T2059, and the infinite loops in T2072 and elsewhere.
This isn't really a fully statisfactory fix. I'll discuss some next steps in T2072.
Test Plan: Created new tasks via MetaMTA/Differential. Ran tasks with `phd debug taskmaster`. Inserted a task 0 and verified it ran and archived correctly.
Reviewers: btrahan, vrana, nh
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2072, T2059
Differential Revision: https://secure.phabricator.com/D3973
Summary: This saves lint errors to the path change of current commit. It requires pushed revision. It doesn't save difference from previous commit mentioned in T2038#comment-4 - I don't plan doing it after all, everything would be much more complicated and the amount of data saved with this approach isn't that bad.
Test Plan: Applied patch, ran script, verified DB.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2038
Differential Revision: https://secure.phabricator.com/D3899
Summary: don't need it now that uploading files is so easy. Plus it made for some buggy jonx if / when there were bad image links coupled with caching. In theory this is a lot less pretty though if folks linked to a bunch of files served elsewhere using images.
Test Plan: http://does-not-exist.com/imaginary.jpg rendered as a link!
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T2000
Differential Revision: https://secure.phabricator.com/D3908
Summary: See D3912 for discussion. InnoDB may reuse autoincrement IDs after restart; provide a way to avoid it.
Test Plan: Unit tests. Scheduled and executed tasks through `drydock lease --type host` and `phd debug taskmaster`.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Differential Revision: https://secure.phabricator.com/D3914
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:
Tightens up a bunch of stuff:
- In `drydock lease`, pull and print logs so the user can see what's happening.
- Remove `DrydockAllocator`, which was a dumb class that did nothing. Move the tiny amount of logic it held directly to `DrydockLease`.
- Move `resourceType` from worker task metadata directly to `DrydockLease`. Other things (like the web UI) can be more informative with this information available.
- Pass leases to `allocateResource()`. We always allocate in response to a lease activation request, and the lease often has vital information. This also allows us to associate logs with leases correctly.
Test Plan: Ran `drydock lease --type host` and saw it perform a host allocation in EC2.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D3870
Summary:
This was the major goal of D3859/D3855, and to a lesser degree D3854/D3852.
As Drydock is allocating a resource, it may need to allocate other resources first. For example, if it's allocating a working copy, it may need to allocate a host first.
Currently, we have the process basically queue up the allocation (insert a task into the queue) and sleep() until it finishes. This is problematic for a bunch of reasons, but the major one is that if allocation takes more resources (host, port, machine, DNS) than you have daemons, they could all end up sleeping and waiting for some other daemon to do their work. This is really stupid. Even if you only take up some of them, you're spending slots sleeping when you could be doing useful work.
To partially get around this and make the CLI experience less dumb, there's this goofy `synchronous` flag that gets passed around everywhere and pushes the workflow through a pile of special cases. Basically the `synchronous` flag causes us to do everything in-process. But this is dumb too because we'd rather do things in parallel if we can, and we have to have a lot of special case code to make it work at all.
Get rid of all of this. Instead of sleep()ing, try to work on the tasks that need to be worked on. If another daemon grabbed them already that's fine, but in the worst case we just gracefully degrade and do everything in process. So we get the best of both worlds: if we have parallelizable tasks and free daemons, things will execute in parallel. If we have nonparallelizable tasks or no free daemons, things will execute in process.
Test Plan: Ran `drydock_control.php --trace` and saw it perform cascading allocations without sleeping or special casing.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D3861
Summary:
Currently, when taskmasters complete a task it is immediately deleted. This prevents us from doing some general things, like:
- Supporting the idea of permanent failure (e.g., after N failures just stop trying).
- Showing the user how fast taskmasters are completing tasks.
- Showing the user how long tasks took to complete.
Having better visibility into this is important to Drydock, which builds on the task system. Also, generally buff debug output for task execution.
Test Plan: Ran `bin/phd debug taskmaster`. Ran `bin/phd debug garbage`. Queued some tasks via various systems.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T2015
Differential Revision: https://secure.phabricator.com/D3852
Summary: you can now add, edit, and delete status events. also added a "description" to status events and surface it in the big calendar view on mouse hover. some refactoring changes as well to make validation logic centralized within the storage class.
Test Plan: added, edited, deleted. yay.
Reviewers: epriestley, vrana
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T407
Differential Revision: https://secure.phabricator.com/D3810
Summary:
This leaves the UI in a pretty rough state, but implements blog policy controls and queries, and 1:1 relationships between posts and blogs. Needs a bunch more cleanup but seemed like an okayish breaking point in terms of cohesiveness.
Posts have these rules:
- Drafts are visible only to the author.
- Published posts are visible to anyone who can see the blog they appear on.
- Posts are only editable by the author.
...so we don't need any special policy UI or state to accommodate these rules.
Posts may have no blog if they're grandfathered in or you write a post to a blog and then lose the ability to see the blog. This is the messiest edge case -- specifically:
- You write a post to blog A.
- You publish the post.
- I edit the "Visible To:" for blog A and set it to exclude you.
What we do in this case is let you see the post in "My Posts", but you can no longer see the blog and you'll see the post as not being part of a blog. We can maybe give you some UI to let you move it later or something.
Test Plan: Hit all (I think?) of the interfaces without issues. Definitely some UI problems still right now.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1373
Differential Revision: https://secure.phabricator.com/D3694
Summary:
Adds "can view" and "can edit" policies to blogs. Replaces "bloggers" with "can join".
This doesn't fully remove "bloggers" because I didn't want this to get too crazy/huge.
Test Plan: Created, edited, deleted blogs.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1373
Differential Revision: https://secure.phabricator.com/D3693
Summary:
Blames to D3659.
It's not that old (< 1 day) to fix it properly (add patch for dropping xhpast DB).
Test Plan:
$ arc unit src/applications/calendar/storage/__tests__/PhabricatorCalendarHolidayTestCase.php
Reviewers: btrahan, epriestley
Reviewed By: btrahan
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3670
Summary:
We need to go slightly farther to stub reply handler functionality for Ponder in at least some configurations, where we rely on the presence of a unique random key to generate per-object or per-object+user reply addresses.
This should probably be formalized in an interface since it's currently pretty ad-hoc.
Test Plan:
- Made comments in Ponder under a per-user email configuration.
- Ran migration, verified mail keys were generated.
- Ran migration again (with --apply), verified existing questions were skipped.
- Created a new question, verified mail key generation.
Reviewers: pieter
Reviewed By: pieter
CC: aran
Maniphest Tasks: T1873
Differential Revision: https://secure.phabricator.com/D3665
Summary:
It isn't deleted by `storage destroy`.
This should be a no-op on current storage because we execute `CREATE DATABASE IF NOT EXISTS`.
Test Plan:
$ bin/storage destroy --dryrun
Reviewers: nh, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3659
Summary: this then enables people to create blog.theircompany.com. And for us, blog.phacility.com...!
Test Plan:
- created custom URIs of various goodness and verified the error messages were sensical.
- verified if "false" in configuration then custom uri stuff disappears
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T1373
Differential Revision: https://secure.phabricator.com/D3542
Summary:
It happens to me quite often that I leave the window with revision (by closing it or by visiting a link from it).
When I return then the comment draft is there so I clowncopterize it but forget that I wanted to take some other action than Comment.
Test Plan: Selected "Add Reviewers", added some reviewers, closed the window, opened it - the action and reviewers were still there.
Reviewers: btrahan, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3530
Summary:
People have occasionally complained about phabricator being slow. We have
the access log to look at to see when slowness happens, but it doesn't tell
us much about why it happened. Since it's usually a sporadic issue that's
reported, it's hard to reproduce and then profile. This change will allow us
to collect sampled profiles so we can look at them when slowness occurs.
Test Plan:
checking that sampling works correctly:
- set rate to 0; do several page loads; check no new entries in table
- set rate to 1; check that there's a new row in the table for each page load
- set rate to 10; check that some requests write to table and some don't
check new ui for samples:
- load /xhprof/list/all/, see a list with a lot of samples
- load /xhprof/list/sampled/, see only sampled runs
- load /xhprof/list/manual/, see only non-sampled runs
- load /xhprof/list/my-runs/, se only my manual runs
Reviewers: vrana, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3458
Summary: Add storage to Pastes for view policies.
Test Plan: Set policies on pastes, see next diff.
Reviewers: vrana, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T603
Differential Revision: https://secure.phabricator.com/D3474
Summary: This is pretty spartan, but it does the job.
Test Plan:
Patch, update storage, add some comment
to your favorite question or answer.
Reviewers: nh, vrana, epriestley
Reviewed By: epriestley
CC: aran, Korvin, starruler, syrneus, me.here, victorzarate7
Maniphest Tasks: T1645
Differential Revision: https://secure.phabricator.com/D3471
Summary: We use numbers here and I see no reason for strings.
Test Plan:
$ bin/storage upgrade
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3303
Summary: This is arguably a more useful view than listing all daemons.
Test Plan: Looked at list, only saw daemons that haven't exited
Reviewers: vrana, epriestley
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3286
Summary:
- In ProjectQuery, always load the viewer's membership in the project because we need it to perform a CAN_VIEW test.
- Add storage for the view, edit and join policies.
- A user can always view a project if they are a member.
- A user can always join a project if they can edit it.
- Editing a project requires both "view" and "edit" permissions, and edit does not imply view.
- This has no effect on the application yet.
Test Plan: See next diff.
Reviewers: vrana, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T603
Differential Revision: https://secure.phabricator.com/D3219
Summary:
Ponder is similar in spirit to the Wiki tool, but uses a Q&A
format and up/downvotes to signal user sentiment. Popular
questions are moved to the top of the feed on a 5-minute
cycle based on age (younger is better) and vote count (higher
is better).
Pre-apologies for noob diff.
Test Plan:
- `./bin/phd list` Should include `PonderHeatDaemon`; phd launch it
if necessary.
- Navigate to /ponder/ ; observe sanity when adding questions,
voting on them, and adding answers.
- Confirm that questions and answers are linkable using Q5 / Q5#A5 formatted object links.
- Confirm that searching for Ponder Questions works using built-in
search.
Feedback on code / schema / whatever organization very welcome.
Reviewers: nh, vrana, epriestley
Reviewed By: epriestley
CC: gmarcotte, aran, Korvin, starruler
Differential Revision: https://secure.phabricator.com/D3136
Summary: See discussion in D3078 for why I've separated this. Pretty sure it's not quite ready yet -- I want to build a couple of things on it so we have a better idea of what we need (autoincrement ID? <factType, objectA, epoch> primary key? objectB column? valueZ?) and don't need to do a ton of schema patches.
Test Plan: Applied patches, ran D3078.
Reviewers: vrana, btrahan, majak
Reviewed By: vrana
CC: aran
Maniphest Tasks: T1581, T1562
Differential Revision: https://secure.phabricator.com/D3088
Summary:
- Store project members in edges.
- Migrate existing members to edge storage.
- Delete PhabricatorProjectAffiliation.
- I left the actual underlying data around just in case something goes wrong; we can delete it evenutally.
Test Plan:
- Ran migration.
- Created a new project.
- Joined and left a project.
- Added and removed project members.
- Manually called PhabricatorOwnersOwner::loadAffiliatedUserPHIDs() to verify its behavior.
Reviewers: vrana, btrahan
Reviewed By: vrana
CC: aran
Maniphest Tasks: T603
Differential Revision: https://secure.phabricator.com/D3186
Summary:
See T1602.
This is just the minimal functional patch; the scripts will continue
working because of the `DEFAULT ''`.
Test Plan:
Can't fully test this until I get more code working, but
nothing broke horribly yet.
Reviewers: epriestley
Reviewed By: epriestley
CC: nh, aran, Korvin
Maniphest Tasks: T1602
Differential Revision: https://secure.phabricator.com/D3147
Summary:
To make it easier to monitor daemons, let's store their current state
(running, died, exited, or unknown) to the db. The purpose of this is to
provide more information on the daemon console about the status of daemons,
especially when they are running on multiple machines. This is mostly backend
work, with only a few frontend changes. (It is also dependent on a change
to libphutil.)
These changes will make dead or stuck daemons more obvious, and will allow
more work on the frontend to hide daemons (and logs) that have exited cleanly,
i.e. ones we don't care about any more.
Test Plan:
- run db migration, check in db that all daemons were marked as exited
- start up a daemon, check in db that it is marked as running
- open web interface, check that daemon is listed as running
- after daemon has been running for a little bit, check in db that dateModified
is being updated (indicating daemon is properly sending heartbeat)
- kill -9 daemon (but don't run bin/phd yet), and check that db still shows it
as running
- edit daemon db entry to show it as being on a different host, and backdate
dateModified field by 3 minutes, and check the web ui to show that the status
is unknown.
- change db entry to have proper host, check in web ui that daemon status is
displayed as dead. Check db to see that the status was saved.
- run bin/phd stop, and see that the formerly dead daemon is now exited.
Reviewers: epriestley, vrana
Reviewed By: epriestley
CC: aran, Korvin
Differential Revision: https://secure.phabricator.com/D3126
Summary:
- Add edges for this relationship.
- Use edges to store this data.
- Migrate old data.
- Fix some warnings with generating feed stories about Aux and Edge transactions.
- Fix a task-task edge issue with "Create Subtask".
Test Plan:
- Migrated data, verified reivsions showed up.
- Attached and detached tasks to revisions and vice versa.
- Created a new revision with attached tasks.
- Created a subtask.
Reviewers: btrahan, vrana
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D3018
Summary:
blogs are collections of posts. a blog also has metadata like a name, description and "bloggers" that can edit the metadata of the blog and contribute posts.
changes include the post edit flow where bloggers can now select which blogs to publish to. also made various small tweaks throughout the UI to make things sensical and clean as the concept of blogs is introduced.
there's edges powering this stuff. bloggers <=> blogs and posts <=> blogs in particular.
Test Plan:
made blogs, deleted blogs, tried to make blogs with no bloggers. all went well.
verified ui to publish only showed up for public posts, published posts to blogs, un-published posts to blogs, re-published posts to blogs, deleted posts and verified they disappeared from blogs.
Reviewers: epriestley
Reviewed By: epriestley
CC: aran, Korvin
Maniphest Tasks: T1373
Differential Revision: https://secure.phabricator.com/D3003
Summary: See D3006. Move this data to the edge store.
Test Plan:
- Created dependencies, migrated, verified dependencies were preserved.
- Added new dependencies, they worked.
Reviewers: vrana, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1162
Differential Revision: https://secure.phabricator.com/D3007
Summary:
- Use edges to store "X depends on Y" information in Maniphest.
- Show both "Depends On" and "Dependent Tasks".
- Migrate all the old edges.
Test Plan:
- Added some relationships, migrated, verified they were preserved.
- Added some new valid relationships, verified tasks got updated with sensible transactions and sent reasonable emails.
- Tried to add a cycle, got an ugly but effective error.
Reviewers: vrana, btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1162
Differential Revision: https://secure.phabricator.com/D3006
Summary:
- See D2945.
- Drop `cache` field from ManiphestTransaction.
- Render task descriptions and transactions through PhabricatorMarkupEngine.
- Also pull the list of macros more lazily.
Test Plan:
- Verified transactions and transaction preview work correctly and interact with cache correctly.
- Verified tasks descriptions and task preview work correctly.
- Verified we don't hit the imagemacro table when we're rendering everything from cache anymore.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Differential Revision: https://secure.phabricator.com/D2946
Summary:
The immediate issue this addresses is T1366, adding a rendering cache to Phriction. For wiki pages with code blocks especially, rerendering them each time is expensive.
The broader issue is that out markup caches aren't very good right now. They have three major problems:
**Problem 1: the data is stored in the wrong place.** We currently store remarkup caches on objects. This means we're always loading it and passing it around even when we don't need it, can't genericize cache management code (e.g., have one simple script to drop/GC caches), need to update authoritative rows to clear caches, and can't genericize rendering code since each object is different.
To solve this, I created a dedicated cache database that I plan to move all markup caches to use.
**Problem 2: time-variant rules break when cached.** Some rules like `**bold**` are time-invariant and always produce the same output, but some rules like `{Tnnn}` and `@username` are variant and may render differently (because a task was closed or a user is on vacation). Currently, we cache the raw output, so these time-variant rules get locked at whatever values they had when they were first rendered. This is the main reason Phriction doesn't have a cache right now -- I wanted `{Tnnn}` rules to reflect open/closed tasks.
To solve this, I split markup into a "preprocessing" phase (which does all the parsing and evaluates all time-invariant rules) and a "postprocessing" phase (which evaluates time-variant rules only). The preprocessing phase is most of the expense (and, notably, includes syntax highlighting) so this is nearly as good as caching the final output. I did most of the work here in D737 / D738, but we never moved to use it in Phabricator -- we currently just do the two operations serially in all cases.
This diff splits them apart and caches the output of preprocessing only, so we benefit from caching but also get accurate time-variant rendering.
**Problem 3: cache access isn't batched/pipelined optimally.** When we're rendering a list of markup blocks, we should be able to batch datafetching better than we do. D738 helped with this (fetching is batched within a single hunk of markup) and this improves batching on cache access. We could still do better here, but this is at least a step forward.
Also fixes a bug with generating a link in the Phriction history interface ($uri gets clobbered).
I'm using PHP serialization instead of JSON serialization because Remarkup does some stuff with non-ascii characters that might not survive JSON.
Test Plan:
- Created a Phriction document and verified that previews don't go to cache (no rows appear in the cache table).
- Verified that published documents come out of cache.
- Verified that caches generate/regenerate correctly, time-variant rules render properly and old documents hit the right caches.
Reviewers: btrahan
Reviewed By: btrahan
CC: aran
Maniphest Tasks: T1366
Differential Revision: https://secure.phabricator.com/D2945
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
