1
0
Fork 0
mirror of https://we.phorge.it/source/phorge.git synced 2024-12-24 06:20:56 +01:00
Commit graph

878 commits

Author SHA1 Message Date
epriestley
34e870819c Remove "bin/repository edit" workflow
Summary:
Ref T10748. In D14250#158181, I accepted this conditional on removing it once Conduit could handle it.

Conduit can now handle it, or at least will be able to as soon as T10748 cuts over.

Test Plan: Grepped for `repository edit`.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15839
2016-05-04 16:10:49 -07:00
epriestley
b8b700c179 Remove "repository.create" Conduit API method
Summary:
Ref T10748. This has had many problems for a long time, can't create hosted repositories, can't create cluster repositories, etc. It is obsoleted by `diffusion.repository.edit`. Remove it.

(Right now `diffusion.repository.edit` isn't a strict replacement, but it will be as soon as the URI stuff cuts over.)

Test Plan: Grepped for `repository.create`.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15838
2016-05-04 16:10:19 -07:00
epriestley
dd2b10b8f8 Guarantee repositories have unique local paths
Summary:
Ref T4039. Long ago these were more freely editable and there were some security concerns around creating a repository, then setting its local path to point somewhere it shouldn't.

Local paths are no longer editable so there's no real reason we need to provide a uniqueness guarantee anymore, but you could still make a mistake with `bin/repository move-paths` by accident, and it's a little cleaner to pull them out into their own column with a key.

(We still don't -- and, largely can't -- guarantee that two paths aren't //equivalent// since one might be symlinked to the other, or symlinked only on some hosts, or whatever, but the primary value here is as a sanity check that you aren't goofing things up and pointing a bunch of repositories at the same working copy by mistake.)

Test Plan:
  - Ran migrations.
  - Grepped for `local-path`.
  - Listed and moved paths with `bin/repository`.
  - Created a new repository, verified its local path populated correctly.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4039

Differential Revision: https://secure.phabricator.com/D15837
2016-05-04 16:09:52 -07:00
epriestley
99718b61d8 Fill in new URI credential edit web UI interfaces
Summary:
Ref T10748. Ref T10366. Allows users to set credential for new URIs.

  - Ref T7221. Our handling of the "git://" protocol is currently incorrect. This protocol is not authenticated, but is considered an SSH protocol. In the new UI, it is considered an anonymous/unauthenticated protocol instead.
  - Ref T10241. This fixes the `PassphraseCredentialControl` so it doesn't silently edit the value if the current value is not visible to you and/or not valid.

Test Plan:
Performed a whole lot of credential edits, removals, and adjustments. I'll give this additional vetting before cutting over to it.

{F1253207}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T7221, T10241, T10366, T10748

Differential Revision: https://secure.phabricator.com/D15829
2016-05-02 04:26:13 -07:00
epriestley
0ba3939ce3 Flesh out more web UI actions for new URI interface
Summary:
Ref T10748.

  - Allow users to add new URIs by clicking a button instead of knowing a secret URI.
  - Validate that URIs are actually valid URIs.
  - Add enable/disable action and strings.

Test Plan:
  - Created a new URI.
  - Tried to create a nonsense URI, created a good URI.
  - Enabled/disabled a URI.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15825
2016-04-29 17:16:15 -07:00
epriestley
c314a3672f Allow callers to query information about repository URIs from diffusion.repository.search
Summary: Ref T10748. Adds a "uris" attachment with URI information.

Test Plan: Queried URI information via Conduit, saw reasonable looking information.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15822
2016-04-29 16:12:41 -07:00
epriestley
da599386f6 Add diffusion.uri.edit for creating and editing repository URIs
Summary: Ref T10748. Brings the rest of the transactions to EditEngine, supports creating via API.

Test Plan:
  - Created a URI via API.
  - Created a URI via web.
  - Tried to apply sneaky transactions, got rejected with good error messages. <_< >_>

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15821
2016-04-29 13:55:48 -07:00
epriestley
128995f1ac Document all the hypothetical URI features we plan to support soon
Summary: Ref T10748. Ref T10366. This documents how everything is planned to work shortly.

Test Plan: Read documentation.

Reviewers: chad

Reviewed By: chad

Subscribers: eadler, scode

Maniphest Tasks: T10366, T10748

Differential Revision: https://secure.phabricator.com/D15817
2016-04-29 09:24:10 -07:00
epriestley
c8711da5ff Add repository URI view pages and IO/Display edit logic
Summary:
Ref T10748.

  - New View page for repository URIs.
  - Make display and I/O behavior (observe, mirror, read, read/write) editable.
  - Add a bunch of checks to prevent you from completely screwing up a repository by making it writable from a bunch of differnet sources.

Test Plan:
{F1249866}

{F1249867}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15816
2016-04-29 09:22:16 -07:00
epriestley
616c9ae887 Rough sketch of new repository URI editing
Summary:
Ref T10748. Ref T10366. This adds a new EditEngine, EditController, Editor, Query, and Transaction for RepositoryURIs.

None of these really do anything helpful yet, and these URIs are still unused in the actual application.

Test Plan: {F1249794}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10366, T10748

Differential Revision: https://secure.phabricator.com/D15815
2016-04-29 09:21:00 -07:00
epriestley
0459e95242 Give users a modal VCS choice when creating a new repository
Summary:
Ref T10748. Allow the new EditEngine workflow to create repositories by giving the user a modal repository type choice upfront.

(The rest of this flow is still confusing/weird, though.)

Test Plan:
  - Created a new repository.

{F1249626}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15813
2016-04-29 09:20:31 -07:00
epriestley
311de580d6 Port "Actions" to new Repository UI
Summary: Ref T10748. This brings the "Actions" items (publish/notify + autoclose enabled) into the new UI.

Test Plan:
  - Edited this stuff via EditEngine and Conduit.
  - Viewed via new Manage UI.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15811
2016-04-27 17:35:36 -07:00
epriestley
8f81930b5d Port Repository "Symbols" to Manage/Panel UI
Summary: Ref T10748. Port this, add EditEngine support, add some type validation to the transaction.

Test Plan:
  - Edited via EditEngine.
  - Edited via Conduit.
  - Viewed via Management UI.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15808
2016-04-27 17:35:03 -07:00
epriestley
dc3a13c5e8 Add bin/repository clusterize and document setup and migration for clusters
Summary: Ref T4292. This provides at least some sort of hint about how to set up cluster repositories.

Test Plan:
  - Read documentation.
  - Ran `bin/repository clusterize` to add + remove clusters.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15798
2016-04-26 10:07:17 -07:00
epriestley
550a82d438 Fix two minor formatting issues with bin/repository move-paths
Summary: This gets over-escaped instead of bolded right now, but I only ever hit it when exporting/importing and never both cleaning it up.

Test Plan: Ran `bin/repository move-paths`, saw bolded "Move" instead of ANSI escape sequences.

Reviewers: chad

Reviewed By: chad

Differential Revision: https://secure.phabricator.com/D15797
2016-04-25 12:29:15 -07:00
epriestley
892a9a1f07 Make cluster repositories more resistant to freezing
Summary:
Ref T10860. This allows us to recover if the connection to the database is lost during a push.

If we lose the connection to the master database during a push, we would previously freeze the repository. This is very safe, but not very operator-friendly since you have to go manually unfreeze it.

We don't need to be quite this aggressive about freezing things. The repository state is still consistent after we've "upgraded" the lock by setting `isWriting = 1`, so we're actually fine even if we lost the global lock.

Instead of just freezing the repository immediately, sit there in a loop waiting for the master to come back up for a few minutes. If it recovers, we can release the lock and everything will be OK again.

Basically, the changes are:

  - If we can't release the lock at first, sit in a loop trying really hard to release it for a while.
  - Add a unique lock identifier so we can be certain we're only releasing //our// lock no matter what else is going on.
  - Do the version reads on the same connection holding the lock, so we can be sure we haven't lost the lock before we do that read.

Test Plan:
  - Added a `sleep(10)` after accepting the write but before releasing the lock so I could run `mysqld stop` and force this issue to occur.
  - Pushed like this:

```
$ echo D >> record && git commit -am D && git push
[master 707ecc3] D
 1 file changed, 1 insertion(+)
# Push received by "local001.phacility.net", forwarding to cluster host.
# Waiting up to 120 second(s) for a cluster write lock...
# Acquired write lock immediately.
# Waiting up to 120 second(s) for a cluster read lock on "local001.phacility.net"...
# Acquired read lock immediately.
# Device "local001.phacility.net" is already a cluster leader and does not need to be synchronized.
# Ready to receive on cluster host "local001.phacility.net".
Counting objects: 3, done.
Delta compression using up to 8 threads.
Compressing objects: 100% (2/2), done.
Writing objects: 100% (3/3), 254 bytes | 0 bytes/s, done.
Total 3 (delta 1), reused 0 (delta 0)
BEGIN SLEEP
```

  - Here, I stopped `mysqld` from the CLI in another terminal window.

```
END SLEEP
# CRITICAL. Failed to release cluster write lock!
# The connection to the master database was lost while receiving the write.
# This process will spend 300 more second(s) attempting to recover, then give up.
```

  - Here, I started `mysqld` again.

```
# RECOVERED. Link to master database was restored.
# Released cluster write lock.
To ssh://local@localvault.phacility.com/diffusion/26/locktopia.git
   2cbf87c..707ecc3  master -> master
```

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10860

Differential Revision: https://secure.phabricator.com/D15792
2016-04-25 11:37:31 -07:00
epriestley
dc75b4bd06 Move all cluster locking logic to a separate class
Summary: Ref T10860. This doesn't change anything, it just separates all this stuff out of `PhabricatorRepository` since I'm planning to add a bit more state to it and it's already pretty big and fairly separable.

Test Plan: Pulled, pushed, browsed Diffusion.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10860

Differential Revision: https://secure.phabricator.com/D15790
2016-04-25 11:20:29 -07:00
epriestley
00885edc47 Don't try to synchronize repositories with no working copy
Summary:
Ref T4292. Sometimes, we may not have a working copy for a repository. The easiest way to get into this condition is to deactivate a repository.

We could try to clone + fetch in this case, but that's kind of complex, and there's an easy command that administrators can run manually. For now, just tell them to do that.

This affects the inactive repositories on `secure`, like rGITCOINS.

Test Plan: Removed working copy, got message.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15786
2016-04-22 08:12:19 -07:00
epriestley
711f13660e Synchronize working copies before doing a "bypassCache" commit read
Summary:
Ref T4292. When the daemons make a query for repository information, we need to make sure the working copy on disk is up to date before we serve the response, since we might not have the inforamtion we need to respond otherwise.

We do this automatically for almost all Diffusion methods, but this particular method is a little unusual and does not get this check for free. Add this check.

Test Plan:
  - Made this code throw.
  - Ran `bin/repository reparse --message ...`, saw the code get hit.
  - Ran `bin/repository lookup-user ...`, saw this code get hit.
  - Made this code not throw.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15783
2016-04-22 08:11:43 -07:00
epriestley
bd4fb3c9fa Implement bin/repository thaw for unfreezing cluster repositories
Summary:
Ref T10751. Add support tooling for manually prying your way out of trouble if disaster strikes.

Refine documentation, try to refer to devices as "devices" more consistently instead of sometimes calling them "nodes".

Test Plan: Promoted and demoted repository devices with `bin/repository thaw`.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10751

Differential Revision: https://secure.phabricator.com/D15768
2016-04-20 10:45:58 -07:00
epriestley
11aa902bd1 Show "Last Writer" and "Last Write At" in the UI, add more documentation
Summary:
Ref T10751. Make the UI more useful and explain what failure states mean and how to get out of them.

The `bin/repository thaw` command does not exist yet, I'll write that soon.

Test Plan: {F1238241}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10751

Differential Revision: https://secure.phabricator.com/D15766
2016-04-20 10:45:03 -07:00
epriestley
b9cf9e6f0d Fix an issue with PHID/handle management in push logs
Summary: Ref T10751. This cleans this up so it's a little more modern, and fixes a possible bad access on the log detail page.

Test Plan: Viewed push log list, viewed push log detail.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10751

Differential Revision: https://secure.phabricator.com/D15765
2016-04-20 04:47:10 -07:00
epriestley
287e761f19 Make repository synchronization safer when leaders are ambiguous
Summary:
Ref T4292. Right now, repository versions only get marked when a write happens.

This potentially creates a problem: if I pushed all the sync code to `secure` and enabled `secure002` as a repository host, the daemons would create empty copies of all the repositories on that host.

Usually, this would be fine. Most repositories have already received a write on `secure001`, so that working copy has a verison and is a leader.

However, when a write happened to a rarely-used repository (say, rKEYSTORE) that hadn't received any write recently, it might be sent to `secure002` randomly. Now, we'd try to figure out if `secure002` has the most up-to-date copy of the repository or not.

We wouldn't be able to, since we don't have any information about which node has the data on it, since we never got a write before. The old code could guess wrong and decide that `secure002` is a leader, then accept the write. Since this would bump the version on `secure002`, that would //make// it an authoritative leader, and `secure001` would synchronize from it passively (or on the next read or write), which would potentially destroy data.

Instead:

  - Refuse to continue in situations like this.
  - When a repository is on exactly one device, mark it as a leader with version "0".
  - When a repository is created into a cluster service, mark its version as "0" on all devices (they're all leaders, since the repository is empty).

This should mean that we won't lose data no matter how much weird stuff we run into.

Test Plan:
  - In single-node mode, used `repository update` to verify that `0` was written properly.
  - With multiple nodes, used `repository update` to verify that we refuse to continue.
  - Created a new repository, verified versions were initialized correctly.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15761
2016-04-19 13:07:02 -07:00
epriestley
6edf181a7e Record which cluster host received a push
Summary: Ref T4292. When we write a push log, also log which node received the request.

Test Plan: {F1230467}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15759
2016-04-19 13:06:30 -07:00
epriestley
d87c500002 Synchronize (hosted, clustered, Git) repositories over Conduit + HTTP
Summary:
Ref T4292. We currently synchronize hosted, clustered, Git repositories when we receive an SSH pull or push.

Additionally:

  - Synchronize before HTTP reads and writes.
  - Synchronize reads before Conduit requests.

We could relax Conduit eventually and allow Diffusion to say "it's OK to give me stale data".

We could also redirect some set of these actions to just go to the up-to-date host instead of connecting to a random host and synchronizing it. However, this potentially won't work as well at scale: if you have a larger number of servers, it sends all of the traffic to the leader immediately following a write. That can cause "thundering herd" issues, and isn't efficient if replicas are in different geographical regions and the write just went to the east coast but most clients are on the west coast. In large-scale cases, it's better to go to the local replica, wait for an update, then serve traffic from it -- particularly given that writes are relatively rare. But we can finesse this later once things are solid.

Test Plan:
  - Pushed and pulled a Git repository over HTTP.
  - Browsed a Git repository from the web UI.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15758
2016-04-19 13:05:45 -07:00
epriestley
31bc023eff Synchronize (hosted, git, clustered, SSH) repositories prior to reads
Summary:
Ref T4292. Before we write or read a hosted, clustered Git repository over SSH, check if another version of the repository exists on another node that is more up-to-date.

If such a version does exist, fetch that version first. This allows reads and writes of any node to always act on the most up-to-date code.

Test Plan: Faked my way through this and got a fetch via `bin/repository update`; this is difficult to test locally and needs more work before we can put it in production.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15757
2016-04-19 13:05:17 -07:00
epriestley
0db6eaca41 Consolidate handling of SSH usernames
Summary:
Ref T4292. This consolidates code for figuring out which user we should connect to hosts with.

Also narrows a lock window.

Test Plan: Browsed Diffusion, pulled and pushed through an SSH proxy.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15754
2016-04-19 13:04:04 -07:00
epriestley
575c01373e Extract repository command construction from Repositories
Summary:
Ref T4292. Ref T10366. Depends on D15751. Today, generating repository commands is purely a function of the repository, so they use protocols and credentials based on the repository configuration.

For example, a repository with an SSH "remote URI" always generate SSH "remote commands".

This needs to change in the future:

  - After T10366, repositories won't necessarily just have one type of remote URI. They can only have one at a time still, but the repository itself won't change based on which one is currently active.
  - For T4292, I need to generate intracluster commands, regardless of repository configuration. These will have different protocols and credentials.

Prepare for these cases by separating out command construction, so they'll be able to generate commands in a more flexible way.

Test Plan:
  - Added unit tests.
  - Browsed diffusion.
  - Ran `bin/phd debug pull` to pull a bunch of repos.
  - Ran daemons.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292, T10366

Differential Revision: https://secure.phabricator.com/D15752
2016-04-19 04:51:48 -07:00
epriestley
f424f9f2d2 Record more details about where a write is taking place while holding a cluster lock
Summary: Ref T4292. This will let the UI and future `bin/repository` tools give administrators more tools to understand problems when reporting or resolving them.

Test Plan:
  - Pushed fully clean repository.
  - Pushed previously-pushed repository.
  - Forced write to abort, inspected useful information in the database.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15748
2016-04-18 11:55:27 -07:00
epriestley
368d2d1ddb Improve robustness of cluster version bookkeeping
Summary:
Ref T4292. Small fixes:

  - There was a bug with the //first// write, where we'd write 1 but expect 0. Fix this.
  - Narrow the window where we hold the `isWriting` lock: we don't need to wait for the client to finish.
  - Release the lock even if something throws.
  - Use a more useful variable name.

Test Plan:
  - Made new writes to a fresh cluster repository.
  - Made sequential writes.
  - Made concurrent writes.
  - Made good writes and bad writes.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15747
2016-04-18 11:54:59 -07:00
epriestley
595f203816 Correct RepositoryURI schema and propagate adjust exit code correctly
Summary:
Fixes T10830.

  - The return code from `storage adjust` did not propagate correct.
  - There was one column issue which I missed the first time around because I had a bunch of unrelated stuff locally.

Test Plan:
  - Ran `bin/storage upgrade -f` with failures, used `echo $?` to make sure it exited nonzero.
  - Got fully clean `bin/storage adjust` by dropping all my extra local tables.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10830

Differential Revision: https://secure.phabricator.com/D15746
2016-04-18 08:11:22 -07:00
epriestley
9352ed8abb Add missing RepositoryURI table + run storage adjustments in tests
Summary:
Fixes T10830. Ref T10366. I wasn't writing to this table yet so I didn't build it, but the fact that `bin/storage adjust` would complain slipped my mind.

  - Add the table.
  - Make the tests run `adjust`. This is a little slow (a few extra seconds) but we could eventually move some steps like this to run server-side only.

Test Plan: Ran `bin/storage upgrade -f`, got a clean `adjust`.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10366, T10830

Differential Revision: https://secure.phabricator.com/D15744
2016-04-18 07:54:02 -07:00
epriestley
fbfe730452 Support more transactions types in RepositoryEditEngine
Summary:
Ref T10748. This supports more transaction types in the modern editor and improves validation so Conduit benefits.

You can technically create repositories via `diffusion.repository.edit` now, although they aren't very useful.

Test Plan:
  - Used `diffusion.repository.edit` to create and edit repositories.
  - Used `/editpro/` to edit repositories.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10748

Differential Revision: https://secure.phabricator.com/D15740
2016-04-17 16:27:02 -07:00
epriestley
92c50de8aa Rough in the new custom URI panel
Summary: Ref T10748. Ref T10366. No support for editing and no impact on the UI, but get some of the basics in place.

Test Plan: {F1223279}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10366, T10748

Differential Revision: https://secure.phabricator.com/D15742
2016-04-17 16:03:24 -07:00
epriestley
adf42db5ea Trivially implement RepositoryEditEngine and API methods
Summary: Ref T10748. Ref T10337. This technically implements this stuff, but it does not do anything useful yet. This skips all the hard stuff.

Test Plan:
  - Technically used `diffusion.repository.search` to get repository information.
  - Technically used `diffusion.repository.edit` to change a repository name.
  - Used `editpro/` to edit a repository name.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10337, T10748

Differential Revision: https://secure.phabricator.com/D15736
2016-04-17 16:02:13 -07:00
epriestley
d96b6506d7 Disable repository read/write synchronization for now
Summary:
This nearly works but I didn't have time to get back to it and it isn't stable enough to turn on in the cluster yet.

We have enough other stuff going out this week, so just disable it before `stable` gets cut. Should be ready by next week if things go well.

Test Plan: Fetched a Git SSH repo locally.

Reviewers: chad

Reviewed By: chad

Differential Revision: https://secure.phabricator.com/D15731
2016-04-15 15:39:36 -07:00
June Rhodes
7150aa8e19 Use Conduit in PhabricatorRepositoryGitCommitChangeParserWorker
Summary:
Ref T2783.  This allows this worker to run on a machine different to the one that stores the repository, by routing the execution of Git over Conduit calls.

This API method is super gross, but fixing it isn't straightforward and it runs into other complicated considerations. We can fix it later; for now, just define it as "internal" to limit how much mess this creates.

"Internal" methods do not appear on the console.

Test Plan: Ran `bin/repository reparse --change <commit> --trace` on several commits, saw daemons make a Conduit call instead of running a `git` command.

Reviewers: hach-que, chad

Reviewed By: chad

Subscribers: joshuaspence, Korvin, epriestley

Maniphest Tasks: T2783

Differential Revision: https://secure.phabricator.com/D11874
2016-04-14 04:53:03 -07:00
epriestley
33060d1652 Ignore post-write repository synchronization if no devices are configured
Summary: Fixes T10789. If we aren't configured with a device, we never grabbed a lock in the first place, and should not expect one to be held.

Test Plan: Pushed non-cluster-configured Git SSH repository.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10789

Differential Revision: https://secure.phabricator.com/D15692
2016-04-12 18:31:11 -07:00
epriestley
4244cad990 Move toward multi-master replicated repositories
Summary:
Ref T4292. This mostly implements the locking/versioning logic for multi-master repositories. It is only active on Git SSH pathways, and doesn't actually do anything useful yet: it just does bookkeeping so far.

When we read (e.g., `git fetch`) the logic goes like this:

- Get the read lock (unique to device + repository).
  - Read all the versions of the repository on every other device.
  - If any node has a newer version:
    - Fetch the newer version.
    - Increment our version to be the same as the version we fetched.
- Release the read lock.
- Actually do the fetch.

This makes sure that any time you do a read, you always read the most recently acknowledged write. You may have to wait for an internal fetch to happen (this isn't actually implemented yet) but the operation will always work like you expect it to.

When we write (e.g., `git push`) the logic goes like this:

- Get the write lock (unique to the repository).
  - Do all the read steps so we're up to date.
  - Mark a write pending.
    - Do the actual write.
  - Bump our version and mark our write finished.
- Release the write lock.

This allows you to write to any replica. Again, you might have to wait for a fetch first, but everything will work like you expect.

There's one notable failure mode here: if the network connection between the repository node and the database fails during the write, the write lock might be released even though a write is ongoing.

The "isWriting" column protects against that, by staying locked if we lose our connection to the database. This will currently "freeze" the repository (prevent any new writes) until an administrator can sort things out, since it'd dangerous to continue doing writes (we may lose data).

(Since we won't actually acknowledge the write, I think, we could probably smooth this out a bit and make it self-healing //most// of the time: basically, have the broken node rewind itself by updating from another good node. But that's a little more complex.)

Test Plan:
  - Pushed changes to a cluster-mode repository.
  - Viewed web interface, saw "writing" flag and version changes.
  - Pulled changes.
  - Faked various failures, got sensible states.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15688
2016-04-12 08:57:57 -07:00
epriestley
58eef68b7c Rough cut of repository cluster status panel
Summary:
Ref T4292. This adds some very basic cluster/device data to the new management view. Nothing interesting yet.

Also deal with disabled bindings a little more cleanly.

Test Plan: {F1214619}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T4292

Differential Revision: https://secure.phabricator.com/D15685
2016-04-12 05:38:10 -07:00
epriestley
0216fac30a Make PullLocal smart about which repositories it should pull
Summary:
Ref T10756. When repositories are properly configured for the cluster (which is hard to set up today), be smart about which repositories are expected to exist on the current host, and only pull them.

This generally allows daemons to pretty much do the right thing no matter how many copies are running, although there may still be some lock contention issues that need to be sorted out.

Test Plan: {F1214483}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10756

Differential Revision: https://secure.phabricator.com/D15682
2016-04-12 05:37:30 -07:00
epriestley
8d6488f290 Fix a typo in bin/repository help update
Summary: Fixes T10741. The workflow is `refs`, not `ref`.

Test Plan: o.O

Reviewers: chad, cspeckmim

Reviewed By: cspeckmim

Maniphest Tasks: T10741

Differential Revision: https://secure.phabricator.com/D15652
2016-04-07 05:39:37 -07:00
epriestley
601aaa5a86 Modularize content sources
Summary:
Ref T10537. For Nuance, I want to introduce new sources (like "GitHub" or "GitHub via Nuance" or something) but this needs to modularize eventually.

Split ContentSource apart so applications can add new content sources.

Test Plan:
This change has huge surface area, so I'll hold it until post-release. I think it's fairly safe (and if it does break anything, the breaks should be fatals, not anything subtle or difficult to fix), there's just no reason not to hold it for a few hours.

- Viewed new module page.
- Grepped for all removed functions/constants.
- Viewed some transactions.
- Hovered over timestamps to get content source details.
- Added a comment via Conduit.
- Added a comment via web.
- Ran `bin/storage upgrade --namespace XXXXX --no-quickstart -f` to re-run all historic migrations.
- Generated some objects with `bin/lipsum`.
- Ran a bulk job on some tasks.
- Ran unit tests.

{F1190182}

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10537

Differential Revision: https://secure.phabricator.com/D15521
2016-03-26 11:59:45 -07:00
epriestley
f82db7524b Add a "Build with CircleCI" build step
Summary: Ref T9456. Some rough edges and we can't complete the build yet since I haven't written a webhook, but this mostly seems to be working.

Test Plan:
  - Ran this build on some stuff.
  - Ran a normal HTTP step build to make sure I didn't break that.

{F880301}

{F880302}

{F880303}

Reviewers: chad

Reviewed By: chad

Subscribers: JustinTulloss, joshma

Maniphest Tasks: T9456

Differential Revision: https://secure.phabricator.com/D14286
2016-03-22 12:12:11 -07:00
epriestley
78e36d6b17 Implement DestructibleInterface on GitLFS refs
Summary: Ref T7789. Make sure these get cleaned up when a repository is destroyed.

Test Plan:
  - Created a new repository.
  - Pushed some LFS data to it.
  - Used `bin/remove destroy` to nuke it.
  - Verified the LFS stuff was cleaned up and the underlying files were destroyed (`SELECT * FROM repository_gitlfsref`, etc).

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T7789

Differential Revision: https://secure.phabricator.com/D15493
2016-03-18 09:36:54 -07:00
epriestley
f46686ff58 Implement a Git LFS link table and basic batch API
Summary:
Ref T7789. This implements:

  - A new table to store the `<objectHash, filePHID>` relationship between Git LFS files and Phabricator file objects.
  - A basic response to `batch` commands, which return actions for a list of files.

Test Plan:
Ran `git lfs push origin master`, got a little further than previously:

```
epriestley@orbital ~/dev/scratch/poemslocal $ git lfs push origin master
Git LFS: (2 of 1 files) 174.24 KB / 87.12 KB
Git LFS operation "upload/b7e0aeb82a03d627c6aa5fc1bbfd454b6789d9d9affc8607d40168fa18cf6c69" is not supported by this server.
Git LFS operation "upload/b7e0aeb82a03d627c6aa5fc1bbfd454b6789d9d9affc8607d40168fa18cf6c69" is not supported by this server.
```

With `GIT_TRACE=1`, this shows the batch part of the API going through.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T7789

Differential Revision: https://secure.phabricator.com/D15489
2016-03-17 17:15:20 -07:00
epriestley
51153a580c Implement "git-lfs-authenticate" over SSH
Summary:
Ref T7789. This implements a (probably) usable "git-lfs-authenticate" on top of the new temporary token infrastructure.

This won't actually do anything yet, since nothing reads the tokens.

Test Plan:
```
$ ./bin/ssh-exec --phabricator-ssh-user admin --ssh-command 'git-lfs-authenticate'
phabricator-ssh-exec: Expected `git-lfs-authenticate <path> <operation>`, but received too few arguments.
```

```
$ ./bin/ssh-exec --phabricator-ssh-user admin --ssh-command 'git-lfs-authenticate x'
phabricator-ssh-exec: Unrecognized repository path "x". Expected a path like "/diffusion/X/" or "/diffusion/123/".
```

```
$ ./bin/ssh-exec --phabricator-ssh-user admin --ssh-command 'git-lfs-authenticate diffusion/22'
Exception: Expected `git-lfs-authenticate <path> <operation>`, but received too few arguments.
```

```
$ ./bin/ssh-exec --phabricator-ssh-user admin --ssh-command 'git-lfs-authenticate diffusion/22 y'
Exception: Git LFS operation "y" is not supported by this server.
```

```
$ ./bin/ssh-exec --phabricator-ssh-user admin --ssh-command 'git-lfs-authenticate diffusion/22 upload'
{"header":{"Authorization":"Basic QGdpdC1sZnM6NmR2bDVreWVsaXNuMmtnNXBtbnZwM3VlaWhubmI1bmI="},"href":"http:\/\/local.phacility.com\/diffusion\/22\/new-callsign-free-repository.git\/info\/lfs"}
```

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T7789

Differential Revision: https://secure.phabricator.com/D15482
2016-03-17 08:08:00 -07:00
epriestley
5520729db3 Remove recently added repository_pathchange key
Summary:
Ref T10560. Reverts D15460. See that task for discussion: we dug up some more information to explain the behavior, and this key was just sort of sidestepping an analyze/cardinality estimate issue on the index.

With proper cardinality estimates it shouldn't be used, so just nuke it.

Test Plan: Ran `bin/storage adjust`, saw key drop.

Reviewers: eadler, chad

Reviewed By: chad

Maniphest Tasks: T10560

Differential Revision: https://secure.phabricator.com/D15486
2016-03-16 14:58:56 -07:00
epriestley
ca4c0db2c1 Add a key to improve Diffusion's cache fill history query
Summary:
Ref T10560. I don't fully understand what MySQL is doing here, but it looks like this key improves the problematic dataset in practice.

(It makes sense that this key helps, I'm just not sure why the two separate keys and the UNION ALL are so bad.)

This key isn't hugely expensive to add, so we can try it and see if there are still issues.

Test Plan: Ran `bin/storage adjust`, saw key added to table. Used `SHOW CREATE TABLE ...` to verify the key exists. Used `EXPLAIN SELECT ...` to make sure MySQL actually uses it.

Reviewers: chad

Reviewed By: chad

Maniphest Tasks: T10560

Differential Revision: https://secure.phabricator.com/D15460
2016-03-10 17:38:36 -08:00
epriestley
abb4c03b47 Remove shouldShowSubscribersProperty() from SubscribableInterface
Summary:
Every caller returns `true`. This was added a long time ago for Projects, but projects are no longer subscribable.

I don't anticipate needing this in the future.

Test Plan: Grepped for this method.

Reviewers: chad

Reviewed By: chad

Differential Revision: https://secure.phabricator.com/D15409
2016-03-06 06:01:36 -08:00