Backport #27555 by @wolfogre
It should be OK to increase the default queue length since the default
type is "level".
IMO, the old default length (100) is a little too small. See
https://github.com/go-gitea/gitea/issues/27540#issuecomment-1754269491
IIRC, a larger length could lead to more memory usage only when the type
is "channel," but it's an obscure case. Otherwise, it's just a limit
(for "level" or "redis").
Co-authored-by: Jason Song <i@wolfogre.com>
# ⚠️ Breaking
Many deprecated queue config options are removed (actually, they should
have been removed in 1.18/1.19).
If you see the fatal message when starting Gitea: "Please update your
app.ini to remove deprecated config options", please follow the error
messages to remove these options from your app.ini.
Example:
```
2023/05/06 19:39:22 [E] Removed queue option: `[indexer].ISSUE_INDEXER_QUEUE_TYPE`. Use new options in `[queue.issue_indexer]`
2023/05/06 19:39:22 [E] Removed queue option: `[indexer].UPDATE_BUFFER_LEN`. Use new options in `[queue.issue_indexer]`
2023/05/06 19:39:22 [F] Please update your app.ini to remove deprecated config options
```
Many options in `[queue]` are are dropped, including:
`WRAP_IF_NECESSARY`, `MAX_ATTEMPTS`, `TIMEOUT`, `WORKERS`,
`BLOCK_TIMEOUT`, `BOOST_TIMEOUT`, `BOOST_WORKERS`, they can be removed
from app.ini.
# The problem
The old queue package has some legacy problems:
* complexity: I doubt few people could tell how it works.
* maintainability: Too many channels and mutex/cond are mixed together,
too many different structs/interfaces depends each other.
* stability: due to the complexity & maintainability, sometimes there
are strange bugs and difficult to debug, and some code doesn't have test
(indeed some code is difficult to test because a lot of things are mixed
together).
* general applicability: although it is called "queue", its behavior is
not a well-known queue.
* scalability: it doesn't seem easy to make it work with a cluster
without breaking its behaviors.
It came from some very old code to "avoid breaking", however, its
technical debt is too heavy now. It's a good time to introduce a better
"queue" package.
# The new queue package
It keeps using old config and concept as much as possible.
* It only contains two major kinds of concepts:
* The "base queue": channel, levelqueue, redis
* They have the same abstraction, the same interface, and they are
tested by the same testing code.
* The "WokerPoolQueue", it uses the "base queue" to provide "worker
pool" function, calls the "handler" to process the data in the base
queue.
* The new code doesn't do "PushBack"
* Think about a queue with many workers, the "PushBack" can't guarantee
the order for re-queued unhandled items, so in new code it just does
"normal push"
* The new code doesn't do "pause/resume"
* The "pause/resume" was designed to handle some handler's failure: eg:
document indexer (elasticsearch) is down
* If a queue is paused for long time, either the producers blocks or the
new items are dropped.
* The new code doesn't do such "pause/resume" trick, it's not a common
queue's behavior and it doesn't help much.
* If there are unhandled items, the "push" function just blocks for a
few seconds and then re-queue them and retry.
* The new code doesn't do "worker booster"
* Gitea's queue's handlers are light functions, the cost is only the
go-routine, so it doesn't make sense to "boost" them.
* The new code only use "max worker number" to limit the concurrent
workers.
* The new "Push" never blocks forever
* Instead of creating more and more blocking goroutines, return an error
is more friendly to the server and to the end user.
There are more details in code comments: eg: the "Flush" problem, the
strange "code.index" hanging problem, the "immediate" queue problem.
Almost ready for review.
TODO:
* [x] add some necessary comments during review
* [x] add some more tests if necessary
* [x] update documents and config options
* [x] test max worker / active worker
* [x] re-run the CI tasks to see whether any test is flaky
* [x] improve the `handleOldLengthConfiguration` to provide more
friendly messages
* [x] fine tune default config values (eg: length?)
## Code coverage:
Some bugs caused by less unit tests in fundamental packages. This PR
refactor `setting` package so that create a unit test will be easier
than before.
- All `LoadFromXXX` files has been splited as two functions, one is
`InitProviderFromXXX` and `LoadCommonSettings`. The first functions will
only include the code to create or new a ini file. The second function
will load common settings.
- It also renames all functions in setting from `newXXXService` to
`loadXXXSetting` or `loadXXXFrom` to make the function name less
confusing.
- Move `XORMLog` to `SQLLog` because it's a better name for that.
Maybe we should finally move these `loadXXXSetting` into the `XXXInit`
function? Any idea?
---------
Co-authored-by: 6543 <6543@obermui.de>
Co-authored-by: delvh <dev.lh@web.de>
Change all license headers to comply with REUSE specification.
Fix#16132
Co-authored-by: flynnnnnnnnnn <flynnnnnnnnnn@github>
Co-authored-by: John Olheiser <john.olheiser@gmail.com>
Convert the old mirror syncing queue to the more modern queue format.
Fix a bug in the from the repo-archive queue PR - the assumption was made that uniqueness could be enforced with by checking equality in a map in channel unique queues - however this only works for primitive types - which was the initial intention but is an imperfect. This is fixed by marshalling the data and placing the martialled data in the unique map instead.
The documentation is also updated to add information about the deprecated configuration values.
Signed-off-by: Andrew Thornton <art27@cantab.net>
There are a couple of settings in `[indexer]` relating to the `issue_indexer` queue
which override settings in unpredictable ways. This PR adjusts this hierarchy and makes
explicit that these settings are deprecated.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: techknowlogick <techknowlogick@gitea.io>
As revealed by #15964 there is inconsistent use of filepath Join and path Join
for these directories. The best thing to do is to use filepath.Join but then ToSlash
them for consistency.
Signed-off-by: Andrew Thornton <art27@cantab.net>
Co-authored-by: John Olheiser <john.olheiser@gmail.com>
This PR suggests a change to the default configuration for queues:
* Use a common DATADIR for the queues
* Set starting workers to 0 and make boost a single worker
Signed-off-by: Andrew Thornton <art27@cantab.net>
Since the move to common leveldb and common redis the disk queue code
will check the connection string before defaulting to the DATADIR.
Therefore we should ensure that the connection string is kept empty
unless it is actually set.
Fix#13023
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Upgrade levelqueue to version 0.2.0
This adds functionality for Unique Queues
* Add UniqueQueue interface and functions to create them
* Add UniqueQueue implementations
* Move TestPullRequests over to use UniqueQueue
* Reduce code duplication
* Add bytefifos
* Ensure invalid types are logged
* Fix close race in PersistableChannelQueue Shutdown
* Make WorkerPools and Queues flushable
Adds Flush methods to Queues and the WorkerPool
Further abstracts the WorkerPool
Adds a final step to Flush the queues in the defer from PrintCurrentTest
Fixes an issue with Settings inheritance in queues
Signed-off-by: Andrew Thornton <art27@cantab.net>
* Change to for loop
* Add IsEmpty and begin just making the queues composed WorkerPools
* subsume workerpool into the queues and create a flushable interface
* Add manager command
* Move flushall to queue.Manager and add to testlogger
* As per @guillep2k
* as per @guillep2k
* Just make queues all implement flushable and clean up the wrapped queue flushes
* cope with no timeout
Co-authored-by: Lauris BH <lauris@nix.lv>
