aboutsummaryrefslogtreecommitdiff
path: root/guix-data-service
Commit message (Collapse)AuthorAge
* Add retrying around inserting channel instance derivationsChristopher Baines2024-08-29
|
* Stop using profiles for channel instancesChristopher Baines2024-08-27
| | | | | Just use the store item instead. This should avoid issues where the profile gets GC'd.
* Handle concatenated data in package-derivation-data->names-and-versionsChristopher Baines2024-08-22
| | | | | As this broke when constructing the package-data from separate queries per system.
* Further reduce the chunk sizeChristopher Baines2024-08-20
|
* Increase the WAL thresholdChristopher Baines2024-08-20
|
* Make the free space requirement configurableChristopher Baines2024-08-20
|
* Increase the inferior idle secondsChristopher Baines2024-08-18
|
* Improve the implementation of checking for idle resourcesChristopher Baines2024-08-18
|
* Improve logging around use of inferiorsChristopher Baines2024-08-17
|
* Log the number of inferiors when reporting WAL issuesChristopher Baines2024-08-16
| | | | And reduce the threshold.
* Check the WAL size more frequently when using inferiorsChristopher Baines2024-08-16
| | | | | | Since getting an inferior from the pool can take some time, it's not sufficient to just check prior to attempting to fetch an inferior from the pool. Instead set a timeout and check periodically.
* Check the WAL size before computing lint warningsChristopher Baines2024-08-16
| | | | | As even though this might not use the store, the inferior still holds a store connection open.
* Use more threads for the serverChristopher Baines2024-08-14
| | | | Just in case this avoids writing responses blocking new requests coming in.
* Move the resource pools to a different threadChristopher Baines2024-08-14
| | | | As the server thread can block for a while when writing big responses.
* Allow setting the scheduler for a resource poolChristopher Baines2024-08-14
|
* Tweak handling web server errorsChristopher Baines2024-08-14
|
* Name the resource poolsChristopher Baines2024-08-14
|
* Include the pool name in the resource pool timeout exceptionChristopher Baines2024-08-14
|
* Make resource pools a recordChristopher Baines2024-08-14
| | | | So that the name is known when requesting a resource from the pool.
* Use system-ids for inserting distribution countsChristopher Baines2024-08-12
|
* Add more logging to backfilling ↵Christopher Baines2024-08-12
| | | | guix_revision_package_derivation_distribution_counts
* Reduce chunk sizeChristopher Baines2024-08-12
| | | | To try and respond quicker when the WAL is growing.
* Check the WAL size for each chunkChristopher Baines2024-08-12
| | | | | | Rather than for each system and target, as this should mean the response (pausing and allowing inferiors and store connections to be closed) is quicker.
* Start trying to handle GC happening while processing revisionsChristopher Baines2024-08-12
|
* Parallelise inserting package derivation distribution countsChristopher Baines2024-08-10
|
* Get the test suite working againChristopher Baines2024-08-07
|
* Move inserting derivations in to the load-new-guix-revision moduleChristopher Baines2024-08-07
| | | | | And start to more closely integrate it. This makes it possible to start making it faster by doing more in parallel.
* Add the fiberize utilityChristopher Baines2024-08-07
|
* Stream the render-revision-package-derivations JSON responseChristopher Baines2024-08-05
| | | | As it can be large.
* Fix the pool metricsChristopher Baines2024-07-29
|
* Tweak timeouts in resource-pool-statsChristopher Baines2024-07-29
|
* Try using 2 threads for the serverChristopher Baines2024-07-26
|
* Add worker thread utilsChristopher Baines2024-07-20
|
* Compute package derivations in chunksChristopher Baines2024-07-19
| | | | | | This allows for keeping the inferiors and store connections around for a more constant period, and allows closing the store connections and allowing the guix-daemon to clear the WAL file if needed.
* Reduce the WAL size thresholdChristopher Baines2024-07-19
|
* Rewrite the key parts of loading data to be even more parallelChristopher Baines2024-07-19
| | | | | | | Use a pool for the database connection, and a fibers promise for the package ids, and run other parts of the process in parallel too. This change also means that inserting starts as soon as some data is available, rather than when all the data is available.
* Add more fibers utilitiesChristopher Baines2024-07-19
|
* Insert package derivations earlierChristopher Baines2024-07-19
|
* Return two values from channel->source-and-derivations-by-systemChristopher Baines2024-07-19
|
* Avoid long running store connectionsChristopher Baines2024-07-19
| | | | As I think this can cause the guix-daemon WAL to grow excessively.
* Catch and ignore the new cross build derivation errorsChristopher Baines2024-07-18
|
* Refactor opening store connections when processing jobsChristopher Baines2024-07-18
| | | | And set the #:built-in-builders.
* Re-work the fibers schedulingChristopher Baines2024-07-16
| | | | | | | | Use a single thread for receiving and responding to requests, and delegate the processing of the requests to a separate set of threads. I'm hoping this will avoid the processing of requests affecting accepting new connections, or the sending of responses.
* Use a bigger buffer for requests/responsesChristopher Baines2024-07-16
| | | | As I'm hoping this will increase the data throughput for responses.
* Stream the package derivation page since it can be largeChristopher Baines2024-07-16
|
* Add more time logging in to insert-missing-derivationsChristopher Baines2024-07-16
|
* Log more timing infromation about inserting derivationsChristopher Baines2024-07-16
|
* Stop inserting missing source file narsChristopher Baines2024-07-16
| | | | | This was more an issue several years ago, so this code is not really needed now.
* Try and speed up large package derivation comparisionsChristopher Baines2024-07-12
|
* Fix delete-revisions-from-branchChristopher Baines2024-07-07
|