aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* Fix for guile-lib v0.2.8HEADmasterChristopher Baines3 days
| | | | | | | | | I'm not sure quite what's changed, but it seems like the way the formatter is being called has changed in a backwards incompatible way. To maintain compatability with both older and newer versions of guile-lib, accept the formatter being called with extra arguments, and just look at the first three.
* Use -O1 to avoid Guile bug(s)Christopher Baines3 days
| | | | | | As compiling at -O2 with guile-next (3.0.9-0.db7efa5) gives errors like: internal error: unexpected kwarg syms ((#:level level #f)) (t803)
* Tweak running sqitchChristopher Baines4 days
|
* Use the new process metrics exporterChristopher Baines9 days
|
* Guard against signaling the error condition not workingChristopher Baines9 days
| | | | As this seems to happen when the process runs out of file descriptors.q
* Fix the sleep-impl argument to retry-on-errorChristopher Baines2024-03-16
|
* Remove the now unused queue builds scriptChristopher Baines2024-03-16
| | | | This has mostly been replaced by the bffe.
* Add a definition of retry-on-error which uses the fibers sleepChristopher Baines2024-03-15
|
* Remove the core-guile-sleep variableChristopher Baines2024-03-15
| | | | As this is unnecessary.
* Remove mutexes from the upload monitoringChristopher Baines2024-03-13
| | | | Instead, store the state in the work queue.
* Actually use non-blocking ports for network requestsChristopher Baines2024-03-02
| | | | In most places at least.
* Allow setting the dynamic auth token via environment variableChristopher Baines2024-02-13
|
* Show tags with build showChristopher Baines2024-02-13
|
* Expose information about setup failuresChristopher Baines2024-02-08
|
* Expose the derived priorities for buildsChristopher Baines2024-01-30
|
* Add inbuilt port monitoringChristopher Baines2024-01-19
| | | | | | There seems to be some issue which can lead to the client and agent ports no longer being listened on. I've got no idea how to track this down, so just try and monitor for it and kill the process if it happens.
* Force output with the custom port logChristopher Baines2024-01-17
|
* Fix worker thread channel destructor loggingChristopher Baines2024-01-17
|
* Change retry-on-error to take #:ignore and #:no-retryChristopher Baines2024-01-12
| | | | And change #:ignore to better reflect ignoring the exception.
* Add more options when updating build prioritiesChristopher Baines2024-01-12
| | | | | | | Computing new derived priorities is slow, so these options allow skipping that if you know what you're doing. For example, if you're updating a batch of related builds and you want them all to have the same priorities and derived priorities, you can now do that much quicker.
* Use client errors in update-build-priorityChristopher Baines2024-01-12
| | | | So these can be passed to the agent.
* Fix build update-priorityChristopher Baines2024-01-12
| | | | As the matching builds is now just a list rather than a stream.
* Update .envrcChristopher Baines2023-12-28
| | | | For the new direnv version that uses guix shell.
* Avoid calling display for loggingChristopher Baines2023-12-28
| | | | | | As I'm getting encoding-error (put-char conversion to port encoding failed) errors from it sometimes. It doesn't happen at all, then suddenly it seems to happen continuously.
* Rewrite process-eventChristopher Baines2023-12-23
| | | | | Just in case this helps to avoid the "conversion to port encoding failed" issue.
* Add some port timeouts in to the build-success-publish-hookChristopher Baines2023-12-13
|
* Improve thread pool and work queue backtracesChristopher Baines2023-12-09
| | | | To try and get more information about "conversion to port encoding failed".
* Add delay logging to SQLite transactionsChristopher Baines2023-12-06
|
* Drop the write duration logging thresholdChristopher Baines2023-12-06
|
* Guard against exceptions in the thread poolChristopher Baines2023-12-06
|
* Don't check too-many-threads? in the thread pool for each jobChristopher Baines2023-12-06
| | | | | As this is excessive. Instead just check this when there are no jobs available.
* Avoid some database queries when running hooksChristopher Baines2023-12-06
|
* Move a hash-set! in the thread poolChristopher Baines2023-12-06
| | | | To avoid unlocking and locking the mutex.
* Re-fetch running jobs in the thread poolChristopher Baines2023-12-06
| | | | As the information might be stale when wait-condition-variable returns.
* Fix the return value for fold-buildsChristopher Baines2023-12-06
|
* Don't pass time to wait-condition-variableChristopher Baines2023-12-02
| | | | | | | | | | My intent here was to avoid threads waiting indefinitely for a signal. Unfortunately, it seems that specifying a timeout can result in wait-condition-variable entering an endless loop internally (#67194). Hopefully avoid this by not passing a time to wait-condition-variable. Maybe some new threads/fibers will need adding that periodically signal these condition variables to replicate the timeout behaviour.
* Extend %file-port-name-canonicalization to hook threadsChristopher Baines2023-11-30
|
* Set %file-port-name-canonicalization to avoid readlink syscallsChristopher Baines2023-11-28
| | | | As Guix does.
* Create index on build_startsChristopher Baines2023-11-24
| | | | As querying it is slow.
* Actually use a separate store connectionChristopher Baines2023-11-23
| | | | Fix the obvious bug in 3711cd8f1d901b424b2073dc54be4eec83c95482.
* Remove set-store-connection-timeoutChristopher Baines2023-11-20
| | | | As non blocking connections are being used now.
* Use a temporary store connection for substituting derivationsChristopher Baines2023-11-20
|
* Use port timeouts when checking substitutesChristopher Baines2023-11-20
| | | | As I think this is better than the alarm syscall.
* Switch to (mostly) non blocking store connections in the agentChristopher Baines2023-11-20
| | | | Pending #67245.
* Pass the store in to delete-outputsChristopher Baines2023-11-20
|
* Pass the store in to get-output-detailsChristopher Baines2023-11-20
|
* Don't shadow store in the agentChristopher Baines2023-11-20
|
* Re-enable fiber preemptionChristopher Baines2023-11-18
| | | | As some fibers might be running for too long.
* Avoid readlink syscalls when reading derivationsChristopher Baines2023-11-17
| | | | | | | | | strace -c reports over 10,000 readlink syscalls when reading the derivation for the hello package. By just setting the %file-port-name-canonicalization fluid, this drops to less than 10. I'm not sure if this actually improves performance, but doing less is surely better.
* Track the metrics endpoint durationChristopher Baines2023-11-17
|
generated by cgit v1.2.3 (git 2.25.4) at 2024-04-26 06:31:03 +0000