TrueCloudLab/restic
16
0
Fork 2
Code Issues Pull requests 1 Projects Releases Packages Wiki Activity Actions
restic/changelog/unreleased/issue-5005

14 lines
598 B
Text
Raw Normal View History

backend/retry: don't trip circuit breaker if context is canceled When the context used for a load operation is canceled, then the result is always an error independent of whether the file could be retrieved from the backend. Do not false positively trip the circuit breaker in this case. The old behavior was problematic when trying to lock a repository. When `Lock.checkForOtherLocks` listed multiple lock files in parallel and one of them fails to load, then all other loads were canceled. This cancelation was remembered by the circuit breaker, such that locking retries would fail.
2024-08-22 21:16:12 +00:00
Bugfix: Fix rare failures to retry locking a repository
Restic 0.17.0 could in rare cases fail to retry locking a repository if
one of the lock files failed to load. The lock operation failed with error
`unable to create lock in backend: circuit breaker open for file <lock/1234567890>`
The error handling has been fixed to correctly retry locking the repository.
lock: introduce short delay between failed locking retries Failed locking attempts were immediately retried up to three times without any delay between the retries. If a lock file is not found while checking for other locks, with the reworked backend retries there is no delay between those retries. This is a problem if a backend requires a few seconds to reflect file deletions in the file listings. To work around this problem, introduce a short exponentially increasing delay between the retries. The number of retries is now increased to 4. This results in delays of 5, 10 and 20 seconds between the retries.
2024-08-23 21:24:43 +00:00
In addition, restic now waits a few seconds between locking retries to
increase chances of success.
backend/retry: don't trip circuit breaker if context is canceled When the context used for a load operation is canceled, then the result is always an error independent of whether the file could be retrieved from the backend. Do not false positively trip the circuit breaker in this case. The old behavior was problematic when trying to lock a repository. When `Lock.checkForOtherLocks` listed multiple lock files in parallel and one of them fails to load, then all other loads were canceled. This cancelation was remembered by the circuit breaker, such that locking retries would fail.
2024-08-22 21:16:12 +00:00
https://github.com/restic/restic/issues/5005
https://github.com/restic/restic/pull/5011
lock: introduce short delay between failed locking retries Failed locking attempts were immediately retried up to three times without any delay between the retries. If a lock file is not found while checking for other locks, with the reworked backend retries there is no delay between those retries. This is a problem if a backend requires a few seconds to reflect file deletions in the file listings. To work around this problem, introduce a short exponentially increasing delay between the retries. The number of retries is now increased to 4. This results in delays of 5, 10 and 20 seconds between the retries.
2024-08-23 21:24:43 +00:00
https://github.com/restic/restic/pull/5012
Reference in a new issue Copy permalink