Merge pull request #14637 from prometheus/superq/update_storage_docs

Reword storage docs on corruption issues
This commit is contained in:
Bryan Boreham 2024-12-03 11:44:43 +00:00 committed by GitHub
commit 6ca88a9b59
No known key found for this signature in database
GPG key ID: B5690EEEBB952194

View file

@ -115,13 +115,12 @@ time series you scrape (fewer targets or fewer series per target), or you
can increase the scrape interval. However, reducing the number of series is can increase the scrape interval. However, reducing the number of series is
likely more effective, due to compression of samples within a series. likely more effective, due to compression of samples within a series.
If your local storage becomes corrupted for whatever reason, the best If your local storage becomes corrupted to the point where Prometheus will not
strategy to address the problem is to shut down Prometheus then remove the start it is recommended to backup the storage directory and restore the
entire storage directory. You can also try removing individual block directories, corrupted block directories from your backups. If you do not have backups the
or the WAL directory to resolve the problem. Note that this means losing last resort is to remove the corrupted files. For example you can try removing
approximately two hours data per block directory. Again, Prometheus's local individual block directories or the write-ahead-log (wal) files. Note that this
storage is not intended to be durable long-term storage; external solutions means losing the data for the time range those blocks or wal covers.
offer extended retention and data durability.
CAUTION: Non-POSIX compliant filesystems are not supported for Prometheus' CAUTION: Non-POSIX compliant filesystems are not supported for Prometheus'
local storage as unrecoverable corruptions may happen. NFS filesystems local storage as unrecoverable corruptions may happen. NFS filesystems