db: improve logging observability into WAL failover #4162
Comments
jbowens
added a commit
to jbowens/pebble
that referenced
this issue
Nov 13, 2024
When we finish replaying a WAL, log the logical log which will encode all the
replayed segment files (if WAL failover resulted in multiple physical files).
```
[JOB 1] WAL 000002: {(data,000)} stopped reading at offset: (data/000002.log: 0); replayed 0 keys in 0 batches
```
Additionally, implement the SafeFormatter interface to ensure that relevant
details are not redacted.
Informs cockroachdb#4162.
jbowens
added a commit
to jbowens/pebble
that referenced
this issue
Nov 13, 2024
When we finish replaying a WAL, log the logical log which will encode all the
replayed segment files (if WAL failover resulted in multiple physical files).
```
[JOB 1] WAL 000002: {(data,000)} stopped reading at offset: (data/000002.log: 0); replayed 0 keys in 0 batches
```
Additionally, implement the SafeFormatter interface to ensure that relevant
details are not redacted.
Informs cockroachdb#4162.
jbowens
added a commit
that referenced
this issue
Nov 14, 2024
When we finish replaying a WAL, log the logical log which will encode all the
replayed segment files (if WAL failover resulted in multiple physical files).
```
[JOB 1] WAL 000002: {(data,000)} stopped reading at offset: (data/000002.log: 0); replayed 0 keys in 0 batches
```
Additionally, implement the SafeFormatter interface to ensure that relevant
details are not redacted.
Informs #4162.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
We should improve the observability into WAL failover via diagnostic logs.
We could at least log when we switch between primary and secondary.
We should also look at improving the WAL replay logic to ensure we log sufficient information about which physical WALs were read during recovery.
Jira issue: PEBBLE-302
The text was updated successfully, but these errors were encountered: