How to remove the "last logged error" msg associated with a process in status output

Using FoundationDB
1

When I log into my fdb cluster, it shows this line:
“FoundationDB processes - 318 (less 0 excluded; 2 with errors)”

The errors associated with the 2 processes in the “status details” output are like this:
“Last logged error: DiskError: io_error at Thu Dec 3 19:25:38 2020”.

I know they were generated in the past, and is no longer relevant. How can I get rid of the error msgs?

2

At least right now, the only way to clear this message is to restart the processes that had errors.

3

Like to give an update on the issue, for other people’s reference if they come into the same issue.

I encountered the same issue again recently. I noticed that there were old SQLite files under /var/lib/foundationdb/data/4500 for the storage server having “Last logged error: DiskError: io_error".

/var/lib/foundationdb/data/4500# ls -l
4.0K -rw-r--r-- 1 foundationdb foundationdb   11 Mar 15 06:58 fitness
4.0K -rw---S--- 1 foundationdb foundationdb 4.0K Nov 25 07:49 processId
 99G -rw---S--- 1 foundationdb foundationdb  99G Apr 21 15:17 storage-754d8b1d8c477ed282aa7fd85d9d197c.sqlite
 16M -rw---S--- 1 foundationdb foundationdb  16M Apr 21 15:17 storage-754d8b1d8c477ed282aa7fd85d9d197c.sqlite-wal
   0 -rw---S--- 1 foundationdb foundationdb    0 Dec  7 22:36 storage-77288e9ee94cd215e606ed4ab3e8b957.sqlite
 17M -rw---S--- 1 foundationdb foundationdb  17M Dec  3 19:25 storage-77288e9ee94cd215e606ed4ab3e8b957.sqlite-wal
4.0K -rw---S--- 1 foundationdb foundationdb 133G Dec  7 22:37 storage-e061d0d173afcdf5aa8477f9ca3f5191.sqlite
 23M -rw---S--- 1 foundationdb foundationdb  23M Dec  7 22:37 storage-e061d0d173afcdf5aa8477f9ca3f5191.sqlite-wal

Tried to delete the old files and restart the fdb server to get rid of the Last Logged Error msg. It worked.

It seems when FDB starts, it tried to open ALL the SQLite files under the “data” dir. If a (old) file is corrupt, you will see the error msg in status json.