Team,
We have our production database going down when there is a audit log rotation happens. This is not during every rotation but occassionally we do see this. we thought this might be due to the size of the audit log so we changed the frequency of the rotation.
limit: 1, singleBatch: true, lsid: { id: UUID("2aaac54e-6ba6-4e8d-8a18-f035c1cc739c") }, $clusterTime: { clusterTime: Timestamp(1693504361, 1), signature:
{ hash: BinData(0, FC15F08BB3FBE72FD3F66B4428E8A91336E9C629), keyId: 7231091013530419201 } }, $db: "admin", $readPreference: { mode: "primaryPreferred" }
}“}}}
{“t”:{”$date":“2023-08-31T23:22:42.864+05:30”},“s”:“E”, “c”:“-”, “id”:29017, “ctx”:“JournalFlusher”,“msg”:“Audit system cannot write {datalen} byte
s to log file {file}. Write failed with fatal error {err_desc}. As audit cannot make progress, the server will now shut down.”,“attr”:{“datalen”:228076,“file
“:”/dberrlog/audit/audit.json”,“err_desc”:“Success”}}
We are using 5.0.18 version of percona mongo. Since it is in a replicaset we do not have much of an impact to the application.
Regards,
Sharath