Hi,
Thank you for the suggestions. However, we are already running with those configurations but still have the issue with many logs.
Any other suggestions?

@Shaban-Mahmood Are you concerned that the logs are large because of those two log entries? How verbose are they? Are they causing the logs to be rotated much sooner than you prefer? I'm trying to understand the real motivation and severity for why these need to be suppressed.

Sorry for the late reply. We have a lot of log entries from rook-ceph and they are spamming our logs, making it hard to see relevant logs. Especially just after our system have deployed, it spams the log and I have seen 80.000 and 150.000 log entries within 5 min after deploying our system and it continues.

can you try to upgrade to latest version to see if those are gone? v1.5.5 is pretty old

I'll try, but I can't guarantee anything
I wasn't the one who set up Rook on this cluster.

Make sure to review the upgrade guide to upgrade between each release.

I did the upgrade and the cluster is currently running the v1.15.3 version
I can see the same logs sadly

@GabMgt Did you try the workaround here?

What is the meaning of the percent value?

Log lines statistics of last 15 minutes. There are a lot of debug level logs, but I didn't change any default log level (the default level in operator helm values is INFO).
Just like this:

Here are examples of the logs we are receiving, which we would like to limit:
cluster [DBG] pgmap v988: 169 pgs: 169 active+clean; 15 GiB data, 31 GiB used, 23 TiB / 23 TiB avail; 1.1 KiB/s rd, 456 KiB/s wr, 43 op/s
audit 2024-05-17T09:14:53.900401+0000 mon.a (mon.0) 2296 : audit [DBG] from='admin socket' entity='admin socket' cmd='mon_status' args=[]: dispatch

One workaround for any log entry with these prefixes can be filtered out completely with the following toolbox command for each daemon where you want to apply it:

# suppress logs with the prefix "debug"
ceph config set mon.a "log to stderr" false

# suppress logs with the prefix "cluster" and "audit"
ceph config set mon.a "mon cluster log to stderr" false

However, this may hide too much in case troubleshooting is needed. But does this help?

Thank you so much for those clues. I searched the web and found the documentation:

ceph config set global log_to_stderr false
ceph config set global mon_cluster_log_to_stderr false

It seems that it's not a Rook issue but a Ceph one.
It worked and thank you again.