Ceph excessive debug logs in /var/log/messages fill up log file system

Issue

• The /var/log/messages file on Ceph nodes grows rapidly with [DBG]messages written every two seconds, eventually filling up the file system.

  Feb 11 13:56:19 st1001 conmon[4055351]: debug 2022-02-11T02:56:19.249+0000 7f11828cd700  0 log_channel(cluster) log [DBG] : pgmap v345081: 33 pgs: 1 active+clean+scrubbing+deep, 32 active+clean; 134 GiB data, 20 TiB used, 349 TiB / 369 TiB avail; 1.7 KiB/s wr, 0 op/s
  Feb 11 13:56:19 st1001 conmon[797842]: :56:19.251166+0000 mgr.st1001.wodfwm (mgr.1391964) 375206 : cluster [DBG] pgmap v345081: 33 pgs: 1 active+clean+scrubbing+deep, 32 active+clean; 134 GiB data, 20 TiB used, 349 TiB / 369 TiB avail; 1.7 KiB/s wr, 0 op/s
  

• These messages appear despite setting the cluster monitor log level to info.

  # ceph config set mon mon_cluster_log_file_level info