Bug #58436
openceph cluster log reporting log level in numeric format for the clog messages
0%
Description
The cluster log now reporting log level in integer value compared to human readable log level e.g DBG, INF etc
1673554794.7540936 mon.a (mon.0) 143 : audit 0 from='client.? 10.74.128.232:0/96232040' entity='client.admin' cmd=[{"prefix": "status"}]: dispatch
1673554796.6712537 mgr.x (mgr.4110) 25 : cluster 0 pgmap v44: 3 pgs: 3 active+clean; 579 KiB data, 3.0 GiB used, 300 GiB / 303 GiB avail
1673554798.6722813 mgr.x (mgr.4110) 26 : cluster 0 pgmap v45: 3 pgs: 3 active+clean; 579 KiB data, 3.0 GiB used, 300 GiB / 303 GiB avail
Updated by Prashant D over 1 year ago
- Status changed from New to Fix Under Review
- Pull request ID set to 49730
Updated by Neha Ojha 11 months ago
- Related to Bug #61547: Logs produce non-human-readable timestamps after monitor upgrade added
Updated by Prashant D 3 months ago
We are encountering this issue in teuthology runs :
https://pulpito.ceph.com/yuriw-2024-01-10_19:18:18-rados-wip-pdhange-testing-distro-default-smithi/7512554/
2024-02-07T18:51:42.108862+0000 mon.a (mon.0) 903 : cluster 1 Health check cleared: OSDMAP_FLAGS (was: nodeep-scrub flag(s) set) 2024-02-07T18:51:42.116129+0000 mon.a (mon.0) 905 : cluster 0 osdmap e148: 8 total, 4 up, 6 in 2024-02-07T18:51:42.108862+0000 mon.a (mon.0) 903 : cluster 1 Health check cleared: OSDMAP_FLAGS (was: nodeep-scrub flag(s) set) 2024-02-07T18:51:42.116129+0000 mon.a (mon.0) 905 : cluster 0 osdmap e148: 8 total, 4 up, 6 in
It should log loglevel in human readable format :
2024-02-07T18:51:42.108862+0000 mon.a (mon.0) 903 : cluster [INF] Health check cleared: OSDMAP_FLAGS (was: nodeep-scrub flag(s) set) 2024-02-07T18:51:42.116129+0000 mon.a (mon.0) 905 : cluster [DBG] osdmap e148: 8 total, 4 up, 6 in 2024-02-07T18:51:42.108862+0000 mon.a (mon.0) 903 : cluster [INF] Health check cleared: OSDMAP_FLAGS (was: nodeep-scrub flag(s) set) 2024-02-07T18:51:42.116129+0000 mon.a (mon.0) 905 : cluster [DBG] osdmap e148: 8 total, 4 up, 6 in
Updated by Yuri Weinstein about 1 month ago
Updated by Sridhar Seshasayee about 1 month ago
Radoslaw Zarzynski wrote:
Do we need to backport?
Yes, this along with https://tracker.ceph.com/issues/64314 needs to be backported to squid.