[ceph-users] Chasing slow ops in mimic

Alex Litvak alexander.v.litvak at gmail.com
Mon Mar 11 14:26:36 PDT 2019


Hello Cephers,

I am trying to find the cause of multiple slow ops happened with my 
small cluster.  I have a 3 node  with 9 OSDs

Intel(R) Xeon(R) CPU E5-2650 v4 @ 2.20GHz
128 GB RAM
Each OSD is SSD Intel DC-S3710 800GB
It runs mimic 13.2.2 in containers.

Cluster was operating normally for 4 month and then recently I had an 
outage with multiple VMs (RBD) showing

Mar  8 07:59:42 sbc12n2-chi.siptalk.com kernel: [140206.243812] INFO: 
task xfsaild/vda1:404 blocked for more than 120 seconds.
Mar  8 07:59:42 sbc12n2-chi.siptalk.com kernel: [140206.243957] 
Not tainted 4.19.5-1.el7.elrepo.x86_64 #1
Mar  8 07:59:42 sbc12n2-chi.siptalk.com kernel: [140206.244063] "echo 0 
 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
Mar  8 07:59:42 sbc12n2-chi.siptalk.com kernel: [140206.244181] 
xfsaild/vda1    D    0   404      2 0x80000000

After examining ceph logs, i found following entries in multiple OSDs
Mar  8 07:38:52 storage1n2-chi ceph-osd-run.sh[20939]: 2019-03-08 
07:38:52.299 7fe0bdb8f700 -1 osd.13 502 get_health_metrics reporting 1 
slow ops, oldest is osd_op(client.148553.0:5996289 7.fe 
7:7f0ebfe2:::rbd_data.17bab2eb141f2.000000000000023d:head [stat,write 
2588672~16384] snapc 0=[] ondisk+write+known_if_redirected e502)
Mar  8 07:38:53 storage1n2-chi ceph-osd-run.sh[20939]: 2019-03-08 
07:38:53.347 7fe0bdb8f700 -1 osd.13 502 get_health_metrics reporting 1 
slow ops, oldest is osd_op(client.148553.0:5996289 7.fe 
7:7f0ebfe2:::rbd_data.17bab2eb141f2.00000000

Mar  8 07:43:05 storage1n2-chi ceph-osd-run.sh[28089]: 2019-03-08 
07:43:05.360 7f32536bd700 -1 osd.7 502 get_health_metrics reporting 1 
slow ops, oldest is osd_op(client.152215.0:7037343 7.1e 
7:78d776e4:::rbd_data.27e662eb141f2.0000000000000436:head [stat,write 
393216~16384] snapc 0=[] ondisk+write+known_if_redirected e502)
Mar  8 07:43:06 storage1n2-chi ceph-osd-run.sh[28089]: 2019-03-08 
07:43:06.332 7f32536bd700 -1 osd.7 502 get_health_metrics reporting 2 
slow ops, oldest is osd_op(client.152215.0:7037343 7.1e 
7:78d776e4:::rbd_data.27e662eb141f2.0000000000000436:head [stat,write 
393216~16384] snapc 0=[] ondisk+write+known_if_redirected e502)

The messages were showing on all nodes and affecting several osds on 
each node.

The trouble started at approximately 07:30 am and end 30 minutes later. 
I have not seen any slow ops since then, nor VMs showed kernel hangups 
since then.  Here is my ceph status.  I also want to note that the load 
on the cluster was minimal at the time.  Please let me know where I 
could start looking as the cluster cannot be in production with this 
failures.

  cluster:
     id:     054890af-aef7-46cf-a179-adc9170e3958
     health: HEALTH_OK

   services:
     mon: 3 daemons, quorum storage1n1-chi,storage1n2-chi,storage1n3-chi
     mgr: storage1n3-chi(active), standbys: storage1n1-chi, storage1n2-chi
     mds: cephfs-1/1/1 up  {0=storage1n2-chi=up:active}, 2 up:standby
     osd: 27 osds: 27 up, 27 in
     rgw: 3 daemons active

   data:
     pools:   7 pools, 608 pgs
     objects: 1.46 M objects, 697 GiB
     usage:   3.0 TiB used, 17 TiB / 20 TiB avail
     pgs:     608 active+clean

   io:
     client:   0 B/s rd, 91 KiB/s wr, 6 op/s rd, 10 op/s wr

Thank you in advance,



More information about the ceph-users mailing list