Re: The effect of changing an osd's class

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



It's not clear to me if you wanted to add some more details after "I see this:" (twice). So you do see backfilling traffic if you out the OSD? Then maybe the remapped PGs are not even on that OSD? Have you checked 'ceph pg ls remapped'? To drain an OSD, you can either set it "out" as you already tried, or set its crush weight to 0:

ceph osd crush reweight osd.0 0



Zitat von Roland Giesler <roland@xxxxxxxxxxxxxx>:

On 2024/11/14 09:37, Eugen Block wrote:
Remapped PGs is exactly what to expect after removing (or adding) a device class. Did you revert the change entirely? It sounds like you maybe forgot to add the original device class back to the OSD where you changed it? Maybe share 'ceph osd tree'? Do you have recovery IO (ceph -s)? Does the number of misplaced objects change?

NodeA:~# ceph osd tree
ID   CLASS  WEIGHT    TYPE NAME           STATUS  REWEIGHT PRI-AFF
 -1         49.82324  root default
 -3         12.36960      host FT1-NodeA
  2    hdd   1.86029          osd.2           up   1.00000 1.00000
  3    hdd   1.86029          osd.3           up   1.00000 1.00000
  4    hdd   1.86029          osd.4           up   1.00000 1.00000
  5    hdd   1.86589          osd.5           up   1.00000 1.00000
  0    ssd   0.28319          osd.0           up   1.00000 1.00000
  1    ssd   0.28319          osd.1           up   1.00000 1.00000
 36    ssd   0.20000          osd.36          up   1.00000 1.00000
 37    ssd   0.28319          osd.37          up   1.00000 1.00000
 38    ssd   0.28319          osd.38          up   1.00000 1.00000
 39    ssd   0.28319          osd.39          up   1.00000 1.00000
 40    ssd   3.30690          osd.40          up   1.00000 1.00000
 -7         12.63338      host FT1-NodeB
 10    hdd   1.86029          osd.10          up   1.00000 1.00000
 11    hdd   1.86029          osd.11          up   1.00000 1.00000
 26    hdd   1.86029          osd.26          up   1.00000 1.00000
 27    hdd   1.86029          osd.27          up   1.00000 1.00000
  6    ssd   0.28319          osd.6           up   1.00000 1.00000
  7    ssd   0.28319          osd.7           up   1.00000 1.00000
  8    ssd   0.28319          osd.8           up   1.00000 1.00000
  9    ssd   0.28319          osd.9           up   1.00000 1.00000
 24    ssd   0.28319          osd.24          up   1.00000 1.00000
 25    ssd   0.28319          osd.25          up   1.00000 1.00000
 41    ssd   3.49309          osd.41          up   1.00000 1.00000
-10         12.18689      host FT1-NodeC
 14    hdd   1.59999          osd.14          up   1.00000 1.00000
 15    hdd   1.86029          osd.15          up   1.00000 1.00000
 16    hdd   1.86029          osd.16          up   1.00000 1.00000
 17    hdd   1.86029          osd.17          up   1.00000 1.00000
 12    ssd   0.28319          osd.12          up   1.00000 1.00000
 13    ssd   0.28319          osd.13          up   1.00000 1.00000
 28    ssd   0.28319          osd.28          up   1.00000 1.00000
 29    ssd   0.28319          osd.29          up   1.00000 1.00000
 30    ssd   0.28319          osd.30          up   1.00000 1.00000
 31    ssd   0.28319          osd.31          up   1.00000 1.00000
 43    ssd   3.30690          osd.43          up   1.00000 1.00000
-13         12.63338      host FT1-NodeD
 20    hdd   1.86029          osd.20          up   1.00000 1.00000
 21    hdd   1.86029          osd.21          up   1.00000 1.00000
 22    hdd   1.86029          osd.22          up   1.00000 1.00000
 23    hdd   1.86029          osd.23          up   1.00000 1.00000
 18    ssd   0.28319          osd.18          up   1.00000 1.00000
 19    ssd   0.28319          osd.19          up   1.00000 1.00000
 32    ssd   0.28319          osd.32          up   1.00000 1.00000
 33    ssd   0.28319          osd.33          up   1.00000 1.00000
 34    ssd   0.28319          osd.34          up   1.00000 1.00000
 35    ssd   0.28319          osd.35          up   1.00000 1.00000
 42    ssd   3.49309          osd.42          up   1.00000 1.00000

No recovery IO, which is what worries me:

NodeA:~# ceph -s
  cluster:
    id:     04385b88-049f-4083-8d5a-6c45a0b7bddb
    health: HEALTH_WARN
            1 pool(s) have no replicas configured

  services:
    mon: 3 daemons, quorum FT1-NodeA,FT1-NodeB,FT1-NodeC (age 3d)
    mgr: FT1-NodeC(active, since 3d), standbys: FT1-NodeA, FT1-NodeB
    mds: 1/1 daemons up, 1 standby
    osd: 44 osds: 44 up (since 14h), 44 in (since 14h); 113 remapped pgs

  data:
    volumes: 1/1 healthy
    pools:   7 pools, 1586 pgs
    objects: 5.79M objects, 12 TiB
    usage:   24 TiB used, 26 TiB / 50 TiB avail
    pgs:     4161/11721328 objects misplaced (0.035%)
             1471 active+clean
             113  active+clean+remapped
             2    active+clean+scrubbing+deep

  io:
    client:   14 MiB/s rd, 8.5 MiB/s wr, 62 op/s rd, 1.27k op/s wr

  progress:

It stays on that "113  active+clean+remapped".

When I "out" osd.0, the status changes for a while (30 minutes estimated) and then, once the recovery IO stops, when I attempt to stop the osd, this is displayed:

The 6 "active+remapped+backfilling" pg were many more at first when i took the osd out of the cluster.  If I try to stop the osd now however, I see this:

Clearly these are more pg's than the 6 that are still backfilling.

Is there a way to force the pg's off this osd, so I safely stop it?



Zitat von Roland Giesler <roland@xxxxxxxxxxxxxx>:

On 2024/11/13 21:05, Anthony D'Atri wrote:
I would think that there was some initial data movement and that it all went back when you reverted.  I would not expect a mess.

  data:
    volumes: 1/1 healthy
    pools:   7 pools, 1586 pgs
    objects: 5.79M objects, 12 TiB
    usage:   24 TiB used, 26 TiB / 50 TiB avail
    pgs:     4161/11720662 objects misplaced (0.036%)
             1463 active+clean
             113  active+clean+remapped
             9    active+clean+scrubbing+deep+repair
             1    active+clean+scrubbing+deep

I have 113 active+clean+remapped pg's that just stay like that. If I try to out the this osd to stop it, the clsuter also never settles.  So then if I try to stop it in the GUI is tells me 73 pg's are still on the OSD...

Can I force those pg's away from the osd?

On Nov 13, 2024, at 12:48 PM, Roland Giesler <roland@xxxxxxxxxxxxxx> wrote:

I created a new osd class and changed the class of an osd to the new one without taking the osd out and stopping it first.  The new class also has a crush rule and a pool created for it.

When I realised my mistake, I reverted to what I had before. However, I suspect that I now have a mess on that osd.

What happened when I did this?

_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx


_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx


_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx




[Index of Archives]     [Information on CEPH]     [Linux Filesystem Development]     [Ceph Development]     [Ceph Large]     [Ceph Dev]     [Linux USB Development]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]     [xfs]


  Powered by Linux