Re: How to abandon PGs that are stuck in "incomplete"?

"wido@xxxxxxxx" <wido@xxxxxxxx> · Sun, 4 Sep 2016 10:51:43 +0200

Short reply from phone. Try object store tool on primary OSD and mark as complete.

ceph-objectstore-tool that is. Run on the primary OSD for that PG.

Wido

Op 3 sep. 2016 om 19:47 heeft Dan Jakubiec <dan.jakubiec@xxxxxxxxx> het volgende geschreven:

I think we are zero'ing in now on root cause for the stuck incomplete.  Looks like the common factor for all our stuck PGs is that they are all showing the removed OSD 8 in their "down_osds_we_would_probe" list (from "ceph pg <id> query").
For reference, I found a few archived threads of other people experiencing similar problems in the past:

  https://www.mail-archive.com/ceph-users@xxxxxxxxxxxxxx/msg13985.html
  http://ceph-users.ceph.narkive.com/jJ2DyVw7/ceph-pgs-stuck-creating-after-running-force-create-pg
  http://lists.ceph.com/pipermail/ceph-users-ceph.com/2014-August/042338.html

The general consensus from those threads is that as long as down_osds_we_would_probe is pointing to any OSD that can't be reached, those PGs will remain stuck incomplete and can't be cured by force_create_pg or even "ceph osd lost".

Question: is there any command we can run to remove the old OSD from down_osds_we_would_probe?

I did try to create an new "fake" OSD.8 today (just created the OSD, but didn't bring it all the way up), and I was able to finally run "ceph osd lost 8".  Did not seem to have any impact.

If there is no command to removed the old OSD, I think our next step will be to bring up a new/real/empty OSD.8 and see if that will clear the log jam.  But seems like there should be a tool to deal with this kind of thing?

Thanks,

-- Dan

On Sep 2, 2016, at 15:01, Dan Jakubiec <dan.jakubiec@xxxxxxxxx> wrote:

Re-packaging this question which was buried in a larger, less-specific thread from a couple of days ago.  Hoping this will be more useful here.
We have been working on restoring our Ceph cluster after losing a large number of OSDs.  We have all PGs active now except for 80 PGs that are stuck in the "incomplete" state.  These PGs are referencing OSD.8 which we removed 2 weeks ago due to corruption.

We would like to abandon the "incomplete" PGs as they are not restorable.  We have tried the following:

Per the docs, we made sure min_size on the corresponding pools was set to 1.  This did not clear the condition.
Ceph would not let us issue "ceph osd lost N" because OSD.8 had already been removed from the cluster.
We also tried "ceph pg force_create_pg X" on all the PGs.  The 80 PGs moved to "creating" for a few minutes but then all went back to "incomplete".

How do we abandon these PGs to allow recovery to continue?  Is there some way to force individual PGs to be marked as "lost"?

====

Some miscellaneous data below:

djakubiec@dev:~$ ceph osd lost 8 --yes-i-really-mean-it
osd.8 is not down or doesn't exist

djakubiec@dev:~$ ceph osd tree
ID WEIGHT   TYPE NAME       UP/DOWN REWEIGHT PRIMARY-AFFINITY
-1 58.19960 root default
-2  7.27489     host node24
 1  7.27489         osd.1        up  1.00000          1.00000
-3  7.27489     host node25
 2  7.27489         osd.2        up  1.00000          1.00000
-4  7.27489     host node26
 3  7.27489         osd.3        up  1.00000          1.00000
-5  7.27489     host node27
 4  7.27489         osd.4        up  1.00000          1.00000
-6  7.27489     host node28
 5  7.27489         osd.5        up  1.00000          1.00000
-7  7.27489     host node29
 6  7.27489         osd.6        up  1.00000          1.00000
-8  7.27539     host node30
 9  7.27539         osd.9        up  1.00000          1.00000
-9  7.27489     host node31
 7  7.27489         osd.7        up  1.00000          1.00000

BUT, even though OSD 8 no longer exists I see still lots of references to OSD 8 in various ceph dumps and query's.

Interestingly, we do still see weird entries in the CRUSH map (should I do something about these?):

# devices
device 0 device0
device 1 osd.1
device 2 osd.2
device 3 osd.3
device 4 osd.4
device 5 osd.5
device 6 osd.6
device 7 osd.7
device 8 device8
device 9 osd.9

And for what it is worth, here is the ceph -s:

    cluster 10d47013-8c2a-40c1-9b4a-214770414234
     health HEALTH_ERR
            212 pgs are stuck inactive for more than 300 seconds
            93 pgs backfill_wait
            1 pgs backfilling
            101 pgs degraded
            63 pgs down
            80 pgs incomplete
            89 pgs inconsistent
            4 pgs recovery_wait
            1 pgs repair
            132 pgs stale
            80 pgs stuck inactive
            132 pgs stuck stale
            103 pgs stuck unclean
            97 pgs undersized
            2 requests are blocked > 32 sec
            recovery 4394354/46343776 objects degraded (9.482%)
            recovery 4025310/46343776 objects misplaced (8.686%)
            2157 scrub errors
            mds cluster is degraded
     monmap e1: 3 mons at {core=10.0.1.249:6789/0,db=10.0.1.251:6789/0,dev=10.0.1.250:6789/0}
            election epoch 266, quorum 0,1,2 core,dev,db
      fsmap e3627: 1/1/1 up {0=core=up:replay}
     osdmap e4293: 8 osds: 8 up, 8 in; 144 remapped pgs
            flags sortbitwise
      pgmap v1866639: 744 pgs, 10 pools, 7668 GB data, 20673 kobjects
            8339 GB used, 51257 GB / 59596 GB avail
            4394354/46343776 objects degraded (9.482%)
            4025310/46343776 objects misplaced (8.686%)
                 362 active+clean
                 112 stale+active+clean
                  89 active+undersized+degraded+remapped+wait_backfill
                  66 active+clean+inconsistent
                  63 down+incomplete
                  19 stale+active+clean+inconsistent
                  17 incomplete
                   5 active+undersized+degraded+remapped
                   4 active+recovery_wait+degraded
                   2 active+undersized+degraded+remapped+inconsistent+wait_backfill
                   1 stale+active+clean+scrubbing+deep+inconsistent+repair
                   1 active+remapped+inconsistent+wait_backfill
                   1 active+clean+scrubbing+deep
                   1 active+remapped+wait_backfill
                   1 active+undersized+degraded+remapped+backfilling

Thanks,

-- Dan

_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com