[ceph-users] HEALTH_ERR pgs are stuck inactive for more than 300 seconds

Traiano Welcome traiano at gmail.com
Tue Nov 21 07:25:01 PST 2017


Hi List

I've just begun using ceph and installed a small cluster on ubuntu
16.04 nodes using this the process in this guide:

https://www.howtoforge.com/tutorial/how-to-install-a-ceph-cluster-on-ubuntu-16-04/

However, once the installation is complete, I see the newly installed
cluster is not healthy, and complaining about pgs stuck in inactive:

---
root at lol-045:~# ceph -s

    cluster 220c92fb-2daa-4860-b511-d65ec88d6060
     health HEALTH_ERR
            448 pgs are stuck inactive for more than 300 seconds
            64 pgs degraded
            256 pgs stale
            64 pgs stuck degraded
            192 pgs stuck inactive
            256 pgs stuck stale
            256 pgs stuck unclean
            64 pgs stuck undersized
            64 pgs undersized
            noout flag(s) set
     monmap e1: 1 mons at {lol-045=17.16.2.20:6789/0}
            election epoch 4, quorum 0 lol-045
     osdmap e66: 7 osds: 4 up, 4 in; 55 remapped pgs
            flags noout,sortbitwise,require_jewel_osds
      pgmap v526: 256 pgs, 1 pools, 0 bytes data, 0 objects
            134 MB used, 6120 GB / 6121 GB avail
                 192 stale+creating
                  64 stale+active+undersized+degraded

---

Why is this, and how can troubleshoot and I fix it? (I've googled
extensively but couldn't find a solution to this).


My osd tree looks like this:

----
ID WEIGHT   TYPE NAME             UP/DOWN REWEIGHT PRIMARY-AFFINITY
-1 10.46080 root default
-2  2.98880     host anx-dp02-046
 0  1.49440         osd.0            down        0          1.00000
 4  1.49440         osd.4              up  1.00000          1.00000
-3  2.98880     host anx-dp02-047
 1  1.49440         osd.1            down        0          1.00000
 5  1.49440         osd.5              up  1.00000          1.00000
-4  2.98880     host anx-dp02-048
 2  1.49440         osd.2            down        0          1.00000
 6  1.49440         osd.6              up  1.00000          1.00000
-5  1.49440     host anx-dp02-049
 7  1.49440         osd.7              up  1.00000          1.00000
----

Many thanks in advance,
Traiano


More information about the ceph-users mailing list