Hi guys,
I upgraded a working cluster from Dumpling to Emperor which went OK. All
mons, osds and mds running 0.72.2 on Fedora 18 now.
I then installed the ceph-extras repo and let it update curl, libcurl
and leveldb.
Next I tried restarting a mon - but it won't start up again. It just
just hangs:
# service ceph start mon.e
=== mon.e ===
Starting Ceph mon.e on stor5...
The only thing logged is this:
2014-03-30 20:31:31.829959 7f3c989877c0 0 ceph version 0.72.2
(a913ded2ff138aefb8cb84d347d72164099cfd60), process ceph-mon, pid 21310
I have debugging cranked all the way up (debug ms=20, mon=20, paxos=20
and auth=20).
When looking at ps, I have a ceph-mon process and nothing else related
to this (i.e. not create keys process or similar hanging).
An strace on the process shows:
futex(0x7fffed5128d8, FUTEX_WAIT_PRIVATE, 2, NULL
"ceph mon stat" shows the following:
e3: 5 mons at
{a=10.0.0.1:6789/0,b=10.0.0.2:6789/0,c=10.0.0.3:6789/0,d=10.0.0.7:6789/0,e=10.0.0.8:6789/0},
election epoch 26362, quorum 0,1,2,3 a,b,c,d
So the rest of the cluster is working (it is mon.e I tried to restart).
Not really much to go on. Anyone seen something similar, or have an idea?
Thanks in advance,
--
Jens Kristian Søgaard, Mermaid Consulting ApS,
jens@xxxxxxxxxxxxxxxxxxxx,
http://www.mermaidconsulting.com/
_______________________________________________
ceph-users mailing list
ceph-users@xxxxxxxxxxxxxx
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
