If I run sudo ceph --admin-daemon /var/run/ceph/ceph-mon.ceph1.asok --format=plain perf dump The monitor crashes, and the command returns: admin_socket: exception: exception: no data returned from admin socket upstart restarts the monitor, and the cluster recovers. I set mon debugging to 20, and captured the attached logs. I'm running 0.67.3 on Ubuntu 12.04.3 LTS. I'm able to crash an OSD the same way. I assume it's the same bug, so I didn't attach any logs. Let me know if you want them. I'm playing around with some monitoring tools, and some of them don't understand JSON. I was hoping to get performance data without needing some python glue. Since I'm just playing around, I'll use some regexps to find what I want in the XML format. I know it's evil, but it's ok for a proof of concept, right? :-) |
Attachment:
ceph-mon.ceph1.log.bz2
Description: BZip2 compressed data
_______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com