I'm less current on the kernel client, so maybe there are some since-fixed bugs I'm forgetting, but: On Fri, May 9, 2014 at 8:55 AM, Aronesty, Erik <earonesty at expressionanalysis.com> wrote: > I can always remount and see them. > > But I wanted to preserve the "broken" state and see if I could figure out why it was happening. (strace isn't particularly revealing.) > > Some other things I noted was that > > - if I reboot the metadata server nobody seems to "fail over" to the hot spare (everything locks up until it's back online). I'm guessing you have to manually make the spare primary, and then switch back? That shouldn't happen. What's the output of "ceph -s"? > - if I reboot the mon that someone is mounted to, his mount locks up (even if I list 4 monitors in the fstab), but other clients still work. Can you elaborate? We discovered an issue in our userspace network code that might have an analogous problem in the kernel, but it generally was only a problem if a NIC disappeared (ie, powered off) without coming back on. -Greg Software Engineer #42 @ http://inktank.com | http://ceph.com
 |