Hi everyone!
I have an issue with my Proxmox 4.
I have a cluster proxmox on 4 servers with using Ceph.
I have a 4 monitors and 8 disks (osd).
I have mount 8 disks on 2 server (cause in ceph deployment requirements on https://pve.proxmox.com/wiki/Ceph_Server wrote that need to use at lest 4 disk per server).
And everything works fine, by sometimes (may be 1 time at week, may be every day - always different) - some of OSDs is self stopping, and storage has HEALTH_ERR status. And VM not working.
I trying to start OSD and manually wrote "ceph pg repair", but this is doesn't help - while i dont restart this 2 servers, my osd stopping with small time period - e.g. 1 hour, 30 minutes, 2 hour, 10 minutes, 1 minute, etc.
I have log with big stack trace, but i dont understand this errors.
Somebody have this trouble?
in attach - logs.
https://drive.google.com/folderview?id=0B_99J69g7QRPUERRazBScV9uMzg&usp=sharing
I have an issue with my Proxmox 4.
I have a cluster proxmox on 4 servers with using Ceph.
I have a 4 monitors and 8 disks (osd).
I have mount 8 disks on 2 server (cause in ceph deployment requirements on https://pve.proxmox.com/wiki/Ceph_Server wrote that need to use at lest 4 disk per server).
And everything works fine, by sometimes (may be 1 time at week, may be every day - always different) - some of OSDs is self stopping, and storage has HEALTH_ERR status. And VM not working.
I trying to start OSD and manually wrote "ceph pg repair", but this is doesn't help - while i dont restart this 2 servers, my osd stopping with small time period - e.g. 1 hour, 30 minutes, 2 hour, 10 minutes, 1 minute, etc.
I have log with big stack trace, but i dont understand this errors.
Somebody have this trouble?
in attach - logs.
https://drive.google.com/folderview?id=0B_99J69g7QRPUERRazBScV9uMzg&usp=sharing