Search results

During a disaster "test" (randomly removing drives), I pulled one osd and one drive that had the journals for 4 other OSDs. So I assumed this would down/out 4 possibly 5 OSDs. Upon re-inserting the drives they were given different /dev/sd[a-z] labels, now the journal disk has the SD[a-z] label...

No problems yet, I just wanted to make sure there were no conflicts before proceeding further. The dashboard gives us a way to separate management roles.

Did you run into any issues while using the dashboard with pveceph?

I have a two node cluster setup with HA both nodes show they are online but they keep trying to fence each other; Here is a bit from the syslog: Feb 14 15:53:52 prox1 fenced[1572]: fencing node prox3 still retrying Feb 14 16:00:44 prox3 fenced[2324]: fencing node prox1 still retrying I have...

This is probably too late for you but I figure I'd let you know how to fix the issue for future refrence. For some unknown reason "blade" is not a valid descriptor for the fence_bladecenter agent, you will have to use "port" instead ex; <clusternode name="prox1" votes="1" nodeid="1">...