Hi, a 'little question' I wonder if anyone can comment,
I've got a pair of ProxmoxVE hosts deployed recently in 2-node config which use an iSCSI target on another device to give 'quorum' as the '3rd data point' for the config. (ie, in case one host goes offline we still have quorum on the remaining host ..)
What I'm noticing is a LOT of messages logged in /var/log/messages that appear thus,
I'm also slightly concerned because yesterday afternoon one of these hosts spontaneously hung up - ie, power was on, console was black and non-responsive, ping was not responding / no access via SSH; had to do a hard power off / power on - and it booted back up normally. At the time this happened there were a similar boat load of these events, ie,
So, I guess 2-fold question,
-- does anyone know what is up with the qdiskd messages, is there an easy way to deal with this / or is it a real sign of an actual problem ?
-- I am not sure if the seize-up yesterday was related or not; probably can't assume that it is linked; but - if anyone knows of such behaviour then any comments are very much appreciated.
Thanks,
Tim
I've got a pair of ProxmoxVE hosts deployed recently in 2-node config which use an iSCSI target on another device to give 'quorum' as the '3rd data point' for the config. (ie, in case one host goes offline we still have quorum on the remaining host ..)
What I'm noticing is a LOT of messages logged in /var/log/messages that appear thus,
Code:
Jul 3 09:59:22 proxmox-a qdiskd[5036]: qdiskd: read (system call) has hung for 5 seconds
Jul 3 09:59:22 proxmox-a qdiskd[5036]: In 5 more seconds, we will be evicted
Jul 3 09:59:23 proxmox-a qdiskd[5036]: qdiskd on node 2 reports hung read()
Jul 3 09:59:23 proxmox-a qdiskd[5036]: qdisk cycle took more than 1 second to complete (6.670000)
Jul 3 10:00:03 proxmox-a qdiskd[5036]: qdisk cycle took more than 1 second to complete (3.760000)
Jul 3 10:00:07 proxmox-a qdiskd[5036]: qdisk cycle took more than 1 second to complete (1.810000)
Jul 3 10:00:39 proxmox-a qdiskd[5036]: qdisk cycle took more than 1 second to complete (1.350000)I'm also slightly concerned because yesterday afternoon one of these hosts spontaneously hung up - ie, power was on, console was black and non-responsive, ping was not responding / no access via SSH; had to do a hard power off / power on - and it booted back up normally. At the time this happened there were a similar boat load of these events, ie,
Code:
Jul 2 14:55:38 proxmox-a qdiskd[5088]: qdisk cycle took more than 1 second to complete (2.810000)
Jul 2 15:30:51 proxmox-a qdiskd[5088]: qdiskd: read (system call) has hung for 5 seconds
Jul 2 15:30:51 proxmox-a qdiskd[5088]: In 5 more seconds, we will be evicted
Jul 2 15:30:52 proxmox-a qdiskd[5088]: qdiskd on node 2 reports hung read()
Jul 2 15:30:52 proxmox-a qdiskd[5088]: qdisk cycle took more than 1 second to complete (6.330000)
Jul 2 15:31:44 proxmox-a qdiskd[5088]: qdiskd: read (system call) has hung for 5 seconds
Jul 2 15:31:44 proxmox-a qdiskd[5088]: In 5 more seconds, we will be evicted
Jul 2 15:31:48 proxmox-a kernel: connection1:0: detected conn error (1011)
Jul 2 15:32:06 proxmox-a kernel: connection1:0: detected conn error (1019)
Jul 2 15:32:09 proxmox-a qdiskd[5088]: qdiskd on node 2 reports hung read()
Jul 2 15:32:09 proxmox-a qdiskd[5088]: qdiskd on node 2 reports hung read()
Jul 2 15:32:09 proxmox-a qdiskd[5088]: qdiskd on node 2 reports hung read()
Jul 2 15:32:09 proxmox-a qdiskd[5088]: qdiskd on node 2 reports hung read()
Jul 2 15:32:09 proxmox-a qdiskd[5088]: qdiskd on node 2 reports hung read()So, I guess 2-fold question,
-- does anyone know what is up with the qdiskd messages, is there an easy way to deal with this / or is it a real sign of an actual problem ?
-- I am not sure if the seize-up yesterday was related or not; probably can't assume that it is linked; but - if anyone knows of such behaviour then any comments are very much appreciated.
Thanks,
Tim