Node shows offline periodically but all VM works

[wahmed]

One of my cluster node seem to be going offline and coming back on after couple minutes. But all VM connectivity in the node remains unaffected. It turns red in GUI then turns green.Following is the syslog screenshot right after the node showed offline:

What may be causing it to show offline randomly?

 

[sdinet]

hey cool credentials. I would like to know more about your business. There are so many interesting and revolutionary technologies in this platform, its just a matter of time before the business community realizes this.

to answer your question: click on the 'datacenter' container, then click on the 'HA' tab. Check to see if the times match. If they are more than 2 minutes apart, then ssh pipe breaks. let me know if this is the problem because I have a documented way to fix this.

 

[wahmed]

I do not use HA in this cluster. But i did check if the time in the node itself was correct. They all have NTP enabled and nodes are on same time.

Not to toot my own horn by yes there are lot more options in business industry than what people actually realizes. Great product such as Proxmox and others just does not get enough publicity in the IT 'specialist' community.

 

[sdinet]

ok. Another reason I have found my node to randomly restart is because of the 'Watchdog" setting. For example by default I think Watchdog is set to 'reset'. If this is the case, as soon as a VM becomes unresponsive, even if it is not designated as an HA resource, the entire node will restart.