Utah c6620 node is unreachable

[Leonid Kondrashov]

Hello,

In my experiment (https://www.cloudlab.us/status.php?uuid=0037bcb8-c8e5-11f0-bc80-e4434b2381fc), I cannot access one of the nodes (er068.utah.cloudlab.us, node-003). Ssh'ing fails with "no route to host" on both control and experiment interfaces. It was working several days ago. The rest of the nodes are reachable without a problem.

Can you take a look at that?

[ajma...@gmail.com]

This node is behaving very oddly.  The iDRAC doesn't seem to think there's anything wrong with it, there's nothing telling in the logs, but the console hangs forever at "no signal" upon power on, and it's not clear it's actually making any progress through POST.  I'll have to check on it in person tomorrow, best case scenario it just needs a hard cycle.

Best,

 - Aleks

[ajma...@gmail.com]

Just as an update, I had business to take care of in the office this morning/early afternoon, and I will be heading to our datacenter soon to look at this.

[ajma...@gmail.com]

The node has, unfortunately, proven unresponsive to my various attempts to revive it.  This is behaving similarly to another server of ours that I recently replaced the motherboard on, so I suspect I will have to do similar here.  This will, unfortunately, likely take longer than the remaining duration of your experiment.  Do you have anything important on er068 that you need to recover before we schedule it out of service?

[Leonid Kondrashov]

Thanks for the update. The node contains nothing important. You can go ahead with the maintenance. Should I delete it from my experiment manually for that?

Regards,

Leonid

[ajma...@gmail.com]

Hi Leonid,

It doesn't really matter, I've scheduled it to go out of service once your experiment expires, so you don't need to explicitly remove it from your experiment right away.

Best,

 - Aleks