[LRUG] Finding out why server was unresponsive

[Andrew Stewart]

On 16 Apr 2013, at 12:28, David Burrows wrote:
> I run pretty much the same setup as you, Hetzner servers running both 12.04 LTS and some old 10.04 LTS and have had the same problem, usually about once per server per year - seized server, nothing in the logs, no core dump, plenty of disk space and no evidence of an issue at all on New Relic. 
> 
> After quite a bit of investigation I just marked it down to a cost of doing business on Hetzner - from the cost I would think they're just using consumer grade hardware that's going to fail from time to time. As the ops people say 'everything fails' so best to get rid of your single points of failure anyway.   

That's interesting to hear.  In this case the server is one of their EX4S's, without ECC RAM.

Perhaps it was a stray cosmic ray...

Cheers,
Andy