[BlueOnyx:16807] Re: Server memory fault

[Chris Gebhardt - VIRTBIZ Internet]

Hi Richard,

On 1/4/2015 3:06 PM, Richard Morgan :: Morgan Web wrote:
> Jan  4 20:57:57 s2 kernel: EDAC MC0: UE page 0x0, offset 0x0, grain
> 1073741824, row 0, labels ":": i3200 UE
>
> A bit of research said the EDAC is Error Detection and Correction and
> one article said as it was detected it was corrected so just monitor it
> to see if more faults occur.
>
> However, I’d like an opinion on whether we should run to the data centre
> (which is a few hours drive) or take a leisurely trip before the end of
> the month?  It’s a production server with BX5208R.

I have not encountered this specifically, but my preference if there is 
a RAM issue is to replace it before it bites me in the rear.  That's 
just my preference.   Understanding it's a bit of a trek for you to get 
to your facility, perhaps you can have your provider take a look and 
swap the RAM for you if you send replacement?   That's the sort of thing 
we routinely do for our customers, but I understand that support 
policies vary.  I know of providers that simply will not, under any 
circumstance, open a colocated server.  Others require your first-born 
and hefty fees up-front.  <shrug>  If your colocation provider is 
amenable, it may be a better idea to simply have some RAM sent to them 
rather than a road-trip.

Regarding the suppression of the error, perhaps this will get you 
started along that process:
http://superuser.com/questions/351387/how-to-stop-kernel-messages-from-flooding-my-console

HTH!

-- 
Chris Gebhardt
VIRTBIZ Internet Services
Access, Web Hosting, Colocation, Dedicated
www.virtbiz.com | toll-free (866) 4 VIRTBIZ