[Clusterusers] power outage

Bassam Kurdali bassam at urchn.org
Fri Aug 9 11:42:01 EDT 2013


Power is flaky today, we've had two outages so far within 20 minutes of
each other..
On Fri, 2013-08-09 at 11:39 -0400, Wm. Josiah Erikson wrote:
> ....and another power outage happened. Grr. We'll see what comes back up 
> this time...
>      -Josiah
> 
> On 8/9/13 10:32 AM, Wm. Josiah Erikson wrote:
> > I'm sending Shawna over.... thanks for the offer!
> > Rack 2 was only down because when the nodes rebuild, they don't bring 
> > the tractor-blade up on first reboot, for a reason I haven't bothered 
> > to figure out yet, because sometimes that's actually kindof nice (and 
> > other times it's totally annoying). If you see nodes up in ganglia and 
> > not in the tractor monitor, you can always become root and execute 
> > "tentakel /etc/init.d/tractor-blade start" and it will bring up 
> > tractor-blade on all the nodes (and just "fail" on the nodes that it's 
> > already running on).... it'll say "remote command timed out" on all of 
> > them, but it actually worked, as you'll see when you go back to the 
> > tractor monitor and see your blades coming up....
> > OK now I really must go pack.... :)
> >     -Josiah
> >
> > On 8/9/13 10:15 AM, Chris Perry wrote:
> >> The tractor monitor says that most of rack two is down. This doesn't 
> >> actually impact me, but I'm on campus - is there anything I can do in 
> >> the server room to help bring them back up?
> >>
> >> - chris
> >>
> >> On Aug 9, 2013, at 10:04 AM, "Wm. Josiah Erikson" 
> >> <wjerikson at hampshire.edu> wrote:
> >>
> >>> I'm not on campus, but I got a lot of text messages about things 
> >>> going down and coming back up (from nagios, our monitoring system), 
> >>> and many (most) of fly's nodes went down... bummer! Just FYI. I 
> >>> can't do anything about it, because I'm not there and won't be for a 
> >>> couple of weeks.... most of them seem to have come back up on their 
> >>> own anyway.
> >>>
> >>> -- 
> >>> -----
> >>> Wm. Josiah Erikson
> >>> Network Engineer
> >>> Hampshire College
> >>> Amherst, MA 01002
> >>>
> >>> _______________________________________________
> >>> Clusterusers mailing list
> >>> Clusterusers at lists.hampshire.edu
> >>> https://lists.hampshire.edu/mailman/listinfo/clusterusers
> >> _______________________________________________
> >> Clusterusers mailing list
> >> Clusterusers at lists.hampshire.edu
> >> https://lists.hampshire.edu/mailman/listinfo/clusterusers
> >
> 




More information about the Clusterusers mailing list