[tech] Manbo downtime, /services offline

Adrian Chadd adrian at ucc.gu.uwa.edu.au
Thu Jan 17 23:13:00 WST 2008


Just shut manbo down to the bare minimum CPU and IO boards? Or won't
that drop the heat generation by much?


On Thu, Jan 17, 2008, David Adam wrote:
> Probably due to extreme weather conditions, Manbo has been having severe 
> operating difficulty, so it has been shut down for an unknown period of 
> time.
> It spent most of this afternoon trying to reboot but failing due to a ZFS 
> misconfiguration (Adrian, I had to mark one of your shares as 
> unmountable), and once this problem the network interface refused to 
> initialise. Additionally, the disk arrays were reporting multiple 
> problems.
> As UCC is currently locked, we turned Manbo off remotely in order to avoid 
> further damage.
> This has taken /services offline, which means (among other things) no 
> main UCC website, wiki or forums. User web space is still working. Windows 
> machine logins may not work, and /away accesses (including Windows 
> home directories) will certainly not.
> We believe this is due to high temperatures in the machine room caused by 
> the failure of one of the airconditioners (which has been doing funny 
> things for a while). It's still under warranty but getting hold of the 
> manufacturer is proving difficult.
> There are a couple of things we're doing to try and restore service:
>  - the cables for our Fibre Channel disk array have arrived, so we're
>    trying to move the files located on Manbo over to Musundo, the V480
>    which is hosting the FC arrays. Musundo runs much cooler (and faster).
>  - if we can't get the aircon fixed in a reasonable time, we'll buy a new
>    one (last time it took less than six hours from "we need a new aircon"
>    to its installation).
> If you have any problems or questions please reply to the list or contact 
> us directly on wheel at ucc.gu.uwa.edu.au
> Thanks,
> David Adam
> UCC Wheel Member
> zanchey at ucc.gu.uwa.edu.au

More information about the tech mailing list