[tech] morwong

Nick Bannon nick at ucc.gu.uwa.edu.au
Tue Sep 19 14:59:00 WST 2000


On Tue, Sep 19, 2000 at 12:44:08PM +0800, Grahame Bowland wrote:
> On Tue, Sep 19, 2000 at 12:10:39PM +0800, David Manchester wrote:
[...]
> Working fine. It's running reasonably well. Personally I think Nick was 
> over-reacting. Sure, it's had some downtime. This has been true of all our 
> boxes, because Cameron Hall power keeps tripping. It was on the same power 

I'm not [over-]reacting yet. It wasn't just the power, and it's
_not_ just one incident. If it goes down too often, rendering every
non-standalone box useless, _then_ we'll have to react. This is
more a "please try to be careful, guys" than a "it's broken, fix
it now".

This is downtime for any reason mind you - wheel members making
mistakes inclusive, deliberately changing the software or hardware
config inclusive. I can think of lots of times this has happened -
has morwong had a two month uptime yet? Uptimes should have gotten
better as we settled in with the machine, with luck we're OK now.

[...]
> We think that was a fork bomb - probably accidental, and we can avoid that 
> by putting process/memory quotas on. If we make them large they won't be 
> overly restrictive.

Sounds good.

David Manchester wrote:
> Nick Bannon wrote:
> > In the couple of years mola was the NFS server I don't think it went
> > down as many times as morwong has.
>
> That's because it wasn't doing half the stuff morwong can and does do.

That's exactly the point. Uptimes are much better with dedicated
servers simply because administration and changes of any sort are rare.
Hell, people even pretend Windows is stable when they follow this policy.

Nick.

-- 
  Nick Bannon  | "I made this letter longer than usual because
nick at it.net.au | I lack the time to make it shorter." - Pascal




More information about the tech mailing list