[Systems] aslo-web on treehouse

[David Farning]

On Mon, Apr 19, 2010 at 5:40 PM, Stefan Unterhauser
<stefan at unterhauser.name> wrote:
> hi all
>
> last time pootle this time aslo-web
> don t what triggers the problem
> http://sunjammer.sugarlabs.org/munin/sugarlabs.org/treehouse.sugarlabs.org.html#Virtual
> machines\
>
> /me needs professional help there :)
> for me the most interesting part is that aslo itself survived
> know that cause i was clicking there around the last 3 days
> (only once i got a broken http answer)

Everything held up.  The load balancer did what it was supposed to do.

The piece I was concerned about last time was that rather than failing
gracefully the network on treehouse hung.  The load balancer kept
sending traffic to the node on treehouse.  This resulted in a spike in
apache processes on the load balancer.  End result was that sunjammer
started to swap because of all the extra processes.

This time, according to the graph at
http://sunjammer.sugarlabs.org/munin/sugarlabs.org/sunjammer.sugarlabs.org-apache_accesses.html
, the Apache processes stayed within the expected level for weekend
traffic.

> rebooted aslo-web
> (dfarning can you double check if now everything is fine?)

Everything looks good. Last time the bring up was problematic because
the network connection on treehouse to sunjammer failed to come back
up.