Re: Load on udd.d.o
On 23/10/11 at 00:44 +0200, gregor herrmann wrote:
> On Sat, 22 Oct 2011 23:38:04 +0200, Tollef Fog Heen wrote:
> > ]] Lucas Nussbaum
> > | In the meantime, DSA, could you reboot samosa (easy option) or restart
> > | apache and postgres and clean up the remaining processes (if any)?
> > I've killed a bunch of postgres processes, I think apache should recover
> > by itself.
> Now bugs.cgi indeed doesn't abort anymore; but it timeouts later
> (which was the same during the better part of the last week).
> Seems like the machine needs a harder kick ...
I've done a vacuum full, and removed the data about etch and lenny from
the tables. This seems to have significantly improved the situation. I
should probably cron a regular vacuum full of all tables.
But the main problem, anyway, is that samosa is a bit low on RAM (only 6
GB), and a bit slow on I/O. In the long term, it would be useful to move
UDD to a faster box...