[rt-users] postgres slowdowns? (pg 7.3.4, rt 3.0.4)

[Kevin Murphy]

Vivek,

Thanks.  Did I mention that a restart of postgres temporarily solves 
the problem?  (Actually, I also restarted Apache at the same time, so I 
can't really say which is relevant.)

> > are you by chance running a vacuum on your DB at the same time?

No.   However, ahem, I hadn't actually, er, vacuumed this database 
before.   I just did it, and I will make this a cron job.  Note, 
however, there was no perceptible improvement in performance after the 
'vacuumdb --analyze rt3'.  Probably because my database is still so 
small that full table scans are still practical (and the machine has 
1.5 GB RAM, some of which the OS might use for disk caching?  The 
machine is doing other things, but not very often, at the moment -- and 
not that I have seen during the slowdowns).

> > have you tuned your PG beyond the defaults for amount of shared 
> memory
> > and sort memory?

No, and I will do this, but I find it hard to believe that this will 
have much of an effect for such a small, low-use database.  The effect 
is quite extreme.  Nevertheless, I will read up about tuning in the PG 
docs.

> > what is the concurrency of use at the slow times?

Very low, usually one.   At any given time, there are around 3 users 
using RT sporadically (once every few minutes or even hours).

-Kevin