Re: does wal archiving block the current client connection?

On Thu, 18 May 2006, Tom Lane wrote:

> Jeff Frost <jeff(at)frostconsultingllc(dot)com> writes:
>> I seem to get alot of these:
>
>> May 17 21:34:04 discord postgres[20573]: [5-1] WARNING: could not rename file
>> "pg_xlog/archive_status/000000010000000000000001.ready" to
>> May 17 21:34:04 discord postgres[20573]: [5-2]
>> "pg_xlog/archive_status/000000010000000000000001.done": No such file or
>> directory
>
> That seems odd ...

Further interesting items:

May 17 22:41:18 discord postgres[23817]: [3-1] LOG: archive command
"/usr/local/pgsql-8.1.3/bin/archive_test.sh "pg_xlog/00000001000000000000000E"
"00000001000000000000000E""
May 17 22:41:18 discord postgres[23817]: [3-2] failed: return code 256

And in the window where I started postgres via pg_ctl, I had this:

cat: pg_xlog/00000001000000000000000E: No such file or directory
cat: pg_xlog/00000001000000000000000E: No such file or directory

Seems bad.

>
>> Currently I'm using pgbench to generate WAL rollover, do you guys have any
>> other handy testing tools for this sort of job or is this the best tool?
>
> pgbench seems like an OK load for this, although it doesn't start/end
> any new connections while running. I *think* that that's not critical
> --- my guess is that your observation of new connections hanging is just
> because each new connection has to run one startup transaction, and
> transactions in general are at risk of hanging --- but that could be
> wrong.
>
> Also, you could increase the rate of WAL generation by decreasing the
> checkpoint segments/timeout parameters, if you need to.
>
> regards, tom lane
>
>

--
Jeff Frost, Owner <jeff(at)frostconsultingllc(dot)com>
Frost Consulting, LLC http://www.frostconsultingllc.com/
Phone: 650-780-7908 FAX: 650-649-1954