On Wed, 17 May 2006, Tom Lane wrote:
Did you try generating a test case using a long sleep() as a replacement
for the archive_command script? If there is a PG bug here it shouldn't
be that hard to expose it in a simple test case. I'm up to my armpits
in other stuff and don't have time to try it myself...
Interesting that you should say that as I had just started setting up a test.
I seem to get alot of these:
May 17 21:34:04 discord postgres[20573]: [4-1] LOG: archived transaction log
file "000000010000000000000001"
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
I'm guessing that the archiver moves the WAL file to <wal filename>.ready, and
successful completion, it renames it to <wal filename>.done? My
archive_command is the utmost in simplicity...looks like this:
----
#!/bin/sh
FULLPATH="$1"
FILENAME="$2"
sleep 300
cat $FULLPATH > /dev/null
----
So far I haven't been able to reproduce, but I'm just getting started.
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?
--
Jeff Frost, Owner <jeff@xxxxxxxxxxxxxxxxxxxxxx>
Frost Consulting, LLC http://www.frostconsultingllc.com/
Phone: 650-780-7908 FAX: 650-649-1954
