On Thu, 2002-11-28 at 10:11, Luis Sousa wrote:
> >You should use copytruncate as a logrotate option, so that you will not
> >need to force the postmaster to close.
> >
> >This is my (working) setup:
> >
> > $ cat /etc/logrotate.d/postgresql
> > /var/log/postgres.log {
> > daily
> > rotate 10
> > copytruncate
> > delaycompress
> > compress
> > notifempty
> > create 640 postgres postgres
> > }
> >
> >
> >
> Oliver,
>
> I'm using the definition that you refer but now I have a problem.
> Now, after one rotate, postgres created an empty file, postgres.log, but
> is witing log on postgres.log.1.
> Why's that ? Why didn't postgre started writing log in postgres.log, the
> new file that was created ?
The above is my setup and it works, so I don't know why yours wouldn't.
Is there a bug in logrotate? What version are you running? Mine is
3.6.5; copytruncate was added at 2.6, but there is no mention of any
problem with it in the subsequent changelog.
The fact that postgres is writing to postgres.log.1 shows that
copytruncate hasn't worked and a new log has been created, whereas the
old log should have been copied to postgres.log.1 and then truncated in
place.
Try removing the create option (which in fact shouldn't be needed at all
with copytruncate.)
--
Oliver Elphick Oliver.Elphick@lfix.co.uk
Isle of Wight, UK http://www.lfix.co.uk/oliver
GPG: 1024D/3E1D0C1C: CA12 09E0 E8D5 8870 5839 932A 614D 4C34 3E1D 0C1C
========================================
"The earth is the LORD'S, and the fullness thereof; the
world, and they that dwell therein." Psalms 24:1