Lewis Kapell wrote:
> Interesting idea about the old log file being deleted. I could
> still see the file listed in the directory, though,
I guess I got confused on that point because the last log file you
listed was for the 4th and all of your posts were on the 5th. I
thought maybe you wer
This is a production server, so I'm not going to stop & restart unless
it's really necessary.
Interesting idea about the old log file being deleted. I could still
see the file listed in the directory, though, and could read/edit it.
Thanks for sharing the ideas.
Lewis
Kevin Grittner wrote:
Lewis Kapell wrote:
> And anyway, there have now been three additional messages written to
> the log file since it was first created.
As long as it continues to log after you start PostgreSQL with the
file already in existence, I guess that's not it. You did try a stop
and a start, and ensur
The filename includes date, but not time, because that is how the
relevant options are set up in the configuration file. I could have
specified that the file name also include hour, minute and second, but
that should make no difference.
And anyway, there have now been three additional message
Lewis Kapell wrote:
> So my problem is solved, although I wish I had some idea what had
> caused it in the first place.
Just a wild shot in the dark -- does it not behave well if the file
already exists? I noticed that your filename includes date but not
time. Perhaps time healed this wound
Using strace, I was able to determine that the logger process was
attempting to output its usual messages when an invalid SQL query was
issued.
As an experiment, I changed the value of log_directory in
postgresql.conf (to a directory which I had created in a different
partition than /var) and
Lewis Kapell writes:
> If I search the output of ps -ax for entries containing both "postgres:"
> and "process" I get the following:
> 2259 ?Ss 0:03 postgres: logger process
Well, that looks reasonable. You might try strace'ing that process
while you do something that's certain t
If I search the output of ps -ax for entries containing both "postgres:"
and "process" I get the following:
2259 ?Ss 0:03 postgres: logger process
2261 ?Ss 0:29 postgres: writer process
2262 ?Ss 0:17 postgres: wal writer process
2263 ?Ss 0:10 p
Lewis Kapell writes:
> 2) The relevant section of the config file looks like this:
> #log_destination = 'stderr'
> logging_collector = on
> log_directory = '/var/log/postgres'
> log_filename = 'postgresql-%Y-%m-%d.log'
> #log_truncate_on_rotation = off
> log_rotation_age = 7d
> #log_rotation_size
Sorry, a couple of things that I intended to mention in my initial post
- I must have forgotten.
1) We (actually I) compiled Postgres from source.
2) The relevant section of the config file looks like this:
#log_destination = 'stderr'
logging_collector = on
log_directory = '/var/log/postgres'
Lewis Kapell writes:
> This is a really strange one. We are running PostgreSQL 8.3.7 on a
> Fedora Core 8 system (uname -a gives: 2.6.23.1-42.fc8).
> On Wednesday I was reading about the upcoming change in the default
> value of default_statistics_target from 10 to 100 - we had never changed
l.org
[mailto:pgsql-admin-ow...@postgresql.org] On Behalf Of Lewis Kapell
Sent: Friday, June 05, 2009 10:44 AM
To: pgsql-admin@postgresql.org
Subject: [ADMIN] Postgres has stopped logging
This is a really strange one. We are running PostgreSQL 8.3.7 on a
Fedora Core 8 system (uname -a gives: 2.6.23.1-4
This is a really strange one. We are running PostgreSQL 8.3.7 on a
Fedora Core 8 system (uname -a gives: 2.6.23.1-42.fc8).
On Wednesday I was reading about the upcoming change in the default
value of default_statistics_target from 10 to 100 - we had never changed
this value, so I bumped it up
13 matches
Mail list logo
