From: | Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> |
---|---|
To: | Rajesh Kumar Mallah <mallah(at)trade-india(dot)com> |
Cc: | "PGSQL List (E-mail)" <pgsql-admin(at)postgresql(dot)org> |
Subject: | Re: fishing out LOG: duration lines from the logfile. |
Date: | 2003-10-31 21:24:54 |
Message-ID: | 200310312124.h9VLOs127219@candle.pha.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-admin |
Rajesh Kumar Mallah wrote:
>
> Hi,
>
> I think its not going to a trivial task to take out only the LOG duration
> lines from a PostgreSQL logfile. We need to extract the duration and
> the actual statement. I think i will put a custom delimeters around the
> statements for the time being so that the log parser can parse it
> unambigiously. Is there any better method?
Seeing that I was involved in implementing this behavior, I felt I
should write a script to pull out this information to see how hard it
would be. Here is an awk script:
awk ' BEGIN {in_statement = "N"}
{
while (getline > 0)
{
if ($1 == "LOG:" && $2 == "duration:" && $5 == "statement:")
{
print $0;
in_statement = "Y";
}
else if (in_statement == "Y" && $0 ~ /^ /) # <-- tab
{
print $0;
}
else
in_statement = "N";
}
}' "$@"
I tested this with the log file you included and it seeme to work fine,
though the email had some line wraps I had to remove.
--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2003-10-31 21:27:31 | Re: SELECT COUNT(*)... returns 0 ROWS |
Previous Message | Tom Lane | 2003-10-31 21:13:17 | Re: ? in explain query |