From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | "Andrus" <eetasoft(at)online(dot)ee> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: How to release locks |
Date: | 2006-03-22 17:46:49 |
Message-ID: | 17923.1143049609@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
"Andrus" <eetasoft(at)online(dot)ee> writes:
>> What's it doing exactly?
> odbc driver was in progress of sending data using libpq when client was
> terminated.
Do you mean that the statement hadn't been fully transmitted yet?
If so, the backend would have just been waiting for the rest of the
statement to arrive. Perhaps you're looking at some sort of network bug
that caused the backend not to be told that the connection was lost.
> Can it happen that process running 2 mb bytea insert command will take
> extremely long time to interrupt ?
> Or will decode() function take long time to interrupt ?
> There is also before insert or update trigger, maybe this influences ?
You seem to have a lot of variables there --- maybe you should do some
experimenting to narrow down what the problem is. I'd expect all of
those things to take some time, but not more than a few seconds...
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | SCassidy | 2006-03-22 17:59:01 | Re: index for inet and >> (contains) function |
Previous Message | Tom Lane | 2006-03-22 17:21:03 | Re: db sever seems to be dropping connections |