From: | Hannu Krosing <hannu(at)tm(dot)ee> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: timeout for "idle in transaction" |
Date: | 2001-10-25 14:13:58 |
Message-ID: | 3BD81E26.5299B515@tm.ee |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Tom Lane wrote:
>
> Hannu Krosing <hannu(at)tm(dot)ee> writes:
> > Is there a way to automatically ROLLBACK transactions that are
> > in "idle in transaction" for too long ?
>
> Nope, we don't have anything for that. Not clear to me that it's
> appropriate as a server-side function anyway.
This can't be done from the client side and we do have other types of
deadlock detection on server side so this seems to quite appropriate
from where I stand.
I guess it would be quite nice to have as a connection-level setting,
so that things that benefit from it can set it to some reasonable
value while others that want to behave unsocially can do it as well ;)
The default could be 1-3 sec of idle time in transaction for typical
client-server and web apps while command line clients (like psql) could
set it to something more automatically.
-----------------
Hannu
From | Date | Subject | |
---|---|---|---|
Next Message | Zeugswetter Andreas SB SD | 2001-10-25 14:24:25 | Re: Index of a table is not used (in any case) |
Previous Message | Tom Lane | 2001-10-25 14:13:29 | Re: Can't cast bigint to smallint? |