Re: Sure-fire "kill"

"Tanel Poder" <tanel@@peldik.com> wrote in message news:<3f147ee2_1_at_news.estpak.ee>...
> There is no init.ora parameters for getting rid of killed sessions AFAIK,
> but there is sqlnet.ora parameter SQLNET.EXPIRE_TIME which you set in
> seconds. You set it on server side and it affects how often server sends
> probes to check whether clients are still alive. If set to 0 (default) no
> probes are sent. I have usually set it to 12-24 hours or even more, because
> had problems w. version 8.0.4.0.0 on WinNT where long running jobs such
> analyze were terminated when exceeded expiry time, IIRC.
>
> Don't set it too low, otherwise you might get lot's of extra network traffic
> and load on your system..
>
> And, I just noticed, don't crosspost to all groups, cdo.server is enough..
> Tanel.
>
> "Fred" <noway_at_jose.com> wrote in message
> news:noway-0F6052.14454715072003_at_vienna7.his.com...
> > There is a setting in INIT.ORA that has the unintended side-effect of
> > making sure the ALTER SYSTEM KILL SESSION command has immediate affect.
> > Without this setting, I've seen some instances where the session reports
> > as being KILLED in V$SESSION but is not physically removed until the
> > instance is bounced. Does anyone remember this value offhand?

Tanel,

I thought that this parameter failed to work on win32 due to only 16 timers being alloted to a process - and the oracle.exe process could certainly support hundreds of threads. Did you ever actually have a dedicated server process (thread under oracle.exe) killed via SQLNET.EXPIRE_TIME? If so, this was news to me. I recall reading notes on Metalink regarding this subject indicating that it was not supported on win32.

Pd Received on Tue Jul 15 2003 - 23:30:15 CDT