Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: Broken pipes with a "failed to start a dedicated server process"
I have gotten the 12500 when there was insufficient memory for another process or the max processes for the id was reached. (The max processes in this case is an OS kernal parameter.)
In article <902up7$okb$1_at_ctb-nnrp2.saix.net>,
"Billy Verreynne" <vslabs_at_onwe.co.za> wrote:
> Running tg4ifmx (transparent gateway for Informix) on a Sun box. I
have
> checked, double-checked and triple checked the initsid.ora file, the
> listener.ora file, the .odbc.ini file... can not see any errors.
>
> The problem: the infamous ORA-12500 error.
>
> The sequence of events that I think is happening.
>
> The listener receives a SQL*Net connection for the gateway sid.
>
> The listener fires up the gateway shell script. This shell script in
turns
> set some environment variables before EXEC'ing the Oracle gateway
software.
>
> The EXEC shell command does not spawn a new process. The software
that is
> EXEC'ed, is executed in the current process space.. the pid thus
remains the
> same.
>
> Set trace level to 16 in the listener. What seems to be happening is
that
> the listener successfully launch the shell script. The shell script
execs
> the gateway software. The listener now attempts to talk to the gateway
> software using IPC. Unix error 32 occurs. "Can not read from a broken
pipe".
>
> The listener returns an ORA-12500.
>
> I thought maybe the shell is to blame. The EXEC command seems to be
handled
> slightly differently between the KSH, SH and CSH. I always use ksh
(he sez
> hanging his shell flag ;-).. Anyway, same problem when using the
standard
> /bin/sh.
>
> Note that the shell script is executed. Thus no problems with file
> permissions or paths or stuff like that. The STDOUT and STDERR of the
Oracle
> gateway software is nothing - nada. I'm redirecting the output to
files and
> these come up empty. No clues there.. (ok, assuming that the fact
that there
> are no STDOUT & STDERR output is not a clue in itself Mr Holmes.. :-)
>
> Any comments, ideas, suggestions?
>
> Thanks.
>
> regards,
> Billy
> --
> Earth first! We'll stripmine the other planets later.
>
>
-- Joseph R.P. Maloney, CCP,CSP,CDP MPiR, Inc. 502-451-7404 some witty phrase goes here, I think. Sent via Deja.com http://www.deja.com/ Before you buy.Received on Wed Nov 29 2000 - 10:28:39 CST