Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: Oracle 8.0 dies on HP-UX 11.0 when /stand/vmunix changes

Re: Oracle 8.0 dies on HP-UX 11.0 when /stand/vmunix changes

From: Liz <lizr_at_geology.com>
Date: 2000/03/16
Message-ID: <MPG.133ad53e8e4f013e989683@news.supernews.com>#1/1

In article <38CA5ACA.F3834960_at_hotmail.com>, assistant_madman_at_hotmail.com says...
> Ulrich Windl wrote:
> >
> > Last Friday we had a reboot of the database server scheduled at 15:00.
> > I began preparing a new kernel about 30 minutes before that
> > time. Unfortunately as soon as the new kernel was put in place, Oracle
> > processes paniced and dumped core like crazy. Maybe the TNS listener
> > stuff is related.
>
> How drastic a change was the kernel rebuild? Did you ensure that the min
> and max shared memory (and other pertinent kernel parameters) were set
> according to the Oracle specs? I have patched and rebuilt the kernel
> numerous times at $PLACE_OF_WORK on both hp-ux 10.20 and 11.00 HP9000
> machines, and every time, the Oracle db restarted and ran as smooth as
> ever.
>

Did you make any changes to the ip stack? I found out the hard way that the oracle checks the system name during startup.

> I'm not the Oracle guru, but the HP-UX one. Anyway: It seems Oracle is
> ridiculously fragile.

I'm both the sysadmin and the DBA. I find Oracle to be far more bullet proof than HPUX. Any operating system that doesn't report hardware errors is not my idea of good.

Liz Received on Thu Mar 16 2000 - 00:00:00 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US