| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
 |
 | |||
| Oracle FAQ | Your Portal to the Oracle Knowledge Grid | |
|
| |||
Home -> Community -> Usenet -> c.d.o.misc -> Re: ORA-3113 from svrmgrl
TGGA wrote:
> "Kenneth C Stahl" <ktsahl_at_yahoo.com> wrote in message
> news:3CB36D73.55909EB4_at_yahoo.com...
> > Daniel Morgan wrote:
> >
> > > Is this a new install or an existing install that has been running fine
> > > up until today?
> > >
> > > Does this Oracle installation have a version number or is it anonymous?
> > >
> > > If it is a new installation, which I strongly suspect, did the person
> > > that installed it read the installation instructions on the CD or just
> > > figure they were so smart they didn't need to?
> > >
> > > ORA-03113 is a common error ... almost always the result of not reading
> > > the installation instructions ... or due to an inability to follow them.
> > >
> > > Daniel Morgan
> > >
> > > Kenneth C Stahl wrote:
> > >
> > > > Has anyone ever seen a case where Oracle is not
> > > > running at all on a Sun Server and you log into
> > > > the account that "owns" the Oracle software (I
> > > > always name it "oracle") and try to run svrmgrl
> > > > and get the infamous ORA-3113? I know that the
> > > > ORACLE_HOME and ORACLE_SID are correct and that
> > > > there are no processes running as oracle.
> > > >
> > > > Any guesses?
> > > >
> > > > I've asked the sysadmin to do a reboot on the box
> > > > just in case there is something going on at that
> > > > level.
> >
> > The version is 8.1.6.3. This is not a new install. The install was done in
> > early 2001. I know it was done right because I did it myself, followed the
> > instruction manual meticulously and had been using the instance fairly
> > regularly until the final quarter of last year. This isn't the first
> install
> > I've done on Solaris 8. I've lost count of how many I did.
> >
> > Anyhow, I had the sysadmin do the reboot and the instance started up
> without
> > a hiccup. I'm beginning to suspect that it was a shared memory problem on
> > the OS level, but now that it is working normally again after the reboot I
> > can't say for sure.
> >
> >
> >
> >
>
Actually I had thought about that. I ran ipcs and it only showed me a single shared message segment owned by root and since I wasn't sure why root had a shared memory segment and since I didn't have root privileges anyhow, I figured that a reboot of the system (this is a lab system, so it doens't have any production load) would be a good place to start. Apparently that was the appropriate guess because when the system came back up the startup script in /etc/rc2.d did its thing and now the database is working without a problem.
What I am beginning to suspect (and I'll research this today) is that someone may have logged in as root and started the database. If that is what happened then it would explain several other things that I saw. Received on Wed Apr 10 2002 - 05:43:49 CDT
 |
 |