RH 7.3 was the last version I ran, but RH8 is being treated like Windows ME. So whatever it is, it must be pretty serious.
I have a server under my desk right now, so I know how you feel. I think it MIGHT cause the problem, but really, at midnight, what would be moving that board around inside the case? I mean if it was being brushed by security as they walked by or something, then it would probably happen during the day, or that would be my thought. From your description I'd actually point at RAM. If Ramtest crashed during the test, there's an issue. It may not turn out to be the RAM, but RAM is cheap, by the description of the server, you could probably swap it's RAM with the stuff from your desktop PC. Less RAM that works 100% would be far better than lots of RAM that doesn't work, so personally, that's the route I'd start on now. It's easy and fast, and I'm all over those types of work-arounds. Plus, it can immediately narrow down the solution. Power surges/spikes will wreak RAM far easier than they wreak anything else, and I think you mentioned a power problem a few weeks back. Personally, I'd also invest in a UPS. Even if it's just a reasonably cheap one, servers should never see utility power (pot calling kettle black here, ours isn't hooked up yet either...). Even a small $100-$150 UPS will clean up the power alot. I still semi-suspect that the thing that happens late at night is automated tasks. If it isn't hardware, that WILL be the issue, I'd bet on it. It's just a matter of which task. That would be easier to narrow down if you knew the time it was crashing. I'd use Servers Alive or a cronjob to monitor that. Kev. ----- Original Message ----- From: "Evan Brown" <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Thursday, November 06, 2003 9:24 AM Subject: Re: (clug-talk) kernel panic, possible answer? > Thanx Kev, that makes me feel really good :P. And if by server room you > mean the computer that sites 2 feet from where I work, then yeah the > server room is a mess :). We are a small(6 people) company so we don't > have all those fancy dancy la dee da server rooms. Now the question I > asked was, could this problem be causeing my kernel panic? OH and what was > broke in RH8? > > Evan > > On Thu, 6 Nov 2003 09:10:49 -0700, Kevin Anderson > <[EMAIL PROTECTED]> wrote: > > > WooHoo! > > > > A server room worse than mine... > > > > Kev. > > > > > > > > ----- Original Message ----- > > From: "Evan Brown" <[EMAIL PROTECTED]> > > To: <[EMAIL PROTECTED]> > > Sent: Thursday, November 06, 2003 8:48 AM > > Subject: (clug-talk) kernel panic, possible answer? > > > > > >> Hey guys > >> > >> I went into the server last night with the intention of switching the > >> socket that the ram was in, just to see if that might help, and then rum > >> memtest86 on it. To my suprise when I touched the mb, it swing back a > >> bit. > >> It seems the pegs that hold the mb off of the case wall had sheared thru > >> and the mb was only really hanging from the very last top corner peg. > >> Could this have caused some of the problems I seem to have been facing? > >> I > >> figure that if the mb had touched the side of the case it could have > >> killed/messed up/crashed the system. What do you guys think? Oh and I > >> left > >> memtest86 to run over night and it seems to have died sometime in the > >> night leaving just a blue screen with a red strip across the top, weird. > >> > >> > >> Evan Brown > >> > >> > >> > > > > > > >
