thanks for the help

Lynn Newton lynn.newton at gmail.com
Wed Apr 4 10:43:42 MST 2007


  On 4/4/07, Micah DesJardins <micahdj at gmail.com> wrote:

  > I'm glad you're back up and running. Any idea what the problem
  > was? I'm curious.

The problem is by no means solved completely, but I'm
back up.

The internal SATA drive worked the first time it was
accessed from Knoppix. Then it behaved intermittently.
Jiggling the cable sometimes worked, but not reliably,
and may have had nothing to do with it.

The USB on the motherboard was working fine. I don't
know why I was unable to use it all this time. It works
now, and I also have my new PCI USB card in, and it
sees all eight ports. (That should cover the two
devices I have to put on it!) We put stuff on both the
internal and external USB and it sees them fine.

I have not been able to do a complete backup since
September 12th, nearest I can tell.

There are other problems. Unfortunately, I installed
both my main system and my external drive with reiser
file systems, not knowing how prone to bugs these were,
being of the misinformed impression that reiserfs was
state of the art. Big mistake. *Very* big mistake.

I *still* have not been able to complete a backup, and
have had to reboot three times since last night. The
first and third time I got syslog messages popping up
that were thoroughly laced with the word "reiser", and
the drives hung.

The second time I tried to umount my automounting
external drive, but it couldn't because of resources
that owned it. I tried logging off completely, logging
onto the console as root, and found there were still
two processes alive that had my loginname attached to
them. When I tried to kill them, the system paniced,
and the dump that appeared also said "reiser" in
numerous lines.

This morning I tried a backup again, figuring that
eventually it should get to the end since I'm using
cpio -pdm, which will skip files that are already out
there. But again I got the pop-up syslog messages. Then
made the mistake of running sync, which has been
running since about 5:30am. I don't want to kill it or
any parent process right now because I'm sure it will
trigger another panic.

Why this situation has suddenly reared its head after
nearly two years of flawless operation I don't know,
but I tend to think it's got to be hardware, which
wears out, because I don't monkey much (at all) with
the system software.

So I'm not out of the water yet. The ideal solution is
going to be to rebuild my system softwarewise and move
quickly away from using reiserfs, which will require me
to get at least one more disk drive.

In addition, I discovered a case fan has gone bad. (Not
the one that surrounds the CPU.) Gotta fix that before
hot weather comes for sure. It gets close to 90 degrees
in my office in the afternoons, and there's not a thing
I can do about it. (Except take my new laptop and work
in the living room where it's at least 10 degrees
cooler.)

As if I've got nothing but time to rebuild systems. I
know that for some subscribers on this list that's what
they do and love. I did it myself for decades, when I
got paid to crash systms daily, but I'm over it. I love
Linux as a work environment, and just want stable
systems that work and that I don't have to rebuild very
often.

I'm presently planning on rebuilding on SuSE 10.2. I
know, someone will say that I should use Ubuntu because
it's Debian based and therefore a lot easier to keep up
to date. But I've never actually seen it, and I've
really *loved* SuSE 9.2 the last two years.

I'll probably run Ubuntu under VMware on my Macbook
Pro.

You asked, now you know.

-- 
Lynn David Newton
Phoenix, Arizona


More information about the PLUG-discuss mailing list