I was (note, operative word) running Stock RH6.1 with the latest upgrades
(via RPM... you saw what it didn't install) on Linus. Unfortunately, it
died horribly sometime yesterday :( I guess I just needed to slick it after
all. I'm not going to trust anything important to this system, either, as
I was sent some email suggesting that I do a manual e2fsck, it DID repair
all my problems, but the problems came back up within 36 hours... which
suggests to me that I have a hardware failure on the Filesystem pending
(I already lost one disk in this machine this year... New Year's day,
my contribution to the Millenium Bug... got a Head Lock on the drive... :(
Looks like I may be doing reinstalls on this machine A LOT. :( And the
graphical install keeps dying on it for the 6.1 installs. Text works
fine, though, as always in the Unix world! :)
But the suggestion to force it through will help on another machine. :)
Bill Ward
-----Original Message-----
From: Gordon Messmer [mailto:[EMAIL PROTECTED]]
Sent: Saturday, March 04, 2000 2:43 PM
To: [EMAIL PROTECTED]
Cc: recipient.list.not.shown; @nswcphdn.navy.mil
Subject: Re: RPMs - Frustration sets in.
Ward William E PHDN wrote:
> I've tried using AutoRPM to update the installed base on system to the
> latest patch levels, and I get the following:
> ---------------------------------------------------------
> ** Failed RPMs **
.....
> ORBit-0.5.0-2 is missing dependencies:
> ORBit = 0.4.95
It looks like AutoRPM is confused a bit. When trying to upgrade ORBit
to version 0.5.0, it's not ORBit-0.5.0 that needs 0.4.95, but
ORBit-devel-0.4.95. Why it's doing that, I don't know. I thought that
I'd heard the latest version of autorpm corrects some such problems.
What version are you running?
In any case, you don't need to hunt down the old versions of any of
these libraries, just switch to the directory where autorpm is storing
all of these updated rpms, and :
rpm -Fvh *rpm
Then, remove any rpms that successfully updated.
> Unfortunately, now I'm getting a new error (which is popping up
> both ways, with /var and //var)
>
> error creating directory /var/lib/rpmrebuilddb.6398: No space left on
device
>
> Ok, now THAT's patently ridiculous.
Maybe so, but it's possible that you have a filesystem problem rather
than an rpm problem.
As root, try:
telinit 1
#Wait for single user mode.
mount / -o remount,ro
fsck -f /dev/hda1
#If there are any errors, fix them.
mount / -o remount,rw
telinit 3
After you've check the filesystem, check the permissions on rpm's
databases:
ls -l /var/lib/rpm/
> I >REALLY< don't want to have to slick this system... I won't lose much
> except time, but it's a pride thing... anything goes wrong in Windows,
> the only alternative is to reinstall fresh. I've told a LOT of folks
that's
> not true in Unix... and pushed them towards Linux. Now I'm finding myself
> stymied on this, and on the verge of slicking it and starting over...
> Like I said in the title to the original message, "Frustration sets in".
There was a bug in some 2.2 kernels where filesystems were being
corrupted in the block device layer of the kernel (I believe).
What kernel are you running?
MSG
--
To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
as the Subject.
--
To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
as the Subject.
