[NTLUG:Discuss] RedHat 8 and broken rpm

[Tom McDonald]

On 17 Dec 2002 21:15:57 -0600
Ryan Sweat <ryans at cecentertainment.com> wrote:

> I have RedHat 8 installed on 3 computers, and 2 of them (both intel i686) have
> broken rpm implementations.  Almost everytime I attempt to install or remove
> a software package, the rpm command freezes and the database is corrupted.  I have
> noticed by searching the redhat-lists and google that many people are having the
> same problem, but RedHat is dismissing it, saying they cannot reproduce it.
> 
> I just wanted to see if anyone is having the same problem and what, if anything, 
> you did to fix the problem.  The way I get around it now is rm -f /var/lib/rpm/__*, 
> and try again..and again..and again.
> 
> Ryan
> 
> 
> _______________________________________________
> https://ntlug.org/mailman/listinfo/discuss

I've had the same problem since Mandrake 7.0, thru Mandrake 9.0, and it even happened on a RedHat 7.2 system. 

After extensive checking I found that the locking for the rpm database was out of sync.  Prior to a rebuild I had to remove the /var/lib/rpm/__db.001 and /var/lib/rpm/__db.002 files. I also remove these files anytime an error occurs during an rpm operation (i.e. rpmdrake fails or rpmfind fails) and it seems to solve my problems.  

I suspect that the problem lies with the berkley db changes as I used to get some odd error messages about directories not being found.

I'm running Mandrake 9.0 and rpm 4.0.4 glibc 2.2.5 and have all the db1 db2 and db3 installed.

Don't know if this will help but it might give you a starting place to look into it a little closer.

Tom
----

Tom McDonald <tom at compuclaim.com>
Compuclaim Inc.