rpm/up2date broken after upgrade to STABLE 83

[ispro]

We have upgraded servers (RHEL3) to the latest STABLE (83). While up2date processed the update we got error:

groupdel: group rpm does not exist
There was a fatal RPM install error. The message was:
There was a rpm unpack error installing the package: procps-2.0.17-13.7

After this we have found that RPM is broken. We have tried to execute:
find /var/lib/rpm -exec chattr -i {} \;
rm -f /var/lib/rpm/_db*;
rm -f /var/lib/rpm/__db*;
rpm --rebuilddb

But it not helped.

up2date broken with:

Traceback (most recent call last):
File "/usr/sbin/up2date", line 1174, in ?
sys.exit(main() or 0)
File "/usr/sbin/up2date", line 668, in main
up2dateAuth.updateLoginInfo()
File "up2dateAuth.py", line 151, in updateLoginInfo
File "up2dateAuth.py", line 105, in login
File "up2dateAuth.py", line 49, in maybeUpdateVersion
File "/usr/share/rhn/up2date_client/up2dateUtils.py", line 228, in getVersion
release, version = getOSVersionAndRelease()
File "/usr/share/rhn/up2date_client/up2dateUtils.py", line 221, in getOSVersionAndRelease
raise up2dateErrors.RpmError(
up2date_client.up2dateErrors.RpmError: RPM error. The message was:
Could not determine what version of Red Hat Linux you are running.
If you get this error, try running

rpm --rebuilddb


Anyone having these issues? Advices how to fix them?..

[chirpy]

groupdel: group rpm does not exist

Did you check the group file for this and re-add it if it is missing?

Btw, when you rebuild the rpm database you must make sure that no process has it open by killing off anything that comes when running:

lsof | grep /var/lib/rpm

Kill all processes -9 that this shows (or reboot the server if they hang in MUTEX/FUTEX),

Then delete the temporary __db* files,

Then run:

rpm --rebuilddb -v -v