Re: [K12OSN] bad rpm database after restore from backup

[Les Mikesell <les futuresource com>]

On Sun, 2004-01-04 at 12:01, Julius Szelagiewicz wrote:

> 	I tried the rpm copy, but the machines were too far apart in
> levels. have double copies of everything, both old 3.1.x and 4.0.0 and the
> install is running now. who needs weekend with friends if you can sit
> alone in the computer room ;-)

The one with a chance of working would be from a 3.1.2 after upgrade.
It may also involve the /lib/libdb* files.  I think your real problem
is that somehow you have a database file from a libdb-3.x and a version
of rpm linked to libdb-4.x or vice versa, and you need to db_dump,
db_restore with appropriate versions to fix it.  The rpm upgrade
procedure would have taken care of this if done when things were
consistent.
  
> 	problem with dual servers is cost - i am trying to make the
> server perform too many duties. i need to rethink the setup.

It is a problem if you only have one server and need a duplicate.
If you have several, keeping one or two spares is cheaper than
service contracts on the hardware and lets you recover faster from
just about anything.  Spare disks might be easier to buy, although
hardware raid makes it harder to deal with them because you have
to keep complete sets together.

> 	problem with upgrade was that i did too many things to "save time"
> - upgrading the disk array forced me to blow away the old system and
> instead of doing full restore and upgrade i opted for new insrtall and
> copy /home. bad idea in retrospect :-)
> 	we'll see how i get going with 3.1.2 - part of the difficul;ty
> yestrday was that the original 3.1.x system was really 3.1.0 dis-upgraded
> to the gills - there are differences. julius

In retrospect it might have been better to have spent more time trying
to solve the 4.0 problem, perhaps by backing in an older kernel but
it's hard to tell what was really wrong.

---
  Les Mikesell
    les futuresource com