On Sun, Sep 03, 2000 at 09:25:38PM +0200, Piotr K. Isajew wrote:
> On Fri, Sep 01, 2000 at 11:45:21AM -0400, Jan Harkes wrote:
> > [...]
> > The way to repair this is:
> > 
> > - Create a temporary rootvolume,
> >     # createvol_rep repair_root E0000100 /vicepa
> > 
> > - Force a client to start using this volume as it's root,
> >     # killall -9 venus
> >     # umount /coda
> - That was the easiest part, but:
> >     # venus -init -r repair_root
> gives something like this:
> 
> Date: Sun 09/03/2000
> 
> 21:21:24 /usr/coda/LOG setup for size 0x88008
> 21:21:25 /usr/coda/DATA initialized at size 0x220020
> 21:21:25 brain-wiping recoverable store
> 21:21:25 loading recoverable store
> 21:21:26 starting VSGDB scan
> 21:21:26        0 vsg entries in table
> 21:21:26        0 vsg entries on free-list
> 21:21:26 starting VDB scan
> 21:21:26        1 vol entries in table (0 MLEs)
> 21:21:26        0 vol entries on free-list (0 MLEs)
> 21:21:26 starting FSDB scan (833, 20000) (25, 75, 4)
> 21:21:26        0 cache files in table (0 blocks)
> 21:21:26        833 cache files on free-list
> 21:21:27 starting HDB scan
> 21:21:27        0 hdb entries in table
> 21:21:27        0 hdb entries on free-list
> 21:21:27 Initial LRDB allocation
> 21:21:27 Getting Root Volume information...
> 21:21:27 GetRootVolume: can't get volinfo for root volume (reproot)!
> 21:21:42 GetRootVolume: can't get volinfo for root volume (reproot)!
> ....
> 
> Any advice, Jan?
> 
> 
> Piotr
> 
> 

The solution was simple. I had some problems with update client on one
of servers. After fixing, the rest of Jan's advice lead to succesful
solution of the problem.


Piotr.