We are running Solaris 8 (we can't upgrade right now) with vxvm 4.0 MP1.
We can't upgrade to MP2, which I think would solve our problem, due to the
that this machine is part of a four-node VCS 4.0 Oracle RAC cluster and we
won't have a
maintenance window for all four machines any time soon.

Here is our issue, any help would be welcome.

We've had to re-install one of the nodes in the cluster. After we vxinstall
try to encapsulate the boot disk, all looks well until we reboot (which
happen twice).

When the machine comes back up the first time we get an error message that
can't be found. I know that there is a link in /dev/vx/[r]dsk bootdg ->
rootdg; which apparently
isn't being created and is a known issue fixed by 4.0 MP2.

The trouble is even when we manually reboot the second time the boot-drive
file systems
become totally corrupted and no amount of fscks seem to have any effect
(it's too far gone
at this point).

We are running off a Fujitsu 1500 xpar partition. So to get back in we have
to boot off the network.
Unfortunately by this time the boot drive is too hosed to work with so we
have to re-jump the

Should we re-install and then check to make sure that the links are there?
Is this the only
known issue?

I just joined the list and I've searched the lists for a similar problem.
If this has been discussed before
my apologies.

Thanks in advance,
Stephen Palmer

Veritas-vx maillist  -  Veritas-vx@mailman.eng.auburn.edu

Reply via email to