Thanks to everyone who replied!
The basic answer is you have to do the rolling upgrade. However it is
better (and considerably quicker) to patch the member 0 disk before the
cluster is built.
The problem I had is that the member0 disk had been reused as swap, so
I was out of luck there!
Thanks especially to Jim Fitzmaurice for pointing out that "cfgmgr" must
be in the /etc/inetd.conf file for rolling upgrades to work!
Regards,
Jonathan
!---------------------------------------------------------------------------
= Jonathan Nicholson - Team Leader : System Support "Special Projects" =
= The Sanger Centre, Wellcome Trust Genome Campus, Hinxton, Cambs, CB10 1SA =
= Email: jjn_at_sanger.ac.uk -=- Tel: 01223 834244 x4987 -=- Fax: 01223 494919 =
----------------------------------------------------------------------------
Received on Thu May 10 2001 - 13:27:26 NZST