I knew it couldn''t possibly go right.
First attempt at upgrading a 0.1.0 host trashed the networking config
majorly. In defense of 0.1.1, I did have an odd networking config (since
XCP had problems with the built in Broadcom ports, I pif-forgetted them
and let Linux handle them. If it weren''t for the mutant network config,
I bet 0.1.1 would have done ok.
More seriously, 0.1.1 will not rejoin the pool because all the versions
aren''t up to the same level (I seem to recall that this should work).
Attempting the join with a --force wiped out the NICs under XCP and xe
won''t work without the management NIC being present (XCP knows the
management address but not the physical port). This looks like a reload
to fix.
At this point, it looks like the plan is going to be:
1. shut down all VMs
2. back up the pool database
3. flatten and reload the hosts to 0.1.1
4. restore the pool database and hope to heck everything joins again
Am I having fun yet? :)
Vern
--
Vern Burke
SwiftWater Telecom
http://www.swiftwatertel.com
ISP/CLEC Engineering Services
Data Center Services
Remote Backup Services
_______________________________________________
Xen-users mailing list
Xen-users@lists.xensource.com
http://lists.xensource.com/xen-users
