[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[hobbit] client on OSF cluster
- To: hobbit (at) hswn.dk
- Subject: [hobbit] client on OSF cluster
- From: David Gore <David.Gore (at) mci.com>
- Date: Mon, 01 Aug 2005 03:22:32 +0000
- User-agent: Mozilla Thunderbird 1.0.6 (Windows/20050716)
The clusters on OSF dgux 4.x work like such:
physical_hostname1 logical_hostname1
physical_hostname2 logical_hostname2
physical_hostname3 logical_hostname3
physical_hostname4 (failover host)
If logical_hostname2 fails, it will failover including the filesystems
to physical_hostname4. So the picture will look like this:
physical_hostname1 logical_hostname1
physical_hostname2 DOWN
physical_hostname3 logical_hostname3
physical_hostname4 logical_hostname2 (failover host)
There are common processes and filesystems for each physical_hostname
that are easily reported. How would I report different processes and
filesystems for the logical_hostnames? Note, physical_hostname1 and
logical_hostname1 do have different IPs.
Not to get into too much detail, but the way I have it setup now, I
check for failover every 10 minutes, and if it has occurred, I copy new
bb-proctabs, bb-dftabs, and bb-exttabs over top the current ones and
restart the BB client. Same, if a fail-back to the home host has
occurred. Perhaps I will have to do something similar with the hobbit
client?
--
David