Christian Iversen wrote:
[EMAIL PROTECTED] wrote:
Hi!
Thank you for your answer.
If you are running a V2 style configuration (crm/pacemaker-based), then
ipfail will not work, use pingd instead. Have a look at:
http://www.linux-ha.org/pingd
Thanks. I remember reading about pingd, but I mis-remembered the name.
Still you should at least have one other connection between the nodes.
This might be serial, though that is not recommended for V2 or ethernet.
Do you use a shared storage or DRBD?
I use DRBD. Why should I have one other connection, btw?
Having only one communication connection in a cluster is a configuration
error. Easily avoidable single point of failure.
My final questions are:
- What's the advantage of running a V2-style configuration?
Resource monitoring, the possibility to define constraints between your
resources, you can have more than 2 nodes, propably more I can't think
of right now and don't find links to right away.
- Are the XML files meant to be human-writable? They're awfully verbose.
Give it a few days :)
Regards
Dominik
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
