On Jan 7, 2008, at 10:32 AM, Thomas Lorenzen wrote:
Hi'
After studying the documentation, I think, that the concepts of
"running-as-master" and "running-as-slave" as introduced in version
2 are, what I need.
Say, that you have two nodes 10.0.0.1 and 10.0.0.2 and that you want
the master node to acquire the (master) 10.0.0.3 address, whereas
the backup node, when up, should acquire the (slave) address
10.0.0.4. If the master node dies, the backup node should of course
drop the 10.0.0.4 address and have the 10.0.0.3 address associated
instead.
This is probably the most simple example of "running-as-master" and
"running-as-slave", but I am really not very familiar with the CIB
XML format, since I have previously only used version 1. I have not
been able to track down an example on this, so a simple recipe to
get me started would be very much appreciated.
master/slave has a different meanings in v2.
what i think you want is something less complex...
* define two IP resources
* set the priority of the .3 resource to (for example) 1000 and to 500
for the .4 resource (this ensures that if only one node is left, that .
3 will "push off" the .4 address)
* define a colocation rule between them and set score=-INFINITY (which
means they cant run on the same node)
if you prefer one node in particular to run the .3 address, use a
rsc_location constraint to indicate this
check out some of the simple examples for something to get hints from:
http://linux-ha.org/v2/Examples/Simple
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
