[Linux-HA] Best effort HA

[Kai Bjørnstad]

Hi all

Back in action after doing some other stuff....

I went back to basics this time, trying out the 2.0.9 version from SuSe,=20
http://software.opensuse.org/download/server:/ha-clustering/SLES_9/i586/

And I am getting even more confused:

I set up a configuration with one IP and three LSB-services. All the=20
LSB-services are set to be colocated with the IP.

Now I list them like this:
lsb_dhcpd       (lsb:dhcpd):    Started dl360g3-2
lsb_conserver   (lsb:conserver):        Started dl360g3-2
lsb_scaproxyd   (lsb:scaproxyd):        Started dl360g3-2
ip_172.19.5.200 (heartbeat::ocf:IPaddr):        Started dl360g3-2

Now, if I corrupt the lsb_conserver service, the resource just gets status=
=20
failed and nothing happens (no failover, nothing).

I then restart everything (including heartbeat) and get everything back up=
=20
again. Then i corrupt the lsb_dhcpd service. Now suddenly the services fail=
=20
over to my backup node.

The rules set for the two lsb services are exactly the same. The only=20
difference is the order they are listed in the CIB. This can't be correct??

I also tried to corrupt the lsb_dhcpd service on my backup node (after a=20
failover). This leads to all my services (and the IP) going down.....


I am starting to think that my best-effort configuration just isn't=20
possible ??

Does anyone have an best-effort example configuration?

What version of Heartbeat should I use?

Cheers=20
Kai

=2D-=20
Kai R. Bj=F8rnstad
Senior Software Engineer
dir. +47 22 62 89 43
mob. +47 99 57 79 11
tel. +47 22 62 89 50
fax. +47 22 62 89 51
kai.b at scali.com

Olaf Helsets vei 6
N0621 Oslo, Norway

Scali - www.scali.com
Scaling the Linux Datacenter