It is enough to start Heartbeat at boot time (if its right configured.)
Then heartbeat is the one, who starts sshd and httpd after itself is started on 
the primary node (the preferred one is the one which you declared with the 
first in /etc/ha.d/haresources)


cat /etc/ha.d/haresources
server1 sshd httpd


server1 (primary)
  /etc/init.d/heartbeat (start at boot)
  /etc/init.d/sshd (started by heartbeat at boot)
  /etc/init.d/httpd (started by heartbeat at boot)

server2 (secondary)
  /etc/init.d/heartbeat (start at boot)
  /etc/init.d/sshd (started by heartbeat, when server1 is down)
  /etc/init.d/httpd (started by heartbeat, when server1 is down)

To you power-question:

When both nodes are affected by the power outage and you'll start them your 
resources will run on server1, again.
When always one server is running you can control the switching of the 
resources (when the other member joines the cluster) with the auto_fallback 
option in /etc/ha.d/ha.cf.


Please correct someone, if something is wrong.

Regards

Florian

-----Ursprüngliche Nachricht-----
Von: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Im Auftrag von Doug Lochart
Gesendet: Mittwoch, 13. Februar 2008 18:19
An: General Linux-HA mailing list
Betreff: [Linux-HA] resource script question (runlevel config)

I have read a few tutorials now.  Some mention turning off highly
available services and some do not mention it.  I am making 'sshd' and
'httpd' highly available on my ha cluster.  Once I have everything
configured one tutorial said to make sure that your services AND
heartbeat are configured NOT to start at boot time on both servers.  I
understand this during the testing aspect but when this is ready to go
live I believe this is the way it should be:

server1 (primary)
  /etc/init.d/heartbeat (start at boot)
  /etc/init.d/sshd (does not start at boot ?? )
  /etc/init.d/httpd ( does not start at boot ??)

server2 (secondary)
  /etc/init.d/heartbeat (start at boot)
  /etc/init.d/sshd ( does not start at boot )
  /etc/init.d/httpd ( does not start at boot )

Could someone please look at my above assumptions and correct me if I am wrong.

I also assume that if server1 went down and server 2 became primary
that is a building wide power outage occurred that when power was
restored that server2 would still know it was primary.

thanks

Doug
-- 
What profits a man if he gains the whole world yet loses his soul?
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems
_______________________________________________
Linux-HA mailing list
[email protected]
http://lists.linux-ha.org/mailman/listinfo/linux-ha
See also: http://linux-ha.org/ReportingProblems

Reply via email to