heartbeat problem with sendmail
hai,
we r trying to make high availability for our mailserver(sendmail).
so we r using heartbeat for that one.till now we r not decided for replicating of data between servers in cluster.let us tell about our current scenario,
now what we made is, we r having two server node1 & node2 with ipaddress x.x.x.1 & x.x.x.2. in both of the server we installed heartbeat.I
In ha.cf file we spacified "UDP etho,node node1 & node node2".
In haresource file "node 1 x.x.x.3 sendmail".
so for this settings how it is working is,
1.when we start the heartbeat in "node 1", along with heartbeat sendmail in getting started.
2.again at the same time when we start the heartbeat in "node 2", heartbeat is getting started and in log file it is showing cluster ip x.x.x.3 is active and node 1 is up. so at that time sendmail is not getting started in "node 2"(ok fine).
3. when we simply stop the service of heartbeat in "node 1", sendmail is also getting stopped.And in node 2 sendmail is automatically getting started.
so what we want to know is whether, if we simply stop the sendmail service alone it should ip take over should occur.
we also seen the mon package, but we cannot able to now more details related to our problem.
so, kindly some one give full details with some sample config files for heartbeat + mon.
advance thanks,
bye
|