Hello, If you dropped ipvsadm & ldirectord, you can make a failover with the following :
heartbeat monit (or any other monitoring tool) Check with monit (http://www.tildeslash.com/monit/doc/) your LDAP status. If you failed X times within a cycle, use monit to kill heartbeat, which will automatically failover to backup mode. You have a serie of example of configuration for monit in the man or the web site. On Thu, May 1, 2008 at 7:57 AM, Bruce Thompson <[EMAIL PROTECTED]> wrote: > Hi, > > We have done as Joseph Mack NA3T suggested, dropping ldirectord and ipvsadm, > as we only require Heartbeat to perform failover in the event our primary > (from Heartbeats perspective) LDAP fails. > > We are running two SUN DSEE 6.3 on separate VM hosts set up with > multi-master configuration. Each instance of the DS is registered with the > "partner" one, and each instance is set as a master in replication topology. > > However, we have been unable to find a configuration that will make > Heartbeat invoke a script in resource.d to determine the status of the > service. > > We have trawled the web for help but there seems to be a paucity of > information relating to SUN DSEE 6.x and Heartbeat. I'm sure there must be > others doing what we are trying to do. > > Below are our Heartbeat configs: > > b2b04util02:/etc/ha.d # cat ha.cf |grep -v ^# |grep -v ^$ > > logfacility local0 > keepalive 2 > ucast eth1 10.32.24.46 > ucast eth1 10.32.24.47 > auto_failback off > node b2b04util01 > node b2b04util02 > ping 10.32.24.46 > ping 10.32.24.47 > respawn hacluster /usr/lib/heartbeat/ipfail > apiauth ipfail gid=haclient uid=hacluster > > b2b04util02:/etc/ha.d # cat haresources |grep -v ^# |grep -v ^$ > > b2b04util01 \ > IPaddr::10.32.24.48 > > b2b04util01 10.32.24.48 hacocoa > > Any suggestions/advice would be greatly appreciated! > > Cheers, > Bruce > > > 2008/4/30 Joseph Mack NA3T <[EMAIL PROTECTED]>: > > > > > On Tue, 29 Apr 2008, Bruce Thompson wrote: > > > > > Hi, > > > > > > We have two directors which are also the real servers which host two > > > instances of LDAP (is it a problem for LVS to route to a local server - > > our > > > ipvsadm output below looks a little strange with "route" and "local"?). > > > > no > > > > > We have heartbeat, ldirectord and ipvsadm configured. We want SUN > > eXchange > > > requests coming in to be redirected to the secondary LDAP server in the > > > event that the primary LDAP fails. > > > > you are talking about failing over the LDAP servers, not > > load balancing them, in which case you should be using > > heartbeat and not LVS to determine the active LDAP server. > > > > Joe > > -- > > Joseph Mack NA3T EME(B,D), FM05lw North Carolina > > jmack (at) wm7d (dot) net - azimuthal equidistant map > > generator at http://www.wm7d.net/azproj.shtml > > Homepage http://www.austintek.com/ It's GNU/Linux! > > > > _______________________________________________ > > LinuxVirtualServer.org mailing list - [email protected] > > Send requests to [EMAIL PROTECTED] > > or go to http://lists.graemef.net/mailman/listinfo/lvs-users > > > > > > > -- > m: +64 (0)21 962 085 > h: +64 (0)4 384 8740 > e: [EMAIL PROTECTED] > skype: bjthompson > _______________________________________________ > > > LinuxVirtualServer.org mailing list - [email protected] > Send requests to [EMAIL PROTECTED] > or go to http://lists.graemef.net/mailman/listinfo/lvs-users > -- Pierrick Grasland ------------------------------------------ [EMAIL PROTECTED] ENSSAT - LSI 3 22300 Lannion FRANCE ----------------------------------------- [EMAIL PROTECTED] NEXCOM Systems 22300 Lannion FRANCE ----------------------------------------- _______________________________________________ LinuxVirtualServer.org mailing list - [email protected] Send requests to [EMAIL PROTECTED] or go to http://lists.graemef.net/mailman/listinfo/lvs-users
