I am actually taking care of other people's HA config, in which I see a line in /etc/ha.d/ha.cf: respawn hacluster /usr/lib64/heartbeat/ipfail
I want to know what this line means and how it behave. Do I need to configure ipfail? if so, where? Thanks. Hai Tao Date: Wed, 25 May 2011 17:37:54 -0500 From: [email protected] To: [email protected] Subject: Re: [Linux-HA] how to configure ipfail On 05/25/2011 05:17 PM, Hai Tao wrote: > > I am not using corosync/pacemaker, and I use mon to monitor my application. I want to configure ipfail so I can monitor the external ethernet connectivity by pinging a gateway. > > In case the active node lost its external (which provide application services) network connection, ipfail can trigger a failover, so the peer can continue to provide service. > > can ipfail do this? if so, how？> A simple mon script could ping your gateway and run 'hb_standby all'. It turns out, however, that pinging the gateway doesn't work all that well: you get nodes flipping back and forth on switch resets etc. So I ended up with mon script that's basically open( IN, "ethtool $nic 2>&1 |" ) or die; while( <IN> ) { if( /Link\s+detected:\s+yes$/ ) { exit( 0 ); ... exit( 1 ); i.e. read the output of ethtool and check for "Link detected: yes". Dima -- Dimitri Maziuk Programmer/sysadmin BioMagResBank, UW-Madison -- http://www.bmrb.wisc.edu _______________________________________________ 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
