[Linux-HA] Configuring HA

[Dan Track]

On 10/10/06, Andrew Beekhof <beekhof at gmail.com> wrote:
> On 10/10/06, Dan Track <dan.track at gmail.com> wrote:
> > Hi
> > I'm trying to setup heartbeat 2.0.7 on rhel 4 es and am having some problems.
> >
> > I'm following the instructions on this page:
> >
> > http://www.linux-ha.org/GettingStarted/OneIPAddress
> >
> > I'm having the following problems
> >
> > 1) Both nodes come up with the virtual floating IP once the primary
> > node is restarted. Only one should come up that gets failed over to
> > the other node. ( at least this is how I understand it)
>
> Sounds like the nodes cant communicate.  Firewall?
>
> And one of the nodes probably needs to change this line to something else
>    ucast eth1 192.168.16.2
>
> They cant both have that ip address :)
>
> > 2) Running the following generated the following error:
> >
> >  python /usr/lib/heartbeat/haresources2cib.py /root/haresources.temp >
> > /var/lib/heartbeat/crm/cib.xml
> > cib.xml or cib.xml.sig exist in /var/lib/heartbeat/crm/
> > please remove them before generating new cib.xml
> >
> > Everytime you remove the file the script recreates a blank file and
> > then complains, buit never populates the file.
> >
> > Could someone please help me with the above?
>
> Not unless you include the contents of  /root/haresources.temp
>
> Also, you cant use ipfail with a crm-based cluster
>
> > My configuration files are
> >
> > node1: ha.cf
> >
> > logfile /var/log/ha-log
> > logfacility     daemon
> > keepalive 1
> > deadtime 10
> > warntime 5
> > initdead 120 # depend on your hardware
> > udpport 694
> > ping 10.11.8.40
> > bcast eth1
> > ucast eth1 192.168.16.2
> > auto_failback off
> > node    zhang
> > node    quan
> > respawn hacluster /usr/lib/heartbeat/ipfail
> > use_logd yes
> >
> > node2: ha.cf
> > logfile /var/log/ha-log
> > logfacility     daemon
> > keepalive 1
> > deadtime 10
> > warntime 5
> > initdead 120 # depend on your hardware
> > udpport 694
> > ping 10.11.8.40
> > bcast eth1
> > ucast eth1 192.168.16.2
> > auto_failback off
> > node    zhang
> > node    quan
> > respawn hacluster /usr/lib/heartbeat/ipfail
> > use_logd yes
> >
> >
> > If you need anymore info please let me know.
> >
> > Thanks in advance
> > Dan
> > _______________________________________________

Hi Andrew,

I made the changes to the ucast but now the IP are coming up very
slowly. Basically if I run /etc/init.d/heartbeat start it takes about
2 minutes before the IP comes up. Any ides why?

Incidentally there's no firewall between the nodes.

Thanks
Dan