[ClusterLabs] Does CMAN Still Not Support Multipe CoroSync Rings?
Eric Robinson
eric.robinson at psmnv.com
Mon Mar 5 01:55:55 EST 2018
Well, I finally got around to trying out the <altname> tag and I managed to get two rings running, but the first ring is not obeying my address and port directives.
Here's my cluster.conf
<cluster config_version="9" name="ha10ab">
<fence_daemon/>
<clusternodes>
<clusternode name="ha10a" nodeid="1">
<multicast addr="239.255.5.1" port="4000"/>
<altname name="ha10a-cl" port="4000" mcast="239.255.5.2"/>
<fence>
<method name="pcmk-method">
<device name="pcmk-redirect" port="ha10a"/>
</method>
</fence>
</clusternode>
<clusternode name="ha10b" nodeid="2">
<multicast addr="239.255.5.1" port="4000"/>
<altname name="ha10b-cl" port="4000" mcast="239.255.5.2"/>
<fence>
<method name="pcmk-method">
<device name="pcmk-redirect" port="ha10b"/>
</method>
</fence>
</clusternode>
</clusternodes>
<cman broadcast="no" expected_votes="1" transport="udp" two_node="1"/>
<fencedevices>
<fencedevice agent="fence_pcmk" name="pcmk-redirect"/>
</fencedevices>
<rm>
<failoverdomains/>
<resources/>
</rm>
</cluster>
The rings are up...
[root at ha10b ~]# corosync-cfgtool -s
Printing ring status.
Local node ID 2
RING ID 0
id = 192.168.10.61
status = ring 0 active with no faults
RING ID 1
id = 198.51.100.61
status = ring 1 active with no faults
HOWEVER, when I run tcpdump, I can see that ring2 running on the appropriate multicast address and port, but ring1 is running on the default address and port...
[root at ha10b ~]# tcpdump -nn -i bond0 net 239.192.0.0/16
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on bond0, link-type EN10MB (Ethernet), capture size 65535 bytes
22:54:36.738395 IP 192.168.10.60.5404 > 239.192.170.111.5405: UDP, length 119
22:54:40.547048 IP 192.168.10.60.5404 > 239.192.170.111.5405: UDP, length 119
How do I get ring1 running on my desired address and port of 239.255.5.1, port 4000
--Eric
More information about the Users
mailing list