[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [Linux-cluster] Use alternate network interfaces for heartbeat in RHCS

Hi Chrissie,
Are you pointing me to the paragraph "what's the right way to ....eth0 ?"
I've tried this at first adding adding a suffix to the interfaces but nothing happened. Suffix -p may be hardcoded (I've used -hb)
Here's an outputof my /etc/hosts (identical on both nodes):   node1 node1.mydomain   node2 node2.mydomain   node1-hb   node2-hb
Still using bond0 ....

2009/8/28, Christine Caulfield <ccaulfie redhat com>:
On 28/08/09 15:10, brem belguebli wrote:
I'm trying to send cluster communications thru dedicated LAN interfaces
(out of band network) separate from the production interfaces but it
doesn't seem to be functional.
I have a 2 nodes cluster (RHEL 5.3) with 2 bonding interfaces each, one
bonding for production on a given Vlan and the other on a dedicated and
isolated LAN (out of band LAN).
According to some cman doc (that may be outdated), the way to do so is
to declare explicitely at each clusternode level  the multicast addr of
the cluster and the interface on which the coms are sent.
<clusternode name="node1.mydomain" nodeid="1" votes="1">
<multicast addr="" interface="bond1"/>
<clusternode name="node2.mydomain" nodeid="2" votes="1">
<multicast addr="" interface="bond1"/>
and the cman multicast addr:
<cman expected_votes="3" two_node="0">
<multicast addr=""/>
(I have a qdiskd).
After restarting the cluster, traffic is still sent through bond0 which
is the production interface.(tcpdump -i bond0 host

I don't know where that document is but it's rather out of date, as you suspected ;-)

The wiki has a up-to-date item on how to assigned different ethernet interfaces for cman:



Linux-cluster mailing list
Linux-cluster redhat com

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]