[Pacemaker] pcs cluster auth shows "Error: Unable to communicate with node" message

[Sihan Goi]

Hi,

I had a basic HA setup working with 2 nodes previously running a simple
Apache web server on a private local network. However, I'm having trouble
getting it to work right now, and I haven't changed anything other than
rebooting a few times.

Firstly, I've noticed that I need to start the pcsd service manually after
every reboot with "systemctl start pcsd". Corosync seems to start
automatically

After starting pcsd and restarting the cluster, the HA cluster used to
work. However, now it doesn't seem to. "pcs status" on the node01 would
show node1 as online and node02 as offline, and vice versa. When I try "pcs
cluster auth node02" from node01, I'd get "Error: Unable to communicate
with node02", even though I'm able to ping both the IP address and hostname
of node02 from node01

node01 and node02 would both serve their own web page when I enter the
virtual IP address in the browser URL bar. However, a 3rd device connected
to the same network is unable to load the webpage from the virtual IP
address.

What's wrong? Thanks!
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.clusterlabs.org/pipermail/pacemaker/attachments/20140909/c87239f0/attachment-0002.html>