I have a realy hard time to make a cluster on a centos 7.3.1611 in prod but was able to make it work in lab. Basically I have followed the following steps:
Code: Select all
- Install centos core system, give an ip, very basic stuff
- yum -y install pacemaker corosync fence-agents-all fence-agents-virsh fence-virt pcs (on 2 nodes)
- echo password|passwd --stdin hacluster (on 2 nodes)
- systemctl enable pcsd (on 2 nodes)
- systemctl start pcsd (on 2 nodes)
Code: Select all
pcs cluster auth node1 node2 -u hacluster -p password
Now in a system in prod using same version of centos, same rpm from the lab I can't make it works...
- the command pcs cluster auth <ip1> <ip2> -u hacluster -p password results with:
Code: Select all
Error: Unable to communicate with <ip1>
Error: Unable to communicate with <ip2>
Code: Select all
Unable to connect to node <ip1>, no token available\n
Unable to connect to node <ip2>, no token available\n
Code: Select all
/usr/bin/ruby -I/usr/lib/pcsd/ /usr/lib/pcsd/pcsd-cli.rb auth
I have also unset the http_proxy and https_proxy
If I try to access https`://<ip1>:2224 after the login I have Internal server error message.
It should be something very simple but I can't find what is not working.. any help will be greatly appreciated.
Thanks!