Hi, sorry for what will probably be a simple question, but I just can't get a clustered deployment to work and am quite new at Elrand and eJabberd.
I am running two 1.1.2 servers on a CentOS 4.3 Linux distribution. Both were installed with the linux installer and checking the 'part of a cluster' option. The first server works fine and I have gotten as far as enabling the node and replicating the database on the second server, but for some reason can never get the second server to start.
I followed the instructions in 10.2 of the documentation to establish the node and as I said, that works. I imagine there is something I need to change in the ejabberdctl script to enable the node and also start the server at the same time.
One other question related to this. On the second box I see that once I change to disc copies of the pertinent tables an Mnesia.node directory gets created. What should the database dir be on this box? The database directory that lives in root from the initial install or the new Mnesia directory that is created as part of establishing the clustered node.
Thanks very much and again apologies for such simple questions....
Hi,I am playing cluster at
Hi,
I am playing cluster at and .
I have made the cluster running but I do not know how to access one from another. I believe that I am using one shared database. On the second node, when you start ejabberd, it will create a directory named Mnesia.[the node name]. And you can see lots files in it. I have deleted all database files in spool directory which is used before when there was no cluster. I have copied the cook from node one to node 2 and I started ejebberd with -mnesia extra_db_nodes. I am running one node on Slackware Linux and another node on FreeBSD 6.1.
My another problem is that the jwchat won't work after some time's running. I have to apachectl restart to make it work. Otherwise it says no service avalaible.
2nd server status
breakevilaxis,
were you able to access the admin panel on both the clusters? How did you modify the start script to get the second server to actually start up. I too can see the second node running but the second server is not starting.
Thanks...
I think I have already
I think I have already started both nodes. My problem is that I can not access one node from another. They are functioning independently. But I am sure both are running because I can access both of them.
I followed the procedure to start the second node. To test them talking, you can in erlang ping one node and the other one will answer with "pang". Also you can find the second node has created a directory to hold the database files.
I really do not know how to make these two guys work together like a cluster. Now it seems they work independently.
**********************
When I restart the second node, I cannot login. I have to click "register a new account". Then I back to the first node. When I use JWChat on first node, it says service unavailable. I did an apachectl restart. Then I logined into the same room I created in the second node!