Got it! On my way.... Thanks Mike!
Nick. On Thu, Sep 22, 2011 at 1:05 PM, mike <[email protected]> wrote: > On 11-09-22 11:45 AM, Nick Khamis wrote: >> Hello Mike, >> Thank you so much for your response. >> >>>> You do not need to install cluster stack on "real" or backend servers >>>> just the nodes that are actually part of the cluster. >> This is the part that I am trying to make sure I absolutely >> understand. For example, >> we would like to setup HA on mysql1, and mysql2. That makes them a node in >> the >> cluster, and therefore the cluster stack must be installed on them? Or >> can we have >> clu1 and clu2 (cluster nodes with the stack installed in them), manage >> mysql1 and >> mysql2, mysqlN? Please don't laugh, i am just trying to make sure I >> understand this ok! ;) >> >> Cheers, >> >> Nick >> _______________________________________________ >> Linux-HA mailing list >> [email protected] >> http://lists.linux-ha.org/mailman/listinfo/linux-ha >> See also: http://linux-ha.org/ReportingProblems >> > > Hi Nick, > I'm not laughing - I was once where you are now. > > A typical MySQL cluster is going to consist of 2 servers both running > mysql AND LinuxHA. When the cluster is running, you'll have one node as > the primary node running a VIP (virtual IP) and Mysql. The other node is > sitting running as a backup. If the primary node dies, it will fail over > the VIP to the backup node and start up mysql. > > As an example, here is a snapshot of a cluster I have running for a > customer: (I've changed the domain names for privacy) > > ============ > Last updated: Thu Sep 22 16:58:01 2011 > Stack: Heartbeat > Current DC: dbsprd1a.domain.com (db80324b-c9de-4995-a66a-eedf93abb42c > ) - partition with quorum > Version: 1.0.6-17fe0022afda074a937d934b3eb625eccd1f01ef > 2 Nodes configured, unknown expected votes > 1 Resources configured. > ============ > > Online: [ dbsprd1b.domain.com dbsprd1a.domain.com ] > > Resource Group: mysql-clu > mysql (lsb:mysqld): Started dbsprd1a.domain.com > ip_mysql (ocf::heartbeat:IPaddr2): Started dbsprd1a.domain.com > > > Migration summary: > * Node dbsprd1a.domain.com > * Node dbsprd1b.domain.com > > /////////////////////////////////////////////////////////////////////////////////////////////////////////////////// > > So you can see that I have a cluster group consisting of a VIP and the > mysql RA. The 2 nodes in the cluster are dbsprd1a and 1b. > > Now the tricky part in this cluster and in any mysql cluster is, how are > you going to access the database? You have 2 instances of mysql but > you'll only have one database. That database is going to have to reside > on some sort of shared filesystem that is accessible on both nodes but > not simultaneously for obvious reasons. But that's getting a little > ahead of ourselves. > > What I would recommend is that you install the cluster software and then > start small. Try and get a VIP up and running, bounce it back and forth, > watch the logs to see what's going on. Then you'll have a good feeling > for what's going on. > > BTW - I am preparing a small wiki that has some how-tos I put together. > If you like I'll send you the link to it and you can read it over and > perhaps it will help with your understanding and installation of HA. > _______________________________________________ > Linux-HA mailing list > [email protected] > http://lists.linux-ha.org/mailman/listinfo/linux-ha > See also: http://linux-ha.org/ReportingProblems > _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
