HA in Distributed monitoring

Is there any documentation on how to setup HA with my 1 master and 2 slaves setup ? At the moment, I want to setup HA only on master in a way if master dies then the slave will get the data from other master. I have a on-premise setup at the moment with RHEL7.

My idea is to first provision another master with the same configuration on a seperate host. and then configure HA using Pacemaker/Corosync/DRBD based solution.
Can you tell me what files are important for the sync to happen?
Later on, If I plan to do HA for the slaves as well ? Is it the same files ?
Since, this is a manual HA cluster steup which will require lot of tuning here and there - Will I experience a data loss risk ?

This is easily possible via the Checkmk Appliance which also uses the Pacemaker/Corosync/DRBD solution and takes care of the files to be synced. Here You can setup a HA cluster between Master(active)-Master(standy) and with this approach, the slaves will always connect to the cluster IP. Same approach you can adopt among the slaves where the master connects the slave via a cluster IP. This way both master and slave are in HA.

PS: Have never tried to setup the cluster manually though but it should be possible.

Some points from my side.
If you have a master-slave, i setup the master most times as an virtual machine with HA capability inside my ESX/HyperV/KVM environment. This is bean done as the master has nothing to check only provide the web interface.
The slaves by itself if they should be HA then the CMK appliance or any other self made Pacemaker cluster will work. The slaves most times are build on bare metal.

Hallo,
you can find some tutorials on several webpages but the easiest way is to use the appliance-
Ralf

Thanks @andreas-doehler @rprengel

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed. Contact @fayepal if you think this should be re-opened.