How to do manual takeover/switchover with UTM

Question

Hi, 
 we currently plan to move one node of our Active/Passive cluster to a different Rack (same DC). We already prepared everything for this but a last question arose from that: How can we do proper takeovers? 
 I know, when up2date the cluster, the slave and master are being updated separately and they do an automatic smooth takeover. I also know that there are Reboot and Shutdown buttons besides the nodes and i know of the CLI command "ha_daemon -c takeover". 
 So what i need is a reliable and smooth way to switchover/takeover from the current master to the current slave (ready state) and - if needed - a way to login to a slave and get him up as a master (in case of something really goes wrong). Can i simply reboot the master, so the cluster makes a switchover to the Slave and the Master comes up as the new Slave (we don't have preferred master setting set). 
 Or can we login to the master via CLI or SSH and use the ha_daemon command the same way? 
 Wasn't able to find reliable information on that. Thanks for helping me! Martin

Steve Weißflog · Accepted Answer

If you have a fully functional HA-Cluster (active/passive) a reboot of the active master will initiate an automatic switchover/takeover to the slave. So the slave will be the new active master UTM. While rebooting the former master this device will sync all data from the new master and after syncing this device will stay as slave. 
 So if you just click reboot inside webadmin on the active master, you'll initiate a takeover. (Normally there will be just 1-3 pings that get lost while switchover and also SSL-VPN will not be disconnected!) 
 
 regards 
 Steve