that indeed has somthing to do with your setup. check the cluster.node.ip inside the db.properties on each mgmt server - value of this should be == the local IP of the mgmt server.
Accessing (externally) UI via HAProxy should not affect this... can you share the "select * from mshost;" You can also set "removed" value for those "old" records in the mshost table (which are already down) to some date in past - so if they have the "removed" date set, ACS considers them removed, and you will not see them in the UI. Best, On Mon, 21 Dec 2020 at 03:48, Niclas Lindblom <niclas_lindb...@icloud.com.invalid> wrote: > Hi all, > > I have 2 management servers load balanced with HA Proxy which all appears > to be working properly. However, each time a reboot a server a new > management server is added in the UI (database) and I now have a long list > of servers showing as down and the 2 currently in use showing as up all > with different uuid's. > > I am guess this have something to do with the HA proxy load balancing, but > not sure how to troubleshoot and would appreciate any pointers. > > Thanks > > Niclas -- Andrija Panić
