During some updates bcl clusters update has been removing all but one of the existing _existing_ nodes and replacing with new nodes along with the new clusters nodes #170
The update process has been doing this for a while now. I posted in Slack about it and we chatted about it briefly. The behavior continues to present itself. Aside from not being the expected behavior this is problematic in so far as when I update I now trigger alerts because the number of healthy running nodes is reduced to one and run the risk of ending up with a fully unhealthy cluster should the remaining node decide to fall out of sync.
The update process has been doing this for a while now. I posted in Slack about it and we chatted about it briefly. The behavior continues to present itself. Aside from not being the expected behavior this is problematic in so far as when I update I now trigger alerts because the number of healthy running nodes is reduced to one and run the risk of ending up with a fully unhealthy cluster should the remaining node decide to fall out of sync.