A quick solution blog regarding joining a new pod into a cloud pod architecture that gets stuck at 50%
Last week, I wrote a blogpost on the issues that I had with upgrading a Horizon environment using the cloud pod architecture (CP). There I wrote about how much time I lost with troubleshooting and in the end redeployed a new pod that took me just a couple of hours to complete.
After completing all the installations of the horizon connection servers, I joined the local pod into the CP architecture.
The process starts with the first connection server getting initialized and the global AD LDS instance is installed on the server. When completed the ADAM database gets replicated from the other pod. When the first connection server has finished, the overall status will be 50%. The second server should start with the same procedure.
But sometimes nothing happens and the ”pod join” keeps stuck on 50%, this is something I have encountered this during the upgrade.
After waiting more than 30 minutes, what would only take a couple of minutes.
When analyzing the VDM.logs on connection server 1, the following entry frequently appears in the logs:
2019-01-21T13:18:37.291+01:00 WARN (17E0-0994) <VipaServerFastThreadPool1-3> [LinkedModeServerUtils] LMV: Mismatching Global/Local Pod endpoints. Unable to find PodEndpoint for broker XXXXXXXXCON12
I was sure the “join” process was in some sort of endless loop, unable to initialize the second server.
I solved the issue by rebooting the services on connection server 2 that stopped the search loop. The join completed successfully within 1-2 minutes.
Comments and questions are always welcome!