How to redeploy vIDM 3.3.2

[luizspinheiro]

Hi guys

After trying to update the vIDM from 3.3.2 to 3.3.3 using vRLM, without success, authentications in vRA were very slow.
Analyzing the problem, the blue screen of vIDM asks to redeploy.
How do I redeploy, since through the LifeCycle console I cannot remove the product?

[shank89]

Before redeploying, have you done any troubleshooting or looked at any logs?

 

Are there any backups of the environment?

[luizspinheiro]

Yes, I saw the logs: "Failed services in runlevel 3: elasticsearch"

I have no backup with healthy iDM.

[shank89]

You will likely have to get GSS involved to troubleshoot the slow auth, however that message can be common and there are ways to remove it.  
Also VCF currently supports vIDM 3.3.2 as part of the BOM here https://docs.vmware.com/en/VMware-Cloud-Foundation/4.1/rn/VMware-Cloud-Foundation-41-Release-Notes.html

You can try cleanly shutting down the appliance and bringing it back up, then run a health check for for VIDM from the global environment.

[luizspinheiro]

Hi  

Thanks for the days and help, but I managed to solve the problem.

After a few attempts to restore trust between the cluster nodes, with LB and discover by vRLCM, both without success.

I shut down all VMs, took a snapshot, preparing to redo the vIDM by hand (unregistering and registering a new one via API - https://docs.vmware.com/en/VMware-vRealize-Suite-Lifecycle-Manager/2018/ com.vmware.vrsuite.lcm_api_21.doc / GUID-21559F8F-C6BD-47B6-A79B-B5FF9F716941.html).

I turned on vRLCM and then turned on vIDM via LifeCycle, when I realized that the management, which was not happening before, worked, from then on I did the whole process of product discovery, re-trust, etc ...
I consulted the NSX, saw that all the nodes appeared as up, did some validations in the logs and realized that the cluster was UP again.

Thanks.

 

Luiz Sérgio Pinheiro