Error: LCMVRAVACONFIG590004 - Initial vRA deploy through Easy Installer

[Christiankkcc]

Hi All,

I've see this error throughout my searches and only a couple within the community but I still haven't been able to find a solution.

During a re-installation of vRealize via the Easy Installer, I've been having this

error: LCMVRAVACONFIG590004

com.vmware.vrealize.lcm.common.exception.EngineException: vRA VA deployment Failed at First Boot check on HostName ci-0027743.evgroup.com at com.vmware.vrealize.lcm.plugin.core.vra80.task.VraVaPostInstallCheckTask.execute(VraVaPostInstallCheckTask.java:104)
at com.vmware.vrealize.lcm.automata.core.TaskThread.run(TaskThread.java:45)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
at java.lang.Thread.run(Thread.java:748)

 

Looking through other solutions:

-Using ESXi 6.7 U3, vCenter 7, and I've tried vRA EasyInstaller 8.2 and 8.0 with no luck.

-DNS Settings are good, reverse and forward entries are good.

-No duplicate PRT records or any other records.

-DNS IP address points directly to my DNS server (no load balancer).

-All other network settings are good.

-NTP server connectivity is good.

 

I'm completely at a loss here. Any help, is much appreciated.

[JimH1]

Same problem here. Checked everything.

[jimmyvandermast]

For some (out of many) reason, LCM cannot reach the new vRA VA and times out.

Do you have a firewall, like NSX, where the old vRA VA is removed and not correctly replaced by the new one?
Perhaps you can check, just before you get this error again, if the LCM can reach the vRA VA over https (use curl)?

Not sure if the firstboot check is via vCenter-VMtools or via https, check both.

[jimmyvandermast]

LoL, currently deploying a new vRA cluster here, and this is what the firstboot check does:

 

SshUtils - -- Executing command --> vracli status first-boot --wait=120

 

So you can test an SSH connection from the LCM to the vRA VA.

And perhaps even do "vracli status firstboot --wait=120" manually to see what it returns.

[bdamian]

Something similar happens to me and the problem was the Reverse Lookup. My PC was in a different DNS and wasn't able to find the PTR entry in the LAB DNS.

My solution was to install Lifecycle Manager ONLY with easy installer and then use the LCM console to install vIDM and vRA.

Hope it helps.

[Christiankkcc]

Hi all,

I believe the solution to my problem was very specific/strange. I talked to a senior technical support engineer and he fixed my issue here. 

As you know, the error code is related to a network error. Although I received this error, the fix wasn't a network issue at all.

Here is what happend:

1. My vRA 8.0 installation via the Easy Installer was stuck at stage 4, then ultimately failed. vRA environment failure could be seen from vRLCM.
2. We attempted to ssh directly into the appliance, but failed because root wouldn't accept my password. (The password is the same I use in vCenter and the same thing I inputted into the Easy Installer)
3. Rebooted the vRA appliance and reset the password in emergency mode.
4.  vRA root password was reset and ssh access into the vRA appliance was successful.
5. Within vRLCM, we retried the vRA failed deployment and the task progressed to stage 6.
6. At stage 6 of the vRA re-deployment, it got stuck again because of an expired container password for the vco-app pod and the pods were also in backoff state. (The actual vRA appliance is all Kubernetes based. This, I didn't know at all. Very interesting.)
7. The senior engineer corrected the deployment.yaml (sorry, I don't know exactly what he did), then deleted the failed vco-app pods. The vco-app pods then were auto deployed with the new deployment.yaml configuration.
8. The vRLCM deployment of vRA then moved through the final stages of the installation and everything after this was successful. vRA was up and running with no problems and I could finally see the web front end and finally log in.

 

Also, being that this entire deployment was vRA 8.0, I took a chance and completely deleted all of it again. Before, I've tried every version, and nothing worked. Anyway, I began a brand new installation of all appliances in 8.3 via the Easy Installer, completely from scratch. During the installation, I had no errors at all and everything worked well!

The engineer also doesn't know why even the new installation of 8.3 works, even after I deleted all of his efforts in our 8.0 installation. He also doesn't know why I received a "LCMVRAVACONFIG590004 network error" when the problem wasn't network related either.

So in short, the appliance didn't take my password, which then broke K8S and it's pods needed to be deleted then re-created.

I think this is a very isolated incident and hopefully it is.

I hope this can help others.

[j3di77]

So I had this similar problem. My DNS and PTR everything was correct.

For some reason the communication between vRLCM and vRA is lost as the vRA password is not set.

I rebooted the vRA appliance and reset the password following the below link as per the password I had entered in vRLCM easy installer

https://docs.vmware.com/en/vRealize-Automation/8.1/Administering/GUID-1BC4BD89-73C5-4C2A-875D-B8787418EE85.html.

Once the password was done, just clicked on retry on vRCLM web portal and voila it progressed like a charm.

Hope this helps