vRO cluster on a new 7.3.1 install shows one node as "The Control Center Node is not Available"

[qc4vmware]

We've just deployed a brand new instance of vRA using 7.3.1.  It has two vRA appliances and we are using the embedded vRO.  The second node always shows as "The Control Center of the node is not available" when we try to verify the cluster settings.  If I shutdown vRO on the primary host when I login to control center it shows the second host as requiring a reboot.  We have of course rebooted it but nothing seems to fix this issue.  One thing I have also noticed that is different on this installation as opposed to our exiting install which was done with 7.0 or 7.1 (can't remember) is the individual nodes hostnames show as the hostname not as the load balanced ip name.  In the environment where things are not working the VIP shows as the hostname on both nodes.  It does correctly show the ip addresses.  This may be as intended but I figured I would call it out as that is about the only thing I can see that seems different.  Any suggestions on what may be going on?

Paul

[daphnissov]

I saw something similar on a new install of 7.3.1. Did you get resolution here, Paul?

[qc4vmware]

Not yet but I do have a support case open.  So far though I've not actually interacted with an engineer.  Hopefully on Monday.

[daphnissov]

Ok, Paul. If you do hear back and get some resolution on it, I'd be very interested to know if this is acknowledged as a defect or if something else is going on.

[IBMChakresh]

Hello ,

I am facing same issue , Did you get resolution?

[Enter123]

Hi,

yes, we installed 7.4

[qc4vmware · Accepted solution]

We did get this fixed.  Required manually editing some of the configuration.  Close notes on our case say:

Have resolved issues after following steps from PR 2111143

Our support request number was 18766844604 for reference.

[qc4vmware]

Sorry forgot to come back here and update this.  We went back and forth for a couple of weeks to finally get a resolution.

[daphnissov]

Thanks qc4vmware​ . Do you have details on what that PR entails? Did they convert it into a public KB?

[qc4vmware]

Not sure if it was converted to a public KB or not.  If I remember right somehow a portion of the cluster configuration that should have been set to use the individual nodes names was set to use the vip name.  Looking back through my case I think this is what fixed it... I was also on the phone a ton but it looks like this was the fix.

1. SSH login to each virtual appliance node.

2. Open the file /etc/bootstrap/everyboot.d/00-20-map-lb-to-localhost for editing.

3. Change line 17 from:

sed -i -r -e "1 i ${entry}" "${hosts}"

to:

sed -i -r -e "\$ i ${entry}" "${hosts}"

4. Save and close the file.

5. Reboot the vRealize Automation environment.

6. Once the environment is back online and all services are registered, verify that the vco-configurator service is started.

7. Login to the Orchestrator Control Center and navigate to the Cluster Management page. Validate that the Control Center connectivity issues are now resolved.

After I also have these notes in the case...

Me:

OK… just got the cluster to sync. I could not get into the control center unless I shut down vRO on the second node both the service and the configurator. Then I had to connect to control center using the vip name. Once I was on the config page of the vip I then had to start the services on the second node. A few minutes after the services started the cluster page shows as sync’d. I could not login to the control center starting from an individual node. I always got redirected to the vip and also got a bunch of certificate errors I could never get past.

Support:

Hello Paul,

Below are additional recommendations from the Engineering :

Please try the following:

1. Delete the browser cache and old cookies.

2. Open a new Firefox Private window and navigate to https:// <https:><vRA_appliance>/ page

3. Click the Control Center link -"vRealize Orchestrator Control Center"

4. Confirm the browser security exceptions and login as administrator user.

5. If step 4 fails, verify that your vRO Control Center load balancer settings are correctly configured.

Me:

OK… that worked!

[daphnissov]

Thanks, Paul. This will be helpful to me and others should they find this page. 7.3.1 had some real problems with this being one.