Blueprint status "In Progress" for over hours and Machine status is Requested on Managed machines interface

[Ousseni]

Get the error below when requested a virtual machine from blueprint: test20150106: sendEBSMessage4(workflow=acd65fbf-30a0-4ac9-b894-5fcde9b264ae) Error in state VMPSMasterWorkflow32.Requested phase PRE event (queue = 24e33afd-84de-4b95-bb5f-6d8d90c5ee33): Timed out while waiting for Event Broker response. When import virtual machine, the status is completed but, see nothing.

[GrantOrchardVMw]

Interesting one. Are you using the event broker to subscribe? That error message indicates some kind of problem there. I'd suggest opening a case with GSS.

Grant

[Ousseni]

Thank you Grant, I open a case since a couple of day and they are taking a loook on  it. But I'm also looking for alternative solution.

[Uridium454]

Ousseni, have you made any headway on this issue?  I am running into exactly the same problem.  I am not using the event broker to subscribe.  I have cut out all calls to vRO and created a basic blueprint, and I have no events firing off the needed in vCenter.  When looking through the logs, I can see my IaaS user logging in and out, but the machines are not being deployed.

[Ousseni · Accepted solution]

Hello Uridium454, Here is how to solve this issue. Logon as tenant or infra admin, then go to the infrastructure tab. So  Infrastructure >  Administration >  Global Settings In global setting under "Group: Extensibility", edit  "Disable extensibility" and change the value to "True". By default it is False. Then reboot all the services in the Manager server or just quick the manager server it self. It works for me hope it will also for you

[x2jlmote]

I am having this problem right now, and I believe I've figured out what it is. I'm still working on deploying the Orchestrator appliance, but I believe that is the fix. The documentation says in a couple places that if you're doing a distributed install you can NOT use the built in vR-Orchestrator. But if you just use the wizard to do the install, it sets everything up with the built in Orchestrator. I think that using the full Orchestrator appliance will fix these issues.

[GrantOrchardVMw]

That's not strictly correct. There are recommendations about breaking it out in the reference architecture for large (30k-50k managed machines) environments. It should not be the cause of this problem.

Grant

[heino_skov]

The above solution is not the correct solution - if you want to use event broker service.

I ran into the exact same issue and this issue is most likely because the IAAS certificate is not trusted - or do not have the correct chain in place. Make sure the IAAS web/manager server has the root and intermediate certs in place.

[jayjay72]

Unfortunately this solution disables event brokers so if you were using them before you are out of luck.  Has anybody found a different solution?

[darrenoid]

For me I had a ticket open with VMWare and after a few months finally got an Engineer to look at it. They said the problem was the "solution user", whatever that is, changed at some point (possibly after an upgarde to vra). So the user that the subscriptions were trying to use for the rabbitmq messaging did not have any permission. When you create the first event broker subscription, an item is added to the vra appliance database that basically says, vra is going to use event subscriptions now, and it will use the following solution user. The user is then hard coded in the database and if the real solution user ever changes, the event broker will timeout.

At least that is what was the case for my issue. The fix was to delete that row from the database, but unfortunately, I didn't take any notes on which one it was. One symptom of this was that I was not able to delete all of my subscriptions from vRA, when I got to the last subscription, it would not let me delete it.

Regards,

Darrenoid

[sbeaver]

When I faced this issue this turned out to be the fix for me to add the polling.  -- Requests Failing After 7.0.1 Upgrade with Event Broker Timeout