vRO Workflow Stuck at Waiting Event

[sc2317]

Hi,

Can someone please help in resolving issue with multiple workflows stuck at waiting event. Basically, users are requesting machnes via vrealize automation which runs vrealize orchestrator workflow at the back end. However, I am not able to identify why its stuck, so is there a way to fix this or make workflow skips this waiting event and proceeds to next element ?

I have already started vro services but no avail.

[stevedrummond]

Considering vRA doesn't run any workflows by default, this must be a custom workflow or EBS pointed at an OOTB workflow.

What is the ​Waiting Event​ step configured to wait for, and is it a custom or ootb workflow?

[sc2317]

Hi Steve,

It is custom workflow. Trigger is the preceding element of "Waiting event" and below is the scripting of  trigger event.

System.log("Creating trigger for catalog item request " + request.vcoId + "...");

trigger = vCACCAFERequestsHelper.createTriggerForCatalogItemRequest(request , timeout)

System.log("Trigger created with time out "+ timeout);

Trigger has 2 input parameters and 1 Output Parameter

Input Parameters

"Request" of type "vCACCAFE:CatalogItemRequest" and "timeout" of type number.

Output Parameter

"trigger" of type "Trigger"

After this, workflow goes to "Waiting event" and stucks there.

[stevedrummond]

You've double checked the correct CatalogItemRequest object is going into the trigger creation?

What version of vRA are you running? I don't use Triggers/Waiting Events so I'm not sure of any particular issues with them.

You asked if that step could be skipped? As it's a custom workflow you can update it however you want. However, the logic had a wait for the request to finish so presumably it needs to know that?

If there are issues with the Triggers you can always just build your own Wait For Request logic.

[sc2317]

Hi Steve,

Many thanks for your reply.

I have managed to fix the previous issue and now workflow is deploying the machine. However, workflow also moves deployed machines to required OU in AD which is not working at the moment and any machine that is deployed is of no use because machine is not moved to required OU and due to this users do not have access to it.

Basically, below scripting element throws an error that "VM not found" & "Current OU not found"

uuid = System.getModule("com.exact.helpers").getUuidForVMbyCatalogRequest(vCAChost,catalogItemRequest) ;

vmn= System.getModule("com.exact.helpers").fndVcVMbyUuid(uuid) ;

result= System.getModule("com.exact.helpers").getComputerDNforVM(vmn) ;

Could you please suggest.

[stevedrummond]

It's difficult to say since those are custom actions being run. Do the actions work for machines that you aren't in the middle of provisioning?

At a guess you could be running into the provisioning race condition issue. Your VM's are joining the domain against a domain controller however it's not guaranteed to be the same domain controller that your have vRO connected to and which has not yet received the replicated data.

Do the machines show up eventually? If not something else is going on, if they are you'll need to work on some logic to connect to the correct DC, wait, or force machines to join a particular DC, or the best solution is to have the machine join the domain into the correct OU to begin with.

[sc2317 · Accepted solution]

It seems like workflow was stuck at executing "customizeos" batch file and after few hours it was failing. I figured out that machines were getting deployed to OU in AD which had windows UAC enabled and due to this customization script was not running.

Basically, any machine that joins AD goes to "New Objects" OU and running scripts as local administrator does not invoke UAC. This was the intended way of doing customization, not sure how it got changed. Anyways, I am happy issue is resolved and also I followed "VMware Knowledge Base " to troubleshoot guest agent of vRA.