Odd template behavior in 9.3

I take it the BR assigns the account definition, right?

I've made some tests and it works fine in my case.

• Modified the template for the CN attr on a custom UnsAccountB
• Created the UNSAccountB and linked a user. The template works.
• or, located the user in Manager, assigned the corresponding account definition and the template applied, too

There's more to it, Markus. 

The info you requested:

PS C:\> dotnet --list-runtimes
Microsoft.AspNetCore.App 8.0.11 [C:\Program Files\dotnet\shared\Microsoft.AspNetCore.App]
Microsoft.AspNetCore.App 9.0.1 [C:\Program Files\dotnet\shared\Microsoft.AspNetCore.App]
Microsoft.NETCore.App 8.0.11 [C:\Program Files\dotnet\shared\Microsoft.NETCore.App]
Microsoft.NETCore.App 9.0.1 [C:\Program Files\dotnet\shared\Microsoft.NETCore.App]
Microsoft.WindowsDesktop.App 8.0.11 [C:\Program Files\dotnet\shared\Microsoft.WindowsDesktop.App]
PS C:\> dotnet --list-sdks
8.0.404 [C:\Program Files\dotnet\sdk]
PS C:\>

The other machine is linux and it's running the jobserver and dbagent on docker containers.

It's been a temporay success. The problem persists. Only worked once after:

• Changing the jobserver and dbagent
• Resetting the administration workstation (the one I've ran the dotnet --list-runtimes)

Now it's there again. The jobserver and dbagent keep running on docker. But I've made changes to the UNSAccountB template and renamed the process. Recompiled the database and , again, the template that's applied is the older one from early this morning and the process name shown in jobqueueinfo keeps being the older one too.

I've changed the Designer settings:

* Save databases locally:OFF

* Use RAM: ON

* Clear caches

reopened Designer. Tried and, again, no changes at all. Template does not apply, process keeps the older name.

Changed the administration workstation, retried. Nothing. The template does not apply, process keeps the older name, though I've changed it 2 times and recompiled the db.

In your case, the JobServer is running in a Linux Docker container, right? If you delete and re-create the Docker Container (just stopping and starting is not enough), does it work then (assuming that you do not persist the caches)?

YES!!!! (I cannot attach images)

It works!!! Recreated the container and now I can see the job process name has been updated in jobqueueinfo and the template is applied correctly.

Changed the process name again, retried and it keeps being the older one. Recreated the container. Process name updated. 

Yes, you've found it Markus Weiss-Ehlers 

So it seems to be a cache problem.

Thanks for drilling down into the issue with me. I will forward this internally.

One additional question came up.

Do you use the Session based HTTP authentication module (SessionHttpAuthentication) to protect the Job Service Web Status page?

If not, using it might be a workaround for the time being.

Verified. I was using a jobserver in docker linux with only http auth based on user/password and that did not work. So I've tried with a legacy jobserver on Windows using session based http auth and , yes, I can confirm. It's a good workaround. That worked.

Thanks Markus!

You may ask support if a hotfix for #474728 has been created in the meantime. Good to hear, that the workaround is working.

Will this fix my non-Docker issue with the template?

I think it will. Or if you dont want to wait for the hotfix, modify your jobserver (legacy and non docker) as Markus said: Configure your Jobserver with the session based http auth module. That has worked for me.

Do I do it on all virtual jobservers or?
What information do I put in under Application URL and Job Provider ID?

Should be done in all jobservers, yes. In my case I have only one in my lab for v93. Maybe the hotfix will apply for all jobservers, I dont know yet.

And about the App URL and Job provider ID , I've left them blank.

Should be done in all jobservers, yes. In my case I have only one in my lab for v93. Maybe the hotfix will apply for all jobservers, I dont know yet.

And about the App URL and Job provider ID , I've left them blank.

I've now added session based http auth module to all of my queues and job servers (3 servers/9 queues) and although I haven't tested my initial problem I'am stll having a cache problem. After changing a process the changes doesn't come into effect, I have to stop all IAM-services on the job servers and remove everything under %localappdata%\One Identity\One Identity Manager\AssemblyCache and then the right "version" of the process runs.

Markus Weiss-Ehlers Do you have any more suggestions?

I do not have any other suggestions besides asking the support for a hotfix for #474728 to check if this helps in your situation.

Thanks, Markus. Shall we see a patch/fix/service pack in v93 for this bug any time soon? 

Please contact support and tell them you are affected #474728 and need a fix, that would help to track this issue. And yes, something is in the making in regard to a hotfix.