E9.2 Change Assistant Error for UDOs

[Seane6]

Hello all,

I've recently been assigned some CNC tasks, I am relatively new to the position and wondered if anyone else has seen this situation.

We are at Update 7, Tools 8.1 but haven't been bringing down UDOs like we apply ESUs until recently. We are trying to get current on UDO bundles in a test environment but when we attempt to deploy them via Change Assistant on our deployment server I continue to get the following error:

      Failed to complete pre-deployment setup. Cannot proceed with deployment.

The change assistant log appears as follows:

com.peoplesoft.deployment.framework.DeployerException (Thread[AWT-EventQueue-2,6,javawsApplicationThreadGroup], 2024.01.15/17:56:21:885)
java.lang.NullPointerException (Thread[AWT-EventQueue-2,6,javawsApplicationThreadGroup], 2024.01.15/17:56:21:885)
Error in Deployment
Failed to complete pre-deployment setup.
null
Cannot proceed with deployment.

I haven't been able to find much on MOS. I was instructed to check the jdbc connection on our deployment server via ctl-alt-shift-j in change assistant and it connected successfully.

We are able to successfully import individual UDO packages successfully, but there are too many UDOs in the various UDO bundles to import them in this way.

This is not a production problem so it's not a critical issue; therefore, I don't think this will get prioritized. But we are trying to get a better handle on UDOs and how to use them and are dead in the water right now.

Our setup:
E1 9.2, Update 7
Oracle 19c Database
Linux Redhat 8
WebLogic 12.2.1.4

Anyone seen this before? Or have any ideas on how to fix it?

Thanks

 

[TFZ]

Turn the security off on your dep server in the JDE.INI, kill the service there if it's running (its in services) and try deploying the bundle again (I want to say I had to pick DEP920 instead of JDEPLAN). At least, I had to do that for the Update 8 bundle recently.

 

[Seane6]

TFZ,
Thanks much for the reply. I initially had security turned off on the deployment server and got the message. I even tried it turned on just to see if it made a difference and it still failed.

I did not try DEP920 before and knowing that this is what I will have to do when I eventually get to the UN8 bundle is good to know. However, when I tried it for UN2's UDO bundle I got the same error.

At some point I got past the error and was presented with a screen I have never seen indicating that the bundle was extracting with an elapsed timer. However, it did not appear to be extracting anything and when I attempted to direct the extract to a folder I received an error "No file or directory selected.". After allowing it to run for several hours in the hopes that it was somehow extracting somewhere, eventually I closed it out. The local batch went to pending rather than error but I could not get it to actually extract. After restarting Change Assistant I had to delete the pending batches and try again, resulting in the same error as initially seen. I have not been able to replicate the extract screen, but believe it may have come up when I simply cancelled out of the E1 sign in screen.

 

[MFreitag]

Anything in jde.log?
...and as you said, regular ESUs work fine, right?

What's the ODBC you're using and are you using JRE 64-Bit with Change Assistant?

 

[Seane6]

MFreitag,

Thanks for the reply.

Nothing in the jde.log. And yes, regular ESUs work OK. Just tested with another regular one this morning to make sure.

When you say what ODBC am I using, not sure what you mean there. We don't have any ODBC connections other than what I'm told is an older connection called ODA.

If you mean the JDBC driver, we are using ojdbc8.jar under E920\MISC. We are using the 64-bit JRE with change assistant. I looked into the file JavaDeployReg.log and it appears to be showing that we are using the 64-bit JRE.

One additional test I ran was to try a newer UDO bundle, in this case the UDO bundle for UN8, just to see if I could get to the environment screen and it gave the same error. I tried both the planner and deployment environments.

It seems like Change Assistant can't get to the point of extracting the contents of the par file? When I use Change Assistant to download a specific UDO (which downloads as a zip file) it seems to work OK, at least to the point of getting past this error and to the path code selection screen.

 

[MFreitag]

Is the latest Planner ESU is installed as well?
Note sure about the 19c, but i've had problems a while back in a 9.1 env with 12c with the odbc8.jar, maybe you wanna give it a shot and try it with odbc7.jar?

 

[TFZ]

I recently went through a fun CHA exercise for batch deploy failing via AIS. I suggest opening a ticket with Oracle, they can give you a file to enable enhanced logging (which they put an expiration date on for some odd reason or I'd just send you the parms.) You can try turning on java console logging - but yeah, for me it wound up being the security server was on..

 

[Seane6]

MFreitag,

Yes, put the latest Planner on prior.

Tried ojdbc7.jar and same error. I made sure ojdbc8.jar was actually removed from the MISC folder as I have seen where the wrong one gets selected if you have more than one available.

Also, not sure it's worth noting but I am able to apply a data pack via Change Assistant, at least up to the point of the path code selection. It just seems like Change Assistant is balking only at UDO bundles when the par contains all of those zip files

 

[Seane6]

TFZ,
Good to know about the enhanced logging on CHA. I was looking for some way to increase the logging level in the tool itself but apparently it needs some intervention on Oracle's part.