Skip to main content
Solved

Disabling Feature caching causes transformer to fail.

  • December 7, 2022
  • 5 replies
  • 102 views

michaelbreen
Contributor
Forum|alt.badge.img+2

Has anyone experienced a workbench erroring when feature caching is disabled?

 

For context I have had this workbench running through task scheduler for around two years (keeping it updated) and recently discovered an issue where it crashed at a specific transformer (feature joiner).

In testing I have run the workbench with feature caching on (it worked fine) so I tested all the normal points - replace transformers, replace readers/writers, change user credentials.

 

When I turned off feature caching is where the error occurs (in the same spot as when running it through task scheduler).

The attached logs show where it crashed on the feature joiner, has anyone experienced this?

 

My last step is to recreate the workbench from scratch and test that as well.

Best answer by ebygomm

I've seen something vaguely similar previously, which was down to an issue with bulk mode and a certain transformer. You could test if this is the issue by putting a declerator before the feature joiner set to 0s and see if your workspace runs then with feature caching turned off.

View original
Did this help you find an answer to your question?

5 replies

ebygomm
Influencer
Forum|alt.badge.img+32
  • Influencer
  • Best Answer
  • December 7, 2022

I've seen something vaguely similar previously, which was down to an issue with bulk mode and a certain transformer. You could test if this is the issue by putting a declerator before the feature joiner set to 0s and see if your workspace runs then with feature caching turned off.


Forum|alt.badge.img+2
  • December 7, 2022

@michaelbreen​ @ebygomm​ is probably correct - it looks like a bulk mode issue in the FeatureJoiner. If adding the Decelerator does resolve the issue, then first try updating to the latest version of FME, if you can (remember that you can have multiple versions of FME Desktop installed - just use a different install folder). If the issue persists then perhaps file a support case with your local FME provider so we can try and address the problem.


michaelbreen
Contributor
Forum|alt.badge.img+2
  • Author
  • Contributor
  • December 12, 2022

Gentlemen @ebygomm​ @Mark Stoakes​, thank you very much for steering me in the right direction.

Adding the Decelerator has resolved the issue and the workbench is now running as normal in either caching or non-caching modes.

I have a suspicion that an updated version would also work a treat as we are still running on 2022.0.0.1 but it would take a bit longer for IT to approve and deploy the update. (I'll try it when I can).

Have a good one and happy FMEing.


Forum|alt.badge.img+2
  • December 19, 2022

looks like this is addressed in newer builds of FME 2022


jamesadams
Contributor
Forum|alt.badge.img+3
  • Contributor
  • October 10, 2024
ebygomm wrote:

I've seen something vaguely similar previously, which was down to an issue with bulk mode and a certain transformer. You could test if this is the issue by putting a declerator before the feature joiner set to 0s and see if your workspace runs then with feature caching turned off.

I too had to use decelerators to solve an issue with FeatureJoiners while not using caching. I was experiencing intermittent cases where the join appears to start before all of the records where ready (left and right). This created a mixed set of results, some with spatial features and other without. Adding a 0s decelerator to both the imports before hand correct the problem. This comes across, to me, as a bug. I’m using 2024 - FME(R) 2024.0.2.1 (20240513 - Build 24217 - WIN64)


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings