This error occurs once and and while and the issue resolves itself on it's own, sometimes the next day or sometimes it persists for several days. The issue seems to be with any transformer/reader that reads in oracle non-spatial tables.
Page 1 / 1
Hi @em9
Were there any changes to the Oracle DB (like an upgrade) that coincide with the error starting? Also, was the workspace created in the same version of FME Desktop, and do you receive the same error when running the workspace in FME Workbench?
Would you be able to share the fmf file the error is referencing?
(Also, I'd recommend upgrading FME Server, as 2017 is a retired release. )
Hi @em9
Were there any changes to the Oracle DB (like an upgrade) that coincide with the error starting? Also, was the workspace created in the same version of FME Desktop, and do you receive the same error when running the workspace in FME Workbench?
Would you be able to share the fmf file the error is referencing?
(Also, I'd recommend upgrading FME Server, as 2017 is a retired release. )
Yes, the workspace was created in the same version. We haven’t done any oracle upgrade since 2020.
The only change that was made, is one of my colleagues uploaded a script to the server that was compatible with python 2.7 or 3.4+ so we initially thought that was the issue, since we did get some python errors because our script wasn’t compatible. After changing that and uploading it again, the python issue resolved and the other issue remained.
I’ll send the fmf file, but right now the scripts aren’t throwing those errors. It seems to happen for a few days in a row, and then it resolves itself. I’ll make sure to send you the fmf file again when the issue occurs again.
We are planning an upgrade to fme server 2022 next year, its difficult to do these big upgrades being part of a big organization and with covid a lot of our projects got pushed back.
Elizabeth
Thanks for the info @em9. Happy to hear an upgrade is planned for the new year! The fmf file won't log anything when this error occurs, but I was curious to see what was in there with regards to mappings for the oracle non-spatial reader. Given that the problem is intermittent, I think it may be data-related. Next time it occurs, I would examine the source data set being read to see if anything jumps out as different from the other records being read successfully.