1 person found this helpful
We need to dive to much more granular level, to identify the cause for this kind of issue. Some analysis on the coredump (search for .dmp) produced by your windows OS using Windebug utility, running through console~obis1.log etc might have to be done to catch the culprit.
Note: I do not remember exactly, but there is are configuration steps available in my oracle support, to enable core dump on windows boxes (at least for Dr Watson though!!). You might want to have a look.
If you think there are any visible messages in the log, you could post here; I could help you read through them.
Thanks for the reply Dhar. I do have more details but left it fairly high level initially because it seemed like one of those things that might be instantly recognizable as a known issue. I'm going to try the 220.127.116.11.1 patch which was just released and see how that goes. If not I'll post back with more details.
I can confirm that thankfully the 18.104.22.168.1 patch(s) has resolved the issue. I think it was an internal/unpublished bug.
Just updating this to say 22.214.171.124 didn't resolve the issue. It continued to fail for some queries. Oracle haven't officially patched it yet afaik but we have managed to resolve it with their support.
Just thought i'd update this to say Oracle eventually gave us a one-off patch for this issue. Not sure when it will be officially fixed