Odd sword DIP upload behaviour

[John Mackey]

Hey all, 

We've got a weird situation with the sword plugin that I can't seem to crack. When using upload DIP to AtoM (2.8) in a fresh install of the latest stable Archivematica, the entire process 'appears' to work from the logs side, but ultimately nothing actually happens and no objects are linked to the target record. 

1. AM reports success and that AtoM is picking up in the job scheduler 
2. AtoM reports the job was received, slug is matched, METS found correctly and processing complete
   1. Cleanup fails, despite the fact we have implemented the tmp fix and the AtoM report indicating it is using the new directory, but I do not believe this to be related to the core issue. 
3. Job is complete (though technically errored as above), but no digital objects are linked to the target description and the target remains entirely unaffected. Standard remedial actions (cache, reindex, restart) do not change the result. 

Similar to the reports in this thread: https://groups.google.com/u/1/g/ica-atom-users/c/S4AbYSEUbjE/m/dEd7SXm3BQAJ

I don't believe this to be related to the indexing bug, we have tested reindex and my understanding is that even if the index were the problem, we would still be able to see the linked object from the record page itself. 

I also don't necessarily think the cleanup issue is related: when using the sword API manually to send the DIP data directly in a request, the AtoM job reports complete success with the same result as above (nothing happens):

I'm quite stumped. I've trawled through the code related to the sword worker and everything seems to follow the logic on that side. The only thing I can think to do next is add additional logging to the php to see if I can track down where it falls over.

I noticed no report appears to have been made off the back of the previous linked post, so I figured a new post was appropriate and I'd be happy to submit a report on Github. If anyone else is able to try replicate this behaviour that would be marvelous. 

Thanks!

John.

[Dan Gillean]

Hi John, 

Yes, if you are willing to file an initial bug report, that would be very helpful. I can ensure that our Maintainers see the issue and attempt to reproduce it here. 

Thanks for the detailed information - hopefully we can get to the bottom of this sooner than later!

Cheers, 

Dan Gillean, MAS, MLIS
AtoM Program Manager
Artefactual Systems, Inc.
604-527-2056

@accesstomemory

he / him

--
You received this message because you are subscribed to the Google Groups "AtoM Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ica-atom-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/ica-atom-users/3f07746f-00c3-4a0c-935f-367dde8ab92cn%40googlegroups.com.