Hi Karlen and all,
To be completely transparent about developer discussions.. as of today, I don't know what the future of Oracle support is for DSpace. So far, in all my outreach,
I've had just two responses -- yours and a private response from a service provider (who doesn't want to be named yet) who
might be able to help some in mid to late 2022. This lack of response implies to me that very few institutions may be using DSpace + Oracle these days.
So, in all honesty, if I were in your shoes, I'd start to advocate for migrating to PostgreSQL *or* advocate for someone to help DSpace fix its Oracle support.
Currently, DSpace doesn't have great Oracle support simply because I've been unable to find anyone to help maintain/support/fix it. I don't have a developer team, and I'm fully reliant on community volunteers and/or service providers to help improve & maintain DSpace. The community developers who used to help us maintain Oracle support left the DSpace community years ago (all moved on to new roles/responsibilities). While the remaining team of volunteers has done their best to keep Oracle support "working", it's obvious we're not great at it (based on Oracle-specific issues like the one you point out in both DSpace 6.x and now 7.x).
In other words, all currently active DSpace developers use PostgreSQL. None of our active developers use Oracle or have Oracle support expertise. This means that Oracle-specific bugs take much much longer to resolve, as I have no one available to help resolve them.
I wish I had better news for those who still use Oracle. But, without help, I'm not sure whether DSpace can continue to support Oracle going forward.
Nonetheless, no final decision has been made as of yet by our DSpace Steering Group. So, if anyone out there is interested in helping, please do reach out. It may be the deciding factor in what comes next for DSpace's Oracle support.
Tim