Reindexing isn't fully implemented and users/community members indexing doesn't work

[tmorrell]

Package version (if known): v12

Describe the bug

The current reindex commands in rdm-records and rdm-communities don't fully reindex requests and request events. In addition the users, groups, and community members bulk_index commands that were added two years ago don't seem to work. I'm also not sure how to re-initialize the usage stats indexes.

Steps to Reproduce

1. Set up a normal install, including adding users, communities, and submitting some requests
2. Delete indexes
3. Reindex records and communities
4. See that the requests index is blank.
5. Try to manually reindex users with

users_service = current_service_registry.get("users")
users_service.rebuild_index(system_identity)

7. See that the users index is blank

Expected behavior

There should be a recipe for re-indexing all content. A script based approach that works is https://github.com/caltechlibrary/caltechauthors/blob/main/scripts/reindex_more_stuff.py, but this isn't ideal. Requests should probably be included in community reindexing and the users/community members indexing is pretty slow.

[zzacharo]

Hey @tmorrell, have you tried using this command ?

[Samk13]

@zzacharo I added this and other rdm commands in the docs as they are not in the reference yet.
Please let me know if the documentation for this command is accurate and covers everything needed:
inveniosoftware/docs-invenio-rdm#608

[zzacharo]

@zzacharo I added this and other rdm commands in the docs as they are not in the reference yet. Please let me know if the documentation for this command is accurate and covers everything needed: inveniosoftware/docs-invenio-rdm#608

Thanks for that! I approved it!

[tmorrell]

The cli command should solve the requests issue. I need to test it, but I don't think it will solve the users issue.

[tmorrell]

This turned out to be a celery issue. I've added more details into the documentation, and this issue can be closed after merging.

[github-actions]

This issue was automatically marked as stale.