I’ve had a look at this with @kuaza because of his history with my plugin.
We removed the offending index from dump.sql that was causing the error above and restored it.
There was an updated script to drop this index but some users of the plugin may not have seen the update.
We now have a build up but there’s no connection from the browser.
We have the container up again without the offending index and all the old script in app.yml to build pgembedding has been removed, so it’s clean but with plugins.
the container is up - you can enter it np.
psql apparently up - you can bring up the database prompt within the container np.
within the container you can open rails c without issue.
/var/discourse/shared/standalone/log/rails# tail -n 200 production.log
does not show any obvious error messages.
I’m currently stumped.
Any ideas for other things to check?