Sands, We did something a bit more through that dropping the Triggers (read on). What version of Postgres are you using Bill?
What happened is that there was a transition somewhere around postgres 7.3 from exposing "Constrains as Triggers" to hiding such details from the user. Yes, you should be able to drop these constraints. But you may find other similar triggers on other tables if you've been maintaining a DSpace instance long enough to have been on postgres before version 7.3. My ultimate solution that cleared up many of our issues was to dump (backup) the data separately from the database "schema". Then restore the database from a fresh new schema (same as the current version of DSpace you were previously using (but minus the INSERT statements that initialize certain data values). After that I reloaded the data. I documented the process here: http://wiki.dspace.org/index.php/ DSpace_Release_1.5.x_Documentation_for_Older_Database_Upgrades It was a rather challenging and dramatic process to do to the database, but the ultimate outcome is that any legacy weirdness caused by upgrading a dc cluster across may versions of postgres was eliminated by doing this. This issue made me aware that its important to keep the schema and any data inserts in separate sql files, I think we should rework the / etc/directory/*.sql files to support this. As I cooked up the solution on our systems, I would be glad to give you any advisement you might need to get through this issue. -Mark On Sep 10, 2008, at 9:23 PM, Sands Fish wrote: > Mark, we simply dropped these in production [EMAIL PROTECTED] when we > encountered this issue, right? > > > -- > sands fish > Software Engineer > MIT Libraries > Technology Research & Development > [EMAIL PROTECTED] > E25-131 > > > > > > On Sep 10, 2008, at 7:53 PM, Bill Jordan wrote: > >> >> We just upgraded to DSpace 1.5.0. When trying to add a new >> bitstream to an >> existing item, I get an org.postgresql.util.PSQLException: >> >> ERROR: constraint <unnamed>: table bundle does not have an attribute >> mets_bitstream_id >> >> I see that the mets_bitstream_id was removed from the bundle table >> as part of >> the schema changes for 1.5, but psql shows that there are still 3 >> RI_ConstraintTriggers in the system catalog referring to >> mets_bitstream_id -- 1 >> on bundle and 2 on bitstream. >> >> Is it safe to just drop those 3 triggers (when I did that on a >> test copy >> of the database, I was able to add the bitstream successfully)? >> >> --Bill >> >> **************************************************************** >> William Jordan >> Associate Dean >> University of Washington Libraries >> Resource Acquisition and Description/ >> Information Technology Services Box 352900, Seattle, WA 98195-2900 >> Voice: (206) 685-1625 Fax: (206) 543-5457 >> **************************************************************** >> >> --------------------------------------------------------------------- >> ---- >> This SF.Net email is sponsored by the Moblin Your Move Developer's >> challenge >> Build the coolest Linux based applications with Moblin SDK & win >> great prizes >> Grand prize is a trip for two to an Open Source event anywhere in >> the world >> http://moblin-contest.org/redirect.php?banner_id=100&url=/ >> _______________________________________________ >> DSpace-tech mailing list >> [email protected] >> https://lists.sourceforge.net/lists/listinfo/dspace-tech > ------------------------------------------------------------------------- This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/ _______________________________________________ DSpace-tech mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/dspace-tech
