Re: [HACKERS] unique index violation after pg_upgrade to PG10

On Tue, Oct 24, 2017 at 01:14:53PM -0500, Justin Pryzby wrote:
> I upgrade another instance to PG10 yesterday and this AM found unique key
> violations.
> 
> Our application is SELECTing FROM sites WHERE site_location=$1, and if it
> doesn't find one, INSERTs one (I know that's racy and not ideal).  We ended up
> with duplicate sites, despite a unique index.  We removed the duplicate rows
> and reindexed fine.  This is just a heads up with all the detail I can fit in a
> mail (but there's more if needed).
> ...
> Note:
> I run a script which does various combinations of ANALYZE/VACUUM (FULL/ANALYZE)
> following the upgrade, and a script runs nightly with REINDEX and pg_repack
> (and a couple of CLUSTER), so you should assume that any combination of those
> maintenance commands have been run.
> 
> In our reindex/repack log I found the first error due to duplicates:
> Tue Oct 24 01:27:53 MDT 2017: sites: sites_idx(repack non-partitioned)...
> WARNING: Error creating index "public"."index_61764": ERROR:  could not create unique index "index_61764"
> DETAIL:  Key (site_office, site_location)=(CRCLMT-DOEMS0, 1120) is duplicated.
> WARNING: Skipping index swapping for "sites", since no new indexes built
> WARNING: repack failed for "sites_idx"
> reindex: warning, dropping invalid/unswapped index: index_61764
> 

Hi Justin,

This sounds like a pg_repack bug and not a PostgreSQL bug. What version are
you running?

Regards,
Ken


-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers