Hackers, felt like reporting this relevant tidbit in case of interest... My site is among the few who've hit this bug.
We observed recently a case of pg_database having joined pg_authid and pg_auth_members in the bad xmin, unable to vacuum shcatalog group, however... pg_database then started working again without any intervention on our part so apparently due to some relevant system activity. We did later do a restart to correct the problem for those other 2 catalogs but , will try the global init file hack mentioned below later if/when we face this anew before running a fixed Pg version. Many thanks! Matheus de Oliveira <matioli.math...@gmail.com> writes: > Hello again... > > On Tue, Jun 19, 2018 at 12:53 PM, Andres Freund <and...@anarazel.de> > wrote: > > ... > > After that, yes, deleting the > global/pg_internal.init file is the way to go, and I can't think > of a > case where it's problematic, even without stopping the server. > > > > Just to let you know. I applied the work around in the affected > server and seemed to work way, so far no error. > > Thank you a lot for all the information and help. > > Best regards, > -- > Matheus de Oliveira > > > > -- Jerry Sievers Postgres DBA/Development Consulting e: postgres.consult...@comcast.net p: 312.241.7800
