<p>On Oct 28, 2011 5:22 AM, "Tom Lane" <<a href="mailto:tgl@sss.pgh.pa.us">tgl@sss.pgh.pa.us</a>> wrote:<br />
><br/> > Bruce Momjian <<a href="mailto:bruce@momjian.us">bruce@momjian.us</a>> writes:<br /> > >
StephenFrost wrote:<br /> > >> Yes, they would have removed it because they didn't want it. As I<br /> >
>>recall, part of the agreement to create an extra database by default was<br /> > >> that it could be
removedif users didn't want it. Turning around and<br /> > >> then saying "but things won't work if it's not
there"isn't exactly<br /> > >> supporting users who decide to remove it.<br /> ><br /> > > Well, you
wouldhave to remove it _after_ you did the pg_upgrade.<br /> ><br /> > As far as the *target* cluster is
concerned,I have no sympathy for<br /> > someone who messes with its contents before running pg_upgrade.<br /> >
That'san RTFM matter: you're supposed to upgrade into a virgin<br /> > just-initdb'd cluster.<br /> ><br /> >
However,it would be nice if pg_upgrade supported transferring from a<br /> > *source* cluster that didn't have the
postgresDB.<br /> ><br /> > What about creating a new, single-purpose database in the source<br /> > cluster
andthen removing it again after we're done?<p>How about naming this newly created database "postgres"? That would make
thecode simple enough - always use the postgres database, just drop it at the end if it didn't exist in the source
cluster.<p>/Magnus <br />