Re: Constraint and Index with same name? (chicken and egg probelm) - Mailing list pgsql-general

From David Brain
Subject Re: Constraint and Index with same name? (chicken and egg probelm)
Date
Msg-id 460934D5.1010507@bandwidth.com
Whole thread Raw
In response to Re: Constraint and Index with same name? (chicken and egg probelm)  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: Constraint and Index with same name? (chicken and egg probelm)  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-general
Not sure I can provide one either I'm afraid, I don't think I did
anything unusual.  But I can provide you with some background as to how
this DB was created so if you can see any steps that may have been
problematic.

This db was created as part of an upgrade to new hardware, using Slony
to replicate the data to copy data and keep things in sync during the
initial test period.

1. Created basic db schema, minus indexes and constraints.
2. Setup Slony replication to copy/sync data from master db.
3. Once syc complete, re-added indexes and constraints to new DB using
pgadmin (by copying the ddl from the original db - again using pgadmin)
4. Removed FK constraint from the old db - worked ok.
5. Attempted to remove FK constraint from new DB - hit the problem I am
seeing now.

The db is now semi production, but I'll happily provide whatever info I
can short of needing a restart (and even a restart could be scheduled if
necessary).

Let me know if I can provide any more info.

David.

Tom Lane wrote:
> David Brain <dbrain@bandwidth.com> writes:
>> This could well be a recurrence of this issue:
>> http://archives.postgresql.org/pgsql-general/2007-01/msg01801.php
>> for which there doesn't seem to have been a resolution.
>
> I never got a reproduceable case out of the other reporter ... can you
> provide one?  His looked like it had something to do with foreign
> key constraints named the same as unique/primary key constraints,
> but AFAICS 8.1 won't let you create such, so I dunno how he got into
> that state.
>
>             regards, tom lane

pgsql-general by date:

Previous
From: "Kevin Field"
Date:
Subject: Re: cutting out the middleperl
Next
From: Oleg Bartunov
Date:
Subject: Re: Is there a shortage of postgresql skilled ops people