Re: pgadmin or pgsql bug? inserting row - Mailing list pgadmin-support

From Dave Page
Subject Re: pgadmin or pgsql bug? inserting row
Date
Msg-id 46BB14C7.2040104@postgresql.org
Whole thread Raw
In response to pgadmin or pgsql bug? inserting row  (Kevin Field <kev@brantaero.com>)
Responses Re: pgadmin or pgsql bug? inserting row  (Kevin Field <kev@brantaero.com>)
List pgadmin-support
Kevin Field wrote:
> One other detail, after it worked I noticed that the ID returned was one
> where there had been a gap in the sequence (i.e., we had entries 1-12
> and 14-24, and the new row had id 13.)  This may be because the existing
> entries had been added with explicit id numbers but the sequence was
> probably still at a lower number.
> 
> This is pretty befuddling to me.  I'm not sure if it's a pgsql issue or
> a pgadmin one, but anyway, there's my report.

Check the current value of the sequence used by the serial column, and
if it's less than the highest value in the table, you're almost
certainly trying to insert a duplicate. Each time you try though, the
sequence increments whether or not the insert worked - so by chance it
was OK in your app when you hit the gap.

The solution would be to set the current value of the sequence to
something higher than any record already in the table.

Regards, Dave


pgadmin-support by date:

Previous
From: Kevin Field
Date:
Subject: pgadmin or pgsql bug? inserting row
Next
From: Kevin Field
Date:
Subject: operators, types, etc....where'd they go?