Thread: Very vague pg_dump question ...

Very vague pg_dump question ...

From
"Marc G. Fournier"
Date:
I'm currently looking into a problem that a client is reporting that 
pg_dump from 8.0.3 is 'skipping' one of their sequences ... I'm waiting 
for more info, but am curious if anyone knows (or can think of?) any 
reason why this might happen?  The only thing I can think of is that the 
sequence is owned by someone other then who the database is being dump'd 
as, and has no permissions to 'read' it ... but anything I'm not thinking 
of?

thanks ...

----
Marc G. Fournier           Hub.Org Networking Services (http://www.hub.org)
Email: scrappy@hub.org           Yahoo!: yscrappy              ICQ: 7615664


Re: Very vague pg_dump question ...

From
Tom Lane
Date:
"Marc G. Fournier" <scrappy@postgresql.org> writes:
> I'm currently looking into a problem that a client is reporting that 
> pg_dump from 8.0.3 is 'skipping' one of their sequences ... I'm waiting 
> for more info, but am curious if anyone knows (or can think of?) any 
> reason why this might happen?  The only thing I can think of is that the 
> sequence is owned by someone other then who the database is being dump'd 
> as, and has no permissions to 'read' it ... but anything I'm not thinking 
> of?

AFAIK, a permissions problem would result in a pretty obvious error
message from pg_dump ... though it's certainly possible for someone
to ignore that, especially if they are running pg_dump noninteractively.

Skipping in what sense --- no DDL, no setval, both?  Is this a
separately created sequence or a SERIAL sequence?
        regards, tom lane


Re: Very vague pg_dump question ...

From
Josh Berkus
Date:
Marc,

> I'm currently looking into a problem that a client is reporting that
> pg_dump from 8.0.3 is 'skipping' one of their sequences ... I'm waiting
> for more info, but am curious if anyone knows (or can think of?) any
> reason why this might happen?  The only thing I can think of is that the
> sequence is owned by someone other then who the database is being dump'd
> as, and has no permissions to 'read' it ... but anything I'm not
> thinking of?

I've encountered databases where a continuous upgrade from 7.2 can cause 
bad dependencies which make certain objects get dropped.   Usually running 
pg_depends fixes this.

-- 
--Josh

Josh Berkus
Aglio Database Solutions
San Francisco


Re: Very vague pg_dump question ...

From
"Marc G. Fournier"
Date:
On Fri, 8 Jul 2005, Tom Lane wrote:

> "Marc G. Fournier" <scrappy@postgresql.org> writes:
>> I'm currently looking into a problem that a client is reporting that
>> pg_dump from 8.0.3 is 'skipping' one of their sequences ... I'm waiting
>> for more info, but am curious if anyone knows (or can think of?) any
>> reason why this might happen?  The only thing I can think of is that the
>> sequence is owned by someone other then who the database is being dump'd
>> as, and has no permissions to 'read' it ... but anything I'm not thinking
>> of?
>
> AFAIK, a permissions problem would result in a pretty obvious error
> message from pg_dump ... though it's certainly possible for someone
> to ignore that, especially if they are running pg_dump noninteractively.
>
> Skipping in what sense --- no DDL, no setval, both?  Is this a
> separately created sequence or a SERIAL sequence?

This is what I'm still looking to find out ... all I got was "the sequence 
isn't being recreated in the dump", and when I message the client back, I 
get a vacation message, so obviously it wasn't a critical bug for them 
*roll eyes*

----
Marc G. Fournier           Hub.Org Networking Services (http://www.hub.org)
Email: scrappy@hub.org           Yahoo!: yscrappy              ICQ: 7615664