Thread: known bugs in sequences ?

known bugs in sequences ?

From
Henk Schets
Date:
Hi,

we just encountered this strange thing in our database.  All our sequences were reset to 1.  Is there any known bug
thatcould be related to this ?  Maybe with a jdbc driver or something ?  We are using version 7.0.3. 
It could very well be an exotic coding problem, but I want to be sure that it has nothing to do with our version of
postgresql.

thanks,

Henk




Re: known bugs in sequences ?

From
Tom Lane
Date:
Henk Schets <henk@poppunt.be> writes:
> we just encountered this strange thing in our database.  All our sequences were reset to 1.  Is there any known bug
thatcould be related to this ?  Maybe with a jdbc driver or something ?  We are using version 7.0.3. 
> It could very well be an exotic coding problem, but I want to be sure that it has nothing to do with our version of
postgresql.

All of your sequences at the same time?  It's really hard to see how any
internal bug in Postgres could cause that.  They're not stored together,
and there's no operation that visits them all.  Sure you didn't have some
client application run around and issue a lot of setvals?

But having said that, 7.0.3 is verging on ancient history, and we are
not fixing bugs in it anymore.  I think you are well overdue for an
update.

            regards, tom lane



Re: known bugs in sequences ?

From
"William N. Zanatta"
Date:
Somebody called 'Tom Lane' tried to say something! Take a look:
> Henk Schets <henk@poppunt.be> writes:
>
>>we just encountered this strange thing in our database.  All our sequences were reset to 1.  Is there any known bug
thatcould be related to this ?  Maybe with a jdbc driver or something ?  We are using version 7.0.3. 
>>It could very well be an exotic coding problem, but I want to be sure that it has nothing to do with our version of
postgresql.
>
>
> All of your sequences at the same time?  It's really hard to see how any
> internal bug in Postgres could cause that.  They're not stored together,
> and there's no operation that visits them all.  Sure you didn't have some
> client application run around and issue a lot of setvals?
>
> But having said that, 7.0.3 is verging on ancient history, and we are
> not fixing bugs in it anymore.  I think you are well overdue for an
> update.
>
>             regards, tom lane
>


   Could this problem be related with CYCLE in SEQUENCEs creation?

   William

--
Perl combines all of the worst aspects of BASIC, C and line noise.
                 -- Keith Packard