fredrik chabot <fredrik(at)f6(dot)nl> writes:
>> NOTICE: Buffer Leak:  (freeNext=-3, freePrev=-3, relname=dsrbk,
>> blockNum=0, flags=0x4, refcount=64 3)
That is a backend bug, without doubt. I fixed a number of problems that
could lead to that message a month or two ago. If you are running a
version older than 6.5.3 it'd be worth upgrading, I think. But there
were some fixes that I didn't risk committing into 6.5.* because they
were too complex to be sure about --- they will have to wait for the
7.0 betatest/release cycle.
> Log included, i know it isn't pretty sql code but it should work right?
Unfortunately this log is quite useless since you didn't supply table
declarations or test data needed to reproduce the problem. So I can't
be sure if my previous fixes take care of your problem or not...
regards, tom lane
pgsql-bugs by date
|Next:||From: Frank Cusack||Date: 1999-11-11 09:00:21|
|Subject: unique()ness not always maintained|
|Previous:||From: Oliver Fischer||Date: 1999-11-10 20:22:28|
|Subject: Re: [BUGS] No index support for date/time types?|