Re: Transaction-scope advisory locks

On 1/23/2011 4:24 AM, Robert Haas wrote:
> On Thu, Jan 20, 2011 at 5:22 AM, Marko Tiikkaja
> <marko(dot)tiikkaja(at)cs(dot)helsinki(dot)fi> wrote:
>> On 2011-01-17 9:28 AM +0200, Itagaki Takahiro wrote:
>>> == Coding ==
>>> I expect documentation will come soon.
>>
>> I'm sorry about this, I have been occupied with other stuff. I'm going to
>> work on this tonight.
>
> Any update on this?

Again, my apologies for the delay :-( Things haven't been going as
planned during the last few weeks.

Here's an updated patch with proposed doc changes. I still didn't
address the issue with pg_advisory_unlock_all() releasing transaction
scoped locks, but I'm going to. Another issue I found while testing the
behaviour here:
http://archives.postgresql.org/pgsql-hackers/2011-01/msg01939.php
is that if a session holds both a transaction level and a session level
lock on the same resource, only one of them will appear in pg_locks. Is
that going to be a problem from the user's perspective? Could it be an
indication of a well-hidden bug? Based on my tests it seems to work,
but I'm not at all confident with the code.

Regards,
Marko Tiikkaja