Re: [HACKERS] [bug fix] dblink leaks unnamed connections - Mailing list pgsql-hackers

From Joe Conway
Subject Re: [HACKERS] [bug fix] dblink leaks unnamed connections
Date
Msg-id 92fd3d0b-f986-9cd6-068e-9d305760674d@joeconway.com
Whole thread Raw
In response to Re: [HACKERS] [bug fix] dblink leaks unnamed connections  (Joe Conway <mail@joeconway.com>)
List pgsql-hackers
On 03/09/2017 08:31 AM, Joe Conway wrote:
> On 03/09/2017 07:54 AM, Tom Lane wrote:
>> Fujii Masao <masao.fujii@gmail.com> writes:
>>> On Wed, Mar 8, 2017 at 3:52 PM, Tsunakawa, Takayuki
>>> <tsunakawa.takay@jp.fujitsu.com> wrote:
>>>> dblink fails to close the unnamed connection as follows when a new unnamed connection is requested.  The attached
patchfixes this. 
>>
>>> This issue was reported about ten years ago and added as TODO item.
>>> http://archives.postgresql.org/pgsql-hackers/2007-10/msg00895.php
>>
>>> I agree that this is a bug, and am tempted to back-patch to all the supported
>>> versions. But it had not been fixed in many years since the first report of
>>> the issue. So I'm not sure if it's ok to just treat this as a bug right now and
>>> back-patch. Or we should fix this only in HEAD? Anyway I'd like to hear
>>> more opinions about this.
>>
>> It looks to me like the issue simply fell through the cracks because Joe
>> wasn't excited about fixing it.  Now that we have a second complaint,
>> I think it's worth treating as a bug and back-patching.
>>
>> (I've not read this particular patch and am not expressing an opinion
>> whether it's correct.)
>
> Ok, will take another look.

I pushed a fix to all supported branches.

Thanks,

Joe

--
Crunchy Data - http://crunchydata.com
PostgreSQL Support for Secure Enterprises
Consulting, Training, & Open Source Development


pgsql-hackers by date:

Previous
From: Tom Lane
Date:
Subject: Re: [HACKERS] \if, \elseif, \else, \endif (was Re: PSQL commands: \quit_if, \quit_unless)
Next
From: Eric Ridge
Date:
Subject: Re: [HACKERS] How to get the 'ctid' from a record type?