Re: Superuser connect during smart shutdown

On Mon, Mar 23, 2015 at 11:02 AM, Kevin Grittner <kgrittn(at)ymail(dot)com> wrote:
> Jim Nasby <Jim(dot)Nasby(at)BlueTreble(dot)com> wrote:
>> On 3/20/15 9:44 AM, Kevin Grittner wrote:
>>> Robert Haas <robertmhaas(at)gmail(dot)com> wrote:
>>>> On Thu, Mar 19, 2015 at 10:42 PM, Bruce Momjian <bruce(at)momjian(dot)us> wrote:
>
>>>>> OK, are we up for changing the default pg_ctl shutdown method
>>>>> for 9.5, ("smart" to "fast"), [...]?
>>>>
>>>> I'm up for it. I think it's long overdue.
>>>
>>> +1
>>
>> +1, but I also like the idea of allowing SU to connect during a
>> smart shutdown. Even if you've intentionally chosen smart
>> instead of fast it still sucks that you can't find out what's
>> actually holding things up (and ps isn't that great a solution).
>
> I like that idea a lot, too. Having been in the position of
> remotely administering about 80 database servers, and getting a
> call that the building containing one of them was on fire, and the
> fire department would be arriving in two or three minutes to cut
> power to the building and start spraying water on everything, I
> found current behavior rather nervous-making as I struggled to get
> a clean shutdown of PostgreSQL followed by a clean shutdown and
> power-off of the server before that happened. The ability to make
> an SU connection during either "fast" or "smart" shutdown can be
> useful in a world of connection pools and long-running report
> queries. And fires.

That, however, is a separate issue from changing the default shutdown mode.

--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company