Re: SQL access to database attributes

On 5/24/14, 8:14 AM, Tom Lane wrote:
> Vik Fearing<vik(dot)fearing(at)dalibo(dot)com> writes:
>> >On 05/24/2014 12:03 AM, Jaime Casanova wrote:
>>> >>Which lead us to the question: you need to connect to the database to
>>> >>modify it, don't you? then, how do you change ALLOW CONNECTIONS to
>>> >>true?
>> >You can ALTER DATABASE from anywhere.
> Perhaps it'd be wise to have a safety check to disallow turning off
> datallowconn for the last connectable database? Although it couldn't be
> bulletproof due to race conditions, so maybe that'd just be nannyism.
> (If you do shoot yourself in the foot that way, I think we ignore
> datallowconn in standalone mode.)

I think this is nannyism that would be well placed. Most people don't know about standalone, and I don't think we want to change that.

BTW, I think the race condition could be eliminated if we did something like (forgive the user-space semantics):

SELECT datallowconn FROM pg_database WHERE datallowconn AND datname <> $$database we're disallowing connections on$$ LIMIT 1 FOR UPDATE;

If you don't get a record back from that you abort; meanwhile no one else can disallow connections on that database until you commit or rollback.
--
Jim C. Nasby, Data Architect jim(at)nasby(dot)net
512.569.9461 (cell) http://jim.nasby.net