On 10/14/2015 04:24 PM, Andres Freund wrote:
> On 2015-10-14 16:17:55 +0300, Amir Rohan wrote:
>> it does fail the "dependent options" test:
>> $ postgres -C "archive_mode"
>> on
>> $ postgres -C wal_level
>> minimal
>
> Yea, because that's currently evaluated outside the config
> mechanism. It'd imo would be good to change that independent of this
> thread.
>
I agree.
>> 5) Because it checks syntax only, you don't get the benefits of having
>> an official place for the community to easily contribute rules that
>> warn you against config pitfalls, so that all users benefits.
>> See my OP for real-world examples and links about this problem.
>
> I don't think we as a community want to do that without review
> mechanisms in place, and I personally don't think we want to add
> separate processes for this.
>
That's what "contribute" means in my book. I'm getting mixed signals
about what the "community" wants. I certainly think if adding rules
involves modifying the postgres server code, that is far too high
a bar and no one will.
I'm not sure what you mean by "separate process". My original
pitch was to have this live in core or contrib, and no one wanted
it. If you don't want it in core, but people thinks its a good idea to
have (with review), what would you suggest?
Amir