Re: parallel.c is not marked as test covered

On Wed, Jun 15, 2016 at 4:59 PM, Peter Eisentraut
<peter(dot)eisentraut(at)2ndquadrant(dot)com> wrote:
> On 6/14/16 12:37 PM, Robert Haas wrote:
>> ERROR: can't generate random numbers because you haven't specified a seed
>>
>> ...to which the user will reply, "oh yes I did; in fact I ran SELECT
>> magic_setseed(42) just before I ran the offending query!". They'll
>> then contact an expert (hopefully) who will very possibly spend a lot
>> of time troubleshooting the wrong thing. If the message says:
>>
>> ERROR: can't generate random numbers because you haven't specified a seed
>> CONTEXT: parallel worker, PID 62413
>>
>> ...then the expert has a very good chance of guessing what has gone
>> wrong right away.
>
> My updated opinion is to keep the context but remove the PID. I can see
> that as this feature is being introduced, we will want to know that
> something is happening inside a parallel worker. Maybe in the future we'll
> have it all figured out and won't need the context anymore. But the PID is
> not useful by the time you see the message. The server log should contain a
> trace of the PIDs if you need to do deep digging.

I'm comfortable with that. Feel free to make it so, unless you want
me to do it for some reason.

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