Обсуждение: propose web form for submission of performance problems
What would the list think of a web form for submitting problems the performance list, similar to the pgsql-bugs form? Alternately, or perhaps additionally, a script (hopefully bundled with postgres) which collects at least the non-query specific info and probably creates .logfile file for attachment. I assume fields would be mostly the content/questions off the SlowQuery wiki page, plus everything else asked with any frequency. There could also be "required" fields.. Justin
On Thu, May 24, 2018 at 4:57 PM, Justin Pryzby <pryzby@telsasoft.com> wrote:
What would the list think of a web form for submitting problems the performance
list, similar to the pgsql-bugs form?
Alternately, or perhaps additionally, a script (hopefully bundled with
postgres) which collects at least the non-query specific info and probably
creates .logfile file for attachment.
I assume fields would be mostly the content/questions off the SlowQuery wiki
page, plus everything else asked with any frequency.
There could also be "required" fields..
Is there something wrong with this email group? I actually to like it. Low tech, but it works. If you think such a web site would be good, you might explain the benefits over a simple email list like this one (which is archived and easily searchable via Google).
Having watched and participated in this mailing list since around 2006, my guess is that well over 95% of the "problems" are people who are learning, and just need advice for tuning their database and/or rewriting their queries. It is rare that someone reports a genuine performance problem in Postgres.
I like watching this mailing list; it gives a sense of the ebb and flow of problem reports, and when one comes along that interests me, I can easily follow it. And occasionally, I even answer questions. I probably wouldn't do that with a bug-reporting type of web site.
It would also be bad to have both an email list and a separate web site. That would simply split the community.
Craig
Justin
On Thu, May 24, 2018 at 06:27:31PM -0700, Craig James wrote:
> On Thu, May 24, 2018 at 4:57 PM, Justin Pryzby <pryzby@telsasoft.com> wrote:
>
> > What would the list think of a web form for submitting problems the
> > performance
> > list, similar to the pgsql-bugs form?
> >
> > Alternately, or perhaps additionally, a script (hopefully bundled with
> > postgres) which collects at least the non-query specific info and probably
> > creates .logfile file for attachment.
> >
> > I assume fields would be mostly the content/questions off the SlowQuery
> > wiki
> > page, plus everything else asked with any frequency.
> >
> > There could also be "required" fields..
> >
>
> Is there something wrong with this email group? I actually to like it. Low
I meant something exactly like the bug form:
https://www.postgresql.org/account/login/?next=/account/submitbug/
https://www.postgresql.org/list/pgsql-bugs/2018-05/
"Entering a bug report this way causes it to be mailed to the
<pgsql-bugs@postgresql.org> mailing list."
The goal was to continue the list but encourage problem reports to include all
of the most commonly-needed(missing) information, and in a standard format too.
But actually I see that nowadays it requires an account, which isn't great and
might discourage many people from using the web form. But maybe that's better
for some people than requiring subscription to the list.
Justin
Craig James schrieb am 25.05.2018 um 03:27: > What would the list think of a web form for submitting problems the performance > list, similar to the pgsql-bugs form? > > Alternately, or perhaps additionally, a script (hopefully bundled with > postgres) which collects at least the non-query specific info and probably > creates .logfile file for attachment. > > I assume fields would be mostly the content/questions off the SlowQuery wiki > page, plus everything else asked with any frequency. > > There could also be "required" fields.. > > Is there something wrong with this email group? I actually to like > it. Low tech, but it works. If you think such a web site would be > good, you might explain the benefits over a simple email list like > this one (which is archived and easily searchable via Google). > I don't think anything is wrong with the list, and I assume neither does Justin. But occasionally the bug submit form is misused for troubleshooting performance problems "Bug #42: Query is too slow" I could imagine that having a separate form to report performance problems and asking for different pieces of information which then sends that to this list (instead of bugs) isn't a bad idea. However I am not sure that the frequency of the bug-list misuse warrants that. Thomas