Re: Re: Proposed Windows-specific change: Enable crash dumps (like core files)
| От | Tom Lane |
|---|---|
| Тема | Re: Re: Proposed Windows-specific change: Enable crash dumps (like core files) |
| Дата | |
| Msg-id | 15764.1292539236@sss.pgh.pa.us обсуждение |
| Ответ на | Re: Re: Proposed Windows-specific change: Enable crash dumps (like core files) (Magnus Hagander <magnus@hagander.net>) |
| Список | pgsql-hackers |
Magnus Hagander <magnus@hagander.net> writes:
> Found another problem in it: when running with an older version of
> dbghelp.dll (which I was), it simply didn't work. We need to grab the
> version of dbghelp.dll at runtime and pick which things we're going to
> dump based on that.
> The attached version of the patch does that. Can you please test that
> it still generates the full dump on your system, which supposedly has
> a much more modern version of dbghelp.dll than mine?
This version of the patch looks fairly sane in terms of how it connects
to the rest of the code; but I'm unqualified to comment on the
Windows-specific code.
One thought is maybe the call point should be in startup_hacks() rather
than the main line of main()? I'm not terribly set on it if you don't
like that.
Also, I notice that the comment for startup_hacks() claims that it isn't
executed in subprocesses, which is evidently an obsolete statement ---
that should get cleaned up independently of this.
regards, tom lane
В списке pgsql-hackers по дате отправления: