Re: Re: Proposed Windows-specific change: Enable crash dumps (like core files)

Magnus Hagander <magnus@hagander.net>

From: Magnus Hagander <magnus@hagander.net>
To: Robert Haas <robertmhaas@gmail.com>
Cc: Craig Ringer <craig@postnewspapers.com.au>, Tom Lane <tgl@sss.pgh.pa.us>, Heikki Linnakangas <heikki.linnakangas@enterprisedb.com>, Dave Page <dpage@pgadmin.org>, pgsql-hackers@postgresql.org
Date: 2010-12-16T14:24:22Z
Lists: pgsql-hackers
On Thu, Dec 16, 2010 at 15:07, Robert Haas <robertmhaas@gmail.com> wrote:
> On Thu, Dec 16, 2010 at 8:01 AM, Magnus Hagander <magnus@hagander.net> wrote:
>> We use the existance of the "crashdumps" directory as an indication we
>> want crashdumps. That's fine when the system is up. But what if we
>> crash *in the postmaster before we have done chdir()*?
>>
>> Should we perhaps instead define a subdirectory of *where the .EXE
>> file is*, and dump the file there?
>
> That seems pretty ugly, for a pretty marginal case.  The number of
> crashes that are going to happen in the postmaster before we've done
> chdir() should be extremely small, especially if we're talking about
> crashes in the field rather than during development.  How about adding
> a command-line option to force a dump to be written in $CWD whether a
> crashdumps directory exists or not?

Is that less ugly? ;)

But yes, we are talking about in the field, so it's fairly small. But
any crash during guc loading for example would go there, I think?

We can do such a commandline. We don't have any platform-specific
commandline options today. Is that something we've intentionally
avoided, or just not needed before?

-- 
 Magnus Hagander
 Me: http://www.hagander.net/
 Work: http://www.redpill-linpro.com/