From: | Magnus Hagander <magnus(at)hagander(dot)net> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Dave Page <dpage(at)pgadmin(dot)org>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Patch to add Windows 7 support |
Date: | 2009-01-27 15:25:52 |
Message-ID: | 497F2780.1060509@hagander.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Tom Lane wrote:
> Dave Page <dpage(at)pgadmin(dot)org> writes:
>> The attached patch adds support for the Windows 7 beta which we've had
>> a few reports of incompatibility with. When we startup using pg_ctl on
>> Windows, we create a job object (a logical grouping of processes on
>> Windows) to which we apply various security options. One of these
>> (JOB_OBJECT_UILIMIT_HANDLES) is used to prevent our processes seeing
>> handles belonging to processes outside of our job, however, when we
>> run under the service control manager, this causes the postmaster to
>> exit immeditately for no apparent reason.
>
>> I'm not entirely sure what has change in the SCM to cause this yet
>> (Windows 7 documentation is somewhat thin on the ground at the
>> moment), but the patch avoids theporblem by only setting
>> JOB_OBJECT_UILIMIT_HANDLES on earlier OSs.
>
> It would be good to understand what the problem actually is and what are
> the risks of running without this flag. I assume we put it in there
> for a reason.
Honestly, I think the reason was simply "enable all flags that we don't
know will break things for us". There was no closer analysis than that.
//Magnus
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2009-01-27 15:27:00 | Re: pgsql: Silence compiler warning on win32. |
Previous Message | Harald Armin Massa | 2009-01-27 15:23:18 | Re: pg_upgrade project status |