Обсуждение: Re: pgsql: add EPERM to the list of return codes to expect from opening

This isn't working. The Windows ports are all saying "permission
denied" but apparently that's not because errno is set to EPERM.
Anyone know how to detect "permission denied" errors from open() on
windows?

On Mon, Mar 1, 2010 at 12:04 AM, Greg Stark <stark@postgresql.org> wrote:
> Log Message:
> -----------
> add EPERM to the list of return codes to expect from opening directories based on Vista results
>
> Modified Files:
> --------------
>    pgsql/src/port:
>        copydir.c (r1.34 -> r1.35)
>        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/port/copydir.c?r1=1.34&r2=1.35)
>
> --
> Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-committers
>



--
greg

You might want to look at _dosmaperror() in src/port/win32error.c - it
contains the different win32 error ccodes that we match to EACCESS. I
don't see us mapping *anything* to EPERM.

Actually, I think that may be your problem - you are testing against
EPERM instead of EACCESS. On my linux manpage, EPERM isn't even a
valid return code from open().

//Magnus

2010/3/1 Greg Stark <gsstark@mit.edu>:
> This isn't working. The Windows ports are all saying "permission
> denied" but apparently that's not because errno is set to EPERM.
> Anyone know how to detect "permission denied" errors from open() on
> windows?
>
> On Mon, Mar 1, 2010 at 12:04 AM, Greg Stark <stark@postgresql.org> wrote:
>> Log Message:
>> -----------
>> add EPERM to the list of return codes to expect from opening directories based on Vista results
>>
>> Modified Files:
>> --------------
>>    pgsql/src/port:
>>        copydir.c (r1.34 -> r1.35)
>>        (http://anoncvs.postgresql.org/cvsweb.cgi/pgsql/src/port/copydir.c?r1=1.34&r2=1.35)
>>
>> --
>> Sent via pgsql-committers mailing list (pgsql-committers@postgresql.org)
>> To make changes to your subscription:
>> http://www.postgresql.org/mailpref/pgsql-committers
>>
>
>
>
> --
> greg
>
> --
> Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-hackers
>



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

Magnus Hagander <magnus@hagander.net> writes:
> Actually, I think that may be your problem - you are testing against
> EPERM instead of EACCESS. On my linux manpage, EPERM isn't even a
> valid return code from open().

Yeah, I had just come to the same conclusion upon seeing the buildfarm
still pink this morning.  On my old HPUX box these codes are defined as

/usr/include/sys/errno.h:#define        EPERM           1       /* Not super-user               */
/usr/include/sys/errno.h:#define        EACCES          13      /* Permission denied            */

so EPERM seems like the wrong mapping for such an error anyway.

BTW, in case anyone with admin privileges is paying attention, the
buildfarm (a) is about two hours off on its system clock again,
and (b) hasn't sent out a daily status-change summary email since
Friday.
        regards, tom lane

So fwiw Narwhal says EACCESS is working.

On Mon, Mar 1, 2010 at 10:01 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> BTW, in case anyone with admin privileges is paying attention, the
> buildfarm (a) is about two hours off on its system clock again,
> and (b) hasn't sent out a daily status-change summary email since
> Friday.

Eh?  The buildfarm sends out a daily status-change summary email?  To where?

...Robert

Robert Haas wrote:
> On Mon, Mar 1, 2010 at 10:01 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>   
>> BTW, in case anyone with admin privileges is paying attention, the
>> buildfarm (a) is about two hours off on its system clock again,
>> and (b) hasn't sent out a daily status-change summary email since
>> Friday.
>>     
>
> Eh?  The buildfarm sends out a daily status-change summary email?  To where?
>
>
>   

To subscribers of the relevant mailing lists. See 
<http://pgfoundry.org/mail/?group_id=1000040>

cheers

andrew

Robert Haas <robertmhaas@gmail.com> writes:
> On Mon, Mar 1, 2010 at 10:01 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> BTW, in case anyone with admin privileges is paying attention, the
>> buildfarm (a) is about two hours off on its system clock again,
>> and (b) hasn't sent out a daily status-change summary email since
>> Friday.

> Eh?  The buildfarm sends out a daily status-change summary email?  To where?

See
http://pgfoundry.org/mail/?group_id=1000040

I'm subscribed to pgbuildfarm-status-green ... and the archives for
it match my local log, which says there hasn't been a message since
Friday.  That's definitely not for lack of changes.  Now that I look
at the archives, it looks like all the status lists stopped getting
mail around that time.
        regards, tom lane

Tom Lane wrote:
>> Eh?  The buildfarm sends out a daily status-change summary email?  To where?
>>     
>
> See
> http://pgfoundry.org/mail/?group_id=1000040
>
> I'm subscribed to pgbuildfarm-status-green ... and the archives for
> it match my local log, which says there hasn't been a message since
> Friday.  That's definitely not for lack of changes.  Now that I look
> at the archives, it looks like all the status lists stopped getting
> mail around that time.
>
>             
>   

There is a DNS failure on the server causing a huge backlog of status 
messages. I have sent a message to the admins list about it. (I can see 
the list of errors on the machine by running mailq, but I can't do 
anything about it.)

cheers

andrew

Greg Stark wrote:
> So fwiw Narwhal says EACCESS is working.
>
>   

dawn_bat is also working. Both of these build using Mingw/gcc, not MSVC.

cheers

andrew