[Dovecot] Allow_nets + MySQL failing when using range notation

[Javier García]

Hello,

Just a couple of lines to definitely confirm that the issue is present 
*only* on the sparc version of Debian while on a box equipped with an 
Intel Pentium processor everything works as expected.
Thanks a lot for your support.

Javier

Javier García escribió:
> x86, mmm..., in fact, I am testing on a sparc box, I will retry on a x86
> system as my production environment is x86. I cannot guess the effect of
> an architecture change, but maybe some library implementations might
> matter... I'll let you know.
> In the meantime, some more details on the actual installation:
>
> prisni:~# uname -a
> Linux prisni 2.6.18-6-sparc64 #1 Tue Feb 12 21:51:30 UTC 2008 sparc64
> GNU/Linux
>
> The installed packages along with dovecot are, in short, postfix, horde,
> apache2 and mysql.
>
> PS: While awaiting for moderator approval (this message was originally 
> too big) I have made a small test on my production environment,
>
> polifemo:~# uname -a
> Linux polifemo 2.6.18-5-686 #1 SMP Mon Dec 24 16:41:07 UTC 2007 i686 
> GNU/Linux
>
> which seems to work. So we could probably restrict the issue to linux 
> on sparc... quite a special scenario, in my opinion.
>
> As this latest test is just preliminary, I'll let you know about 
> further results.
>
> Regards,
> Javier
>
> Timo Sirainen escribió:
>> On Wed, 2008-05-07 at 18:15 +0200, Javier García wrote:
>>  
>>> I am afraid that I must come back with this issue. Following advice 
>>> from the Debian package maintainers, I installed a backported 1.0.13 
>>> version which keeps behaving wrongly. To be more specific:
>>>     
>>
>> Do you use x86 or something else? Just wondering if this could be
>> because of some endianess issue.
>>
>>  
>>> I wonder if this option is rare enough to this issue have remained 
>>> undiscovered through versions...
>>>     
>>
>> Could be, but it worked in my tests.. I guess it's also possible I
>> messed up my tests somehow.
>>
>>   
>
>
>