Public SourceMage Discussion List

[Javier Vasquez <j.e.vasquez.v AT gmail.com>]

On Sun, Sep 1, 2013 at 12:02 AM, Javier Vasquez <j.e.vasquez.v AT gmail.com> 
wrote:
> On Mon, Aug 26, 2013 at 4:53 PM, Javier Vasquez <j.e.vasquez.v AT gmail.com> 
> wrote:
>> On Mon, Aug 26, 2013 at 11:25 AM, Remko van der Vossen <wich AT yuugen.jp> 
>> wrote:
>>> On Mon, Aug 26, 2013 at 08:30:47AM -0600, Javier Vasquez wrote:
>>>> On Mon, Aug 26, 2013 at 12:29 AM, Remko van der Vossen <wich AT yuugen.jp> 
>>>> wrote:
>>>> > Hmmn, interesting, I did not get any such problems. Though I have to 
>>>> > say I only tested with my own configuration which is with all optional 
>>>> > deps disabled. Could you try disabling optional deps to see if you can 
>>>> > find if any of them are the problem?
>>>> >
>>>> > Also, is the rest of your system up to date?
>>>> >
>>>> > Regards, Remko
>>>>
>>>> It's up to date...
>>>>
>>>> BTW, it failed on x86_64 laptop.  And it passed on a x86_32 one.
>>>
>>> The only systems I tested it on were x86_64
>>>
>>>> The failure is due to nss version according to the message, so I don't
>>>> believe it's the other deps.  I'll try again later, no idea what might
>>>> have gone wrong, :-)...
>>>
>>> Have you tried just recasting nss?
>>
>> Not yet, but that was going to be my next move any ways.  See, on the
>> x86_32 I casted firefox with "-r", which I didn't do with x86_64, and
>> that's the difference I can notice.  But I saw on x86_64 nss getting
>> re-casted automatically...
>>
>> ...
>
> Unfortunately FF 23.0.1 still refuses to install (it already compiles)
> on x86_64.  And I haven't figured out why, given it installed OK on
> x86_32, :-(
>
> I noticed 1 difference, but can't make any sense of it yet.
>
> On 32 bits (compiles+installs OK):
>
> ******************
> ...
> ******************
>
> While on 64 bits (compiles OK, but fails to install):
>
> ******************
> ...
> ******************
>
> Attached go the 2 logs in case someone can help identifying wha the
> issue might be (1st for 32 and 2nd for 64)...

That made me suspect of a race condition, so I tried:

mk_add_options MOZ_MAKE_FLAGS="-j1"

But that didn't work, and I got the same error, :-(  Just in case
someone might have thought of that as well...

Thanks,

-- 
Javier.