[Pkg-javascript-devel] Bug#880936: nodejs: updated version available upstream

[Félix Sipma]

On 2017-12-14 17:45+0100, Jérémy Lal wrote:
>>
>> On 2017-12-13 14:55+0100, Félix Sipma wrote:
>>> On 2017-12-02 13:21+0100, Sébastien Villemot wrote:
>>>> On Sat, Dec 02, 2017 at 01:08:55PM +0100, Félix Sipma wrote:
>>>>> On 2017-11-05 21:27+0100, Jérémy Lal wrote:
>>>>>> 2017-11-05 21:19 GMT+01:00 Félix Sipma <felix+debian at gueux.org>:
>>>>
>>>>>>> node version 8.9.0 LTS is available upstream. Could you consider
>> packaging
>>>>>>> it?
>>>>
>>>>>> Sure, but nodejs 6.11.x first has to go in testing...
>>>>>
>>>>> Is there still something preventing 6.12.x from migrating to testing?
>>>>
>>>> Yes, it fails to build on mips and mipsel:
>>>>
>>>> https://buildd.debian.org/status/package.php?p=nodejs
>>>
>>> OK, thanks. More specifically, this test is failing:
>>>
>>> not ok 1219 parallel/test-zlib
>>> ---
>>> duration_ms: 154.551
>>> severity: fail
>>> stack: |-
>>>   buffer.js:11
>>>       super(arg1, arg2, arg3);
>>>       ^
>>>
>>>   RangeError: Array buffer allocation failed
>>>       at Buffer.Uint8Array (native)
>>>       at FastBuffer (buffer.js:11:5)
>>>       at createUnsafeBuffer (buffer.js:38:12)
>>>       at allocate (buffer.js:181:12)
>>>       at Function.Buffer.allocUnsafe (buffer.js:141:10)
>>>       at Function.Buffer.concat (buffer.js:309:23)
>>>       at InflateRaw.stream.Stream.stream.Stream.node.pipe.pipe.on.on.common.mustCall
>> (/<<PKGBUILDDIR>>/test/parallel/test-zlib.js:153:29)
>>>       at InflateRaw.<anonymous> (/<<PKGBUILDDIR>>/test/common/
>> index.js:445:15)
>>>       at emitNone (events.js:91:20)
>>>       at InflateRaw.emit (events.js:185:7)
>>> ...
>>>
>>> Could it be that the hosts ran out of RAM?
>>
> 2017-12-14 17:38 GMT+01:00 Félix Sipma <felix+debian at gueux.org>:
> 
>> It would be interesting to have 8.x in experimental to see if the build
>> still
>> fails. Who knows? We may be lucky :-).
>>
>> @Jérémy: could you consider uploading this?
> 
> If everything goes well i'll upload it in a few hours.
> 
> Jérémy

Great, thanks! I went on irc, and bremner told me that mips is architecture
limited to 2G per process, even if https://db.debian.org/machines.cgi (I didn't
knew this ressource!) says mips-sil-01.debian.org has 8G of RAM... Maybe a clue?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: not available
URL: <http://lists.alioth.debian.org/pipermail/pkg-javascript-devel/attachments/20171214/4d5a386e/attachment.sig>