Close pysph RC bug #948736 (Was: Re: zarr and numcodecs sponsorship)

Antonio Valentino antonio.valentino at tiscali.it
Mon Feb 8 06:55:35 GMT 2021


Thanks Anton,

Il 07/02/21 20:46, Anton Gladky ha scritto:
> Hi Antonio,
> 
> uploaded!
> 
> Anton

It seems thet the fix works properly.
There are still a couple of buuilds missing in this moment,

ciao
antonio

> Am So., 7. Feb. 2021 um 10:54 Uhr schrieb Antonio Valentino <
> antonio.valentino at tiscali.it>:
> 
>> Dear Anton,
>> thanks a lot for sponsoring the previous upload.
>>
>> Now I finally received form upstream a patch that actually fixes #948736.
>> It is the lat RC bug remaining for pysph.
>>
>> I have prepared a new version of the package including the patch.
>> As for the previous upload the build in CI fails so I have set
>> distribution to experimental to have a chance to make a build in a more
>> realistic environment.
>>
>> In any case a local build on a s390x docker container seems to work
>> correctly.
>>
>> I would really appreciate if you could sponsor this upload also (I have
>> also put a request in SoB).
>>
>>
>> kind regards
>>
>> Antonio
>>
>>
>> Il 31/01/21 10:45, Antonio Valentino ha scritto:
>>> Dear Anton,
>>>
>>> Il 30/01/21 19:37, Anton Gladky ha scritto:
>>>> Hi Antonio,
>>>>
>>>> I have just tried to build pysph, but tests are failing. The same
>>>> is in CI-pipelines. Could you please check?
>>>
>>>
>>> The failure in CI is expected IMHO for i386 because
>>> libtrilinos-zoltan-dev an all dependent packages are not available on
>>> that platform.
>>>
>>> pysph builds successfully in my git-pbuilder on x86_64.
>>> I suspect that the problem in CI is due to the fact that at least to
>>> core are necessary to run parallel tests.
>>> Is is the case on you build environment?
>>> ... an most of all how many cores are available in a standard builder?
>>>
>>> The relevant part of the error message is the following IMHO:
>>>
>>> """
>>> There are not enough slots available in the system to satisfy the 2
>>> slots that were requested by the application:
>>>     /usr/bin/python3.9
>>> Either request fewer slots for your application, or make more slots
>>> available for use.
>>> A "slot" is the Open MPI term for an allocatable unit where we can
>>> launch a process.  The number of slots available are defined by the
>>> environment in which Open MPI processes are run:
>>>     1. Hostfile, via "slots=N" clauses (N defaults to number of
>>>        processor cores if not provided)
>>>     2. The --host command line parameter, via a ":N" suffix on the
>>>        hostname (N defaults to 1 if not provided)
>>>     3. Resource manager (e.g., SLURM, PBS/Torque, LSF, etc.)
>>>     4. If none of a hostfile, the --host command line parameter, or an
>>>        RM is present, Open MPI defaults to the number of processor cores
>>> In all the above cases, if you want Open MPI to default to the number
>>> of hardware threads instead of the number of processor cores, use the
>>> --use-hwthread-cpus option.
>>> Alternatively, you can use the --oversubscribe option to ignore the
>>> number of available slots when deciding the number of processes to
>>> launch.
>>> """
>>>
>>> Probably I could try to patch the code to force the number of slots to 2
>>> even on a single core machine but I'm not sure that it makes sense to
>>> test a parallel program in this conditions.
>>> The alternative could be to disable offending tests.
>>>
>>>
>>> Which is the most convenient option for you?
>>>
>>>
>>> kind regards
>>> antonio
>>>
>>>> Am Di., 8. Dez. 2020 um 09:26 Uhr schrieb Antonio Valentino <
>>>> antonio.valentino at tiscali.it>:
>>>>
>>>>> Hi Anton,
>>>>>
>>>>> Il 05/12/20 21:53, Anton Gladky ha scritto:
>>>>>> Am Sa., 5. Dez. 2020 um 21:46 Uhr schrieb Antonio Valentino
>>>>>> <antonio.valentino at tiscali.it>:
>>>>>>>
>>>>>>> Shall I remove the sponsor request form SoB until [1] is fixed?
>>>>>>>
>>>>>> Just let me know, when the package is in buildable state.
>>>>>
>>>>> Now numcodecs builds again.
>>>>> I have also bumped the Standards-Version and re-tagged.
>>>>>
>>>>> Please feel free to upload if you want.
>> --
>> Antonio Valentino
>>
>>
> 

-- 
Antonio Valentino



More information about the debian-science-maintainers mailing list