You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mxnet.apache.org by Nan Zhu <zh...@gmail.com> on 2017/08/13 14:35:14 UTC

how CI system works in MxNet?

Hi, all

I just noticed something which raises this question

Yesterday afternoon, I checked
https://github.com/apache/incubator-mxnet/commits/master and the first
commit passed all tests

However, when I checked again this morning, the test result was changed to
fail...

(one of my PRs also exposes this symptom)

Would anyone please share the insights about which part of our CI system
can produce this?

Best,

Nan

Re: how CI system works in MxNet?

Posted by Nan Zhu <zh...@gmail.com>.
I see, thanks

On Sun, Aug 13, 2017 at 12:32 PM, Ly Nguyen <ng...@gmail.com> wrote:

> As mentioned by Sheng this is an ongoing issue and a JIRA ticket has been
> filed to Apache Infra team, will follow up tomorrow if status hasn't been
> updated. https://issues.apache.org/jira/browse/INFRA-14840
>
> On Sun, Aug 13, 2017 at 9:21 AM, Nan Zhu <zh...@gmail.com> wrote:
>
> > no, I mean the CI system in our github repo......
> >
> > On Sun, Aug 13, 2017 at 8:06 AM, shiwen hu <ya...@gmail.com>
> wrote:
> >
> > > It may be that Sphinx cannot work incrementally
> > >
> > > 2017-08-13 23:05 GMT+08:00 shiwen hu <ya...@gmail.com>:
> > >
> > > > I think it was the last build of a residual file. Try make clean.
> first
> > > >
> > > > 2017-08-13 22:35 GMT+08:00 Nan Zhu <zh...@gmail.com>:
> > > >
> > > >> Hi, all
> > > >>
> > > >> I just noticed something which raises this question
> > > >>
> > > >> Yesterday afternoon, I checked
> > > >> https://github.com/apache/incubator-mxnet/commits/master and the
> > first
> > > >> commit passed all tests
> > > >>
> > > >> However, when I checked again this morning, the test result was
> > changed
> > > to
> > > >> fail...
> > > >>
> > > >> (one of my PRs also exposes this symptom)
> > > >>
> > > >> Would anyone please share the insights about which part of our CI
> > system
> > > >> can produce this?
> > > >>
> > > >> Best,
> > > >>
> > > >> Nan
> > > >>
> > > >
> > > >
> > >
> >
>

Re: how CI system works in MxNet?

Posted by Ly Nguyen <ng...@gmail.com>.
As mentioned by Sheng this is an ongoing issue and a JIRA ticket has been
filed to Apache Infra team, will follow up tomorrow if status hasn't been
updated. https://issues.apache.org/jira/browse/INFRA-14840

On Sun, Aug 13, 2017 at 9:21 AM, Nan Zhu <zh...@gmail.com> wrote:

> no, I mean the CI system in our github repo......
>
> On Sun, Aug 13, 2017 at 8:06 AM, shiwen hu <ya...@gmail.com> wrote:
>
> > It may be that Sphinx cannot work incrementally
> >
> > 2017-08-13 23:05 GMT+08:00 shiwen hu <ya...@gmail.com>:
> >
> > > I think it was the last build of a residual file. Try make clean. first
> > >
> > > 2017-08-13 22:35 GMT+08:00 Nan Zhu <zh...@gmail.com>:
> > >
> > >> Hi, all
> > >>
> > >> I just noticed something which raises this question
> > >>
> > >> Yesterday afternoon, I checked
> > >> https://github.com/apache/incubator-mxnet/commits/master and the
> first
> > >> commit passed all tests
> > >>
> > >> However, when I checked again this morning, the test result was
> changed
> > to
> > >> fail...
> > >>
> > >> (one of my PRs also exposes this symptom)
> > >>
> > >> Would anyone please share the insights about which part of our CI
> system
> > >> can produce this?
> > >>
> > >> Best,
> > >>
> > >> Nan
> > >>
> > >
> > >
> >
>

Re: how CI system works in MxNet?

Posted by Nan Zhu <zh...@gmail.com>.
no, I mean the CI system in our github repo......

On Sun, Aug 13, 2017 at 8:06 AM, shiwen hu <ya...@gmail.com> wrote:

> It may be that Sphinx cannot work incrementally
>
> 2017-08-13 23:05 GMT+08:00 shiwen hu <ya...@gmail.com>:
>
> > I think it was the last build of a residual file. Try make clean. first
> >
> > 2017-08-13 22:35 GMT+08:00 Nan Zhu <zh...@gmail.com>:
> >
> >> Hi, all
> >>
> >> I just noticed something which raises this question
> >>
> >> Yesterday afternoon, I checked
> >> https://github.com/apache/incubator-mxnet/commits/master and the first
> >> commit passed all tests
> >>
> >> However, when I checked again this morning, the test result was changed
> to
> >> fail...
> >>
> >> (one of my PRs also exposes this symptom)
> >>
> >> Would anyone please share the insights about which part of our CI system
> >> can produce this?
> >>
> >> Best,
> >>
> >> Nan
> >>
> >
> >
>

Re: how CI system works in MxNet?

Posted by shiwen hu <ya...@gmail.com>.
It may be that Sphinx cannot work incrementally

2017-08-13 23:05 GMT+08:00 shiwen hu <ya...@gmail.com>:

> I think it was the last build of a residual file. Try make clean. first
>
> 2017-08-13 22:35 GMT+08:00 Nan Zhu <zh...@gmail.com>:
>
>> Hi, all
>>
>> I just noticed something which raises this question
>>
>> Yesterday afternoon, I checked
>> https://github.com/apache/incubator-mxnet/commits/master and the first
>> commit passed all tests
>>
>> However, when I checked again this morning, the test result was changed to
>> fail...
>>
>> (one of my PRs also exposes this symptom)
>>
>> Would anyone please share the insights about which part of our CI system
>> can produce this?
>>
>> Best,
>>
>> Nan
>>
>
>

Re: how CI system works in MxNet?

Posted by shiwen hu <ya...@gmail.com>.
I think it was the last build of a residual file. Try make clean. first

2017-08-13 22:35 GMT+08:00 Nan Zhu <zh...@gmail.com>:

> Hi, all
>
> I just noticed something which raises this question
>
> Yesterday afternoon, I checked
> https://github.com/apache/incubator-mxnet/commits/master and the first
> commit passed all tests
>
> However, when I checked again this morning, the test result was changed to
> fail...
>
> (one of my PRs also exposes this symptom)
>
> Would anyone please share the insights about which part of our CI system
> can produce this?
>
> Best,
>
> Nan
>