Nightly testing stalled?

[Farid Zaripov-2]

  The latest build results on 4.2.x branch generated at Mon Sep 8 15:00:25...

Farid.

[Scott Zhong-2]

http://svn.apache.org/repos/asf/stdcxx/branches/4.2.x not changed since
694682

stdcxx 4.2.x nightly builds only occur when a change has occurred over
the last run.

Scott Z.

> -----Original Message-----
> From: Farid Zaripov [mailto:Farid_Zaripov@...]
> Sent: Wednesday, September 17, 2008 7:23 AM
> To: dev@...
> Subject: Nightly testing stalled?
>
>   The latest build results on 4.2.x branch generated at Mon Sep 8
> 15:00:25...
>
> Farid.

[Scott Zhong-2]

It could also be that Martin is out until mid next week and the scripts
that generate the build page haven't been run yet.

Scott Z.

> -----Original Message-----
> From: Scott Zhong [mailto:Scott.Zhong@...]
> Sent: Wednesday, September 17, 2008 8:25 AM
> To: dev@...
> Subject: RE: Nightly testing stalled?
>
> http://svn.apache.org/repos/asf/stdcxx/branches/4.2.x not changed
since

> 694682
>
> stdcxx 4.2.x nightly builds only occur when a change has occurred over
> the last run.
>
> Scott Z.
>
> > -----Original Message-----
> > From: Farid Zaripov [mailto:Farid_Zaripov@...]
> > Sent: Wednesday, September 17, 2008 7:23 AM
> > To: dev@...
> > Subject: Nightly testing stalled?
> >
> >   The latest build results on 4.2.x branch generated at Mon Sep 8
> > 15:00:25...
> >
> > Farid.

[Martin Sebor]

Farid Zaripov wrote:
>   The latest build results on 4.2.x branch generated at Mon Sep 8 15:00:25...

I'm traveling this week and won't be back until next Wednesday
but unless it gets resolved in the meantime I'll look into it
as soon as I get back.

In the meantime, Travis, can you check to see if the builds
are running on our end?

Martin

>
> Farid.
>

[Martin Sebor]

Scott Zhong wrote:
> http://svn.apache.org/repos/asf/stdcxx/branches/4.2.x not changed since
> 694682
>
> stdcxx 4.2.x nightly builds only occur when a change has occurred over
> the last run.

AFAICS, rev 694682 was made on the 12th, but the page hasn't
been updated since the 8th. I get daily updates from the cron
job that updates it that it is up to date and doesn't need to
be refreshed. I think we'll need to dig in to the cron job on
people.apache.org to see what's going on.

http://svn.apache.org/viewvc?view=rev&revision=694682

Martin

>
> Scott Z.
>
>> -----Original Message-----
>> From: Farid Zaripov [mailto:Farid_Zaripov@...]
>> Sent: Wednesday, September 17, 2008 7:23 AM
>> To: dev@...
>> Subject: Nightly testing stalled?
>>
>>   The latest build results on 4.2.x branch generated at Mon Sep 8
>> 15:00:25...
>>
>> Farid.
>

[Travis Vitek-4]

 

Martin Sebor wrote:

>
>Farid Zaripov wrote:
>> The latest build results on 4.2.x branch generated at Mon
>> Sep 8 15:00:25...
>
>I'm traveling this week and won't be back until next Wednesday
>but unless it gets resolved in the meantime I'll look into it
>as soon as I get back.
>
>In the meantime, Travis, can you check to see if the builds
>are running on our end?
>

There are builds for every day since 9/10 with the exception of 9/15.
None of these builds appear on the public build log pages.

It is unclear to me exactly how the data from the build result files
gets out to Apache. It seems that they would need to be pushed out from
the Rogue Wave private network.

Travis

[Andrew Black-5]

Travis Vitek wrote:
> It is unclear to me exactly how the data from the build result files
> gets out to Apache. It seems that they would need to be pushed out from
> the Rogue Wave private network.

The process is run via Martin's crontab on one of the build machines
(dolly I believe).  The scripts used for the process reside in Perforce,
and I just checked in a change to one of them with change 246223 that
should fix the problem.  We recently migrated the systems used for the
majority of the builds from one domain to another, but hadn't updated
the domain reference in this script to reflect the change.  Looking at
the timestamp for the script, it looks like it's automatically updated
as part of the export process, so with a little luck we'll get a good
export tomorrow.

--Andrew Black

[Martin Sebor]

Andrew Black wrote:

> Travis Vitek wrote:
>> It is unclear to me exactly how the data from the build result files
>> gets out to Apache. It seems that they would need to be pushed out from
>> the Rogue Wave private network.
>
> The process is run via Martin's crontab on one of the build machines
> (dolly I believe).  The scripts used for the process reside in Perforce,
> and I just checked in a change to one of them with change 246223 that
> should fix the problem.  We recently migrated the systems used for the
> majority of the builds from one domain to another, but hadn't updated
> the domain reference in this script to reflect the change.  Looking at
> the timestamp for the script, it looks like it's automatically updated
> as part of the export process, so with a little luck we'll get a good
> export tomorrow.

Great, thanks for fixing it up! Let's keep an eye on the results
tomorrow to see if they've been refreshed.

Martin

>
> --Andrew Black
>