reef-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mariia Mykhailova <mamyk...@microsoft.com.INVALID>
Subject RE: Disable tests on OpenJDK 7?
Date Mon, 30 Oct 2017 23:11:21 GMT
We've started using sudo:false as was suggested at some point (about 2 years ago) by Travis
as part of migration to new infrastructure (from what they call "legacy environment" to container-based
environment). There was no effect on our tests.

Now they're doing a new migration in which a combination of sudo:false (use container-based
environment with Docker) and environment:precise will not be supported anymore. Looks like
sudo:required (which is what they're using with precise starting September) breaks our tests,
though I didn't look into how. Eventually environment:precise itself will be removed, so we
need to figure out what to switch to.

I've checked that environment:trusty (replacement for precise) with sudo:false runs our tests
successfully. The only catch is that it doesn't have Oracle JDK 1.7 anymore, so if we switch
to it we'll have to limit ourselves with just Oracle JDK 1.8 and OpenJDK 1.7.

Shall we switch and drop Oracle JDK 1.7 build job?

-Mariia

-----Original Message-----
From: Markus Weimer [mailto:markus@weimo.de] 
Sent: Monday, October 30, 2017 2:21 PM
To: REEF Developers Mailinglist <dev@reef.apache.org>
Subject: Re: Disable tests on OpenJDK 7?

On Mon, Oct 30, 2017 at 11:53 AM, Mariia Mykhailova < mamykhai@microsoft.com.invalid>
wrote:

> I estimated timeframe looking at the history of our commits at
> https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithu
> b.com%2Fapache%2Freef%2Fcommits%2Fmaster&data=02%7C01%7Cmamykhai%40mic
> rosoft.com%7Cf881f9ad032149793ece08d51fdc25e7%7C72f988bf86f141af91ab2d
> 7cd011db47%7C1%7C0%7C636449952746658052&sdata=tlpHSaRR4GrRxhAZW6PiOAfs
> iYIkHvQ%2B3oPedMmvO6s%3D&reserved=0 - commit e0634a4 (netty
> dependency) on Sep 2 has successful Travis build, commit 6489d43 
> (driver side interfaces) on Sep 8 has Travis failure, and ever since 
> Travis has been failing on master.
>

Thanks! I think we can rule out changes on our side then. There do seem to be some changes
at that time on the Travis side[0]:

we will begin ignoring sudo: false when we find an explicit dist: precise.


Does anyone know whether we relied on sudo:false in our tests? We do set it...

Markus
Mime
View raw message