ChaiBapchya commented on pull request #18622:
URL: https://github.com/apache/incubator-mxnet/pull/18622#issuecomment-651446932


   Based on the existing CI infra setup, both centos-gpu & unix-gpu run on the 
same AMI
   Same AMI translates essentially to same HW/SW.
   However, after the instance is created, it runs the CI on a docker file. 
That dockerfile setup is different from unix & centos
   
   Centos BUILD: 
https://github.com/apache/incubator-mxnet/blob/master/ci/docker/Dockerfile.build.centos7
   Centos Test: 
https://github.com/apache/incubator-mxnet/blob/master/ci/docker/Dockerfile.publish.test.centos7
   
   The variables used in these dockerfiles are defined in 
https://github.com/apache/incubator-mxnet/blob/master/ci/docker/docker-compose.yml
   
   For e.g. 
   In your case: 
http://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Fcentos-gpu/detail/PR-18622/9/pipeline
   Fails at 
   ```
   ci/build.py --docker-registry mxnetci --nvidiadocker --platform 
centos7_gpu_cu92 --docker-build-retries 3 --shm-size 500m 
/work/runtime_functions.sh unittest_centos7_gpu
   ```
   
   Thus the platform `centos7_gpu_cu92` defined here
   
   
https://github.com/apache/incubator-mxnet/blob/master/ci/docker/docker-compose.yml#L44-L53
   
   


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


Reply via email to