[GitHub] [incubator-mxnet] leezu opened a new issue #19183: CI: Sanity step spends >50% of time on running filepermission update script

[GitBox <gi...@apache.org>]

leezu opened a new issue #19183:
URL: https://github.com/apache/incubator-mxnet/issues/19183


   ## Description
   Out of 8m 16s total runtime for Lint step, 5 minutes are spent on `RUN /work/docker_filepermissions.sh`
   
   ```
   [2020-09-18T22:58:53.540Z] Step 10/14 : COPY install/docker_filepermissions.sh /work/
   
   [2020-09-18T22:58:53.796Z]  ---> Using cache
   
   [2020-09-18T22:58:53.796Z]  ---> 7095374256c4
   
   [2020-09-18T22:58:53.796Z] Step 11/14 : RUN /work/docker_filepermissions.sh
   
   [2020-09-18T23:04:00.262Z]  ---> Running in 83961ba8efc8
   
   [2020-09-18T23:04:00.262Z] + [[ 1001 -gt 0 ]]
   
   [2020-09-18T23:04:00.262Z] + useradd -m --no-log-init --uid 1001 --system jenkins_slave
   
   [2020-09-18T23:04:00.262Z] + mkdir /work/mxnet
   
   [2020-09-18T23:04:00.262Z] + mkdir /work/build
   
   [2020-09-18T23:04:00.262Z] + chown -R jenkins_slave /work/
   ```
   
   See https://jenkins.mxnet-ci.amazon-ml.com/blue/organizations/jenkins/mxnet-validation%2Fsanity/detail/PR-19182/1/pipeline/39/
   
   This case may have been extreme (unlucky), but ideally we can figure out how to prevent it.
   
   This may be related to the fact that `RUN /work/docker_filepermissions.sh` is the first step that doesn't rely on the Docker cache.


----------------------------------------------------------------
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:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@mxnet.apache.org
For additional commands, e-mail: issues-help@mxnet.apache.org