jzgithub1 opened a new pull request #1112: Fixes #1052 - Review DistributedCache usage - Updated TokenFileIT URL: https://github.com/apache/accumulo/pull/1112 I tried to demonstrate that DistributedCaching is working and demonstrated how we can create a pseudonym for a cached file by putting at the end of a URI the # symbol followed by a pseudonym like in this example: job.addCacheFile(new URI("hdfs://localhost:8020/tokenfile.json#tokenfile")); I tested this new code with a running instance of Hadoop on port 8020 and it works. I had to remove much of the code that relies on this so the TokenFileIT will run and clean up properly during the build process. The system call: 'hdfs dfs -put tokenfilepath /' in the setTokenFileInHDFS function is vestigial now. So is the 'hdfs -rm call' . Since I had to test a real HDFS file system I have left this code in for further investigation in case someone wants to test it the way I did originally. The system calls in the annotated After cleanUpHDFS function are vestigial too but left in to demonstrate the steps taken to evaluate this issue of concern that distributed caching was not working as expected.
---------------------------------------------------------------- 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: [email protected] With regards, Apache Git Services
