ahmarsuhail opened a new pull request, #4298:
URL: https://github.com/apache/hadoop/pull/4298

   ### Description of PR
   
   This PR adds support for unbuffer. 
   
   Unbuffer is used by certain applications (eg: Impala) when they want to hold 
onto an input stream but free the resource it's using. This is useful as when 
it needs to read from the stream again, it doesn't have to open the stream 
again, and can save on HEAD calls.
   
   For prefetching, unbuffer needs to free up the buffer pool, delete any local 
files, clear state about blocks in the file etc. Also, when reading after an 
unbuffer, the input stream should reinitialise all this state. It should also 
read from the last active position before the read.
   
   ### How was this patch tested?
   
   Tested in eu-west-1 by running
   
   `mvn -Dparallel-tests -DtestsThreadCount=16 clean verify`
   
   `ITestS3AInputStreamPerformance` is failing, unrelated to this PR. Created 
issue: https://issues.apache.org/jira/browse/HADOOP-18231
   
   `ITestS3AUnbuffer` fails. instance of assertion & `isObjectStreamOpen()` 
fails.  Similar to the above issue, there are a few different ways to fix this 
test. I'm not sure what the best way is. Parameterized tests and different 
assertions based on if prefetching is enabled/new tests? I've left it failing 
for now. 
   
   All unbuffer contract tests are passing now. 


-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to