RACK_LOCAL Tasks Failed to finish

[洪奇 <qi...@alibaba-inc.com>]

Hi all,When running GraphX applications on Spark, task scheduler may schedule some tasks to be executed on RACK_LOCAL executors,but the tasks get halting in that case, repeating print the following log information:
14-07-14 15:59:14 INFO [Executor task launch worker-6] BlockFetcherIterator$BasicBlockFetcherIterator: Started 1 remote fetches in 3 ms
14-07-14 15:59:14 INFO [Executor task launch worker-1] BlockManager: Found block rdd_29_38 locally
14-07-14 15:59:14 INFO [Executor task launch worker-1] BlockManager: Found block rdd_29_38 locally
14-07-14 15:59:14 INFO [Executor task launch worker-1] BlockManager: Found block rdd_29_38 locally
14-07-14 15:59:14 INFO [Executor task launch worker-1] BlockFetcherIterator$BasicBlockFetcherIterator: maxBytesInFlight: 50331648, targetRequestSize: 10066329
14-07-14 15:59:14 INFO [Executor task launch worker-1] BlockFetcherIterator$BasicBlockFetcherIterator: Getting 300 non-empty blocks out of 300 blocks
14-07-14 15:59:14 INFO [Executor task launch worker-1] BlockFetcherIterator$BasicBlockFetcherIterator: Started 1 remote fetches in 3 ms
14-07-14 15:59:14 INFO [Executor task launch worker-0] BlockFetcherIterator$BasicBlockFetcherIterator: maxBytesInFlight: 50331648, targetRequestSize: 10066329
14-07-14 15:59:14 INFO [Executor task launch worker-0] BlockFetcherIterator$BasicBlockFetcherIterator: Getting 300 non-empty blocks out of 300 blocks
14-07-14 15:59:14 INFO [Executor task launch worker-0] BlockFetcherIterator$BasicBlockFetcherIterator: Started 1 remote fetches in 3 ms
14-07-14 15:59:14 INFO [Executor task launch worker-2] BlockManager: Found block rdd_29_2 locally
14-07-14 15:59:14 INFO [Executor task launch worker-2] BlockManager: Found block rdd_29_2 locally
14-07-14 15:59:14 INFO [Executor task launch worker-2] BlockManager: Found block rdd_29_2 locally
BlockManager's `get` and `getMultiple` are being called continually, and I don't know why. Are there some rdds being recomputed?Thanks for your help.Qiping

答复：RACK_LOCAL Tasks Failed to finish

[洪奇 <qi...@alibaba-inc.com>]

I just running PageRank(included in GraphX) on a dataset which has 55876487 edges. I submit the application to YARN with options`--num-executors 30 --executor-memory 30g --driver-memory 10g --executor-cores 8`.

Thanks------------------------------------------------------------------发件人：Ankur Dave <an...@gmail.com>发送时间：2014年7月15日(星期二) 11:52收件人：user <us...@spark.apache.org>，李奇平(洪奇) <qi...@alibaba-inc.com>主　题：Re: RACK_LOCAL Tasks Failed to finish
What GraphX application are you running? If it's a custom application that calls RDD.unpersist, that might cause RDDs to be recomputed. It's tricky to do unpersisting correctly, so you might try not unpersisting and see if that helps.

Ankur

Re: RACK_LOCAL Tasks Failed to finish

[Ankur Dave <an...@gmail.com>]

What GraphX application are you running? If it's a custom application that
calls RDD.unpersist, that might cause RDDs to be recomputed. It's tricky to
do unpersisting correctly, so you might try not unpersisting and see if
that helps.

Ankur <http://www.ankurdave.com/>