Did you find a work around for this?

Could it be class path ordering. I would expect the "file://..." protocol
to work when you have the MapR jars on the classpath..?

On Tue, Jan 20, 2015 at 4:36 AM, Ted Yu <[email protected]> wrote:

> Your classpath has some MapR jar.
>
> Is that intentional ?
>
> Cheers
>
> On Mon, Jan 19, 2015 at 6:58 AM, Jianguo Li <[email protected]>
> wrote:
>
>> Hi,
>>
>> I created some unit tests to test some of the functions in my project
>> which use Spark. However, when I used the sbt tool to build it and then ran
>> the "sbt test", I ran into "java.io.IOException: Could not create
>> FileClient":
>>
>> 2015-01-19 08:50:38,1894 ERROR Client
>> fs/client/fileclient/cc/client.cc:385 Thread: -2 Failed to initialize
>> client for cluster 127.0.0.1:7222, error Unknown error(108)
>> num lines: 21
>> [info] TextFileAdapterTestSuite:
>> [info] - Checking the RDD Vector Length *** FAILED ***
>> [info]   java.io.IOException: Could not create FileClient
>> [info]   at
>> com.mapr.fs.MapRFileSystem.lookupClient(MapRFileSystem.java:351)
>> [info]   at
>> com.mapr.fs.MapRFileSystem.lookupClient(MapRFileSystem.java:363)
>> [info]   at
>> com.mapr.fs.MapRFileSystem.getMapRFileStatus(MapRFileSystem.java:795)
>> [info]   at
>> com.mapr.fs.MapRFileSystem.getFileStatus(MapRFileSystem.java:822)
>> [info]   at
>> org.apache.hadoop.fs.FileSystem.getFileStatus(FileSystem.java:1419)
>> [info]   at
>> org.apache.hadoop.fs.FileSystem.globStatusInternal(FileSystem.java:1092)
>> [info]   at
>> org.apache.hadoop.fs.FileSystem.globStatus(FileSystem.java:1031)
>> [info]   at
>> org.apache.hadoop.mapred.FileInputFormat.listStatus(FileInputFormat.java:231)
>> [info]   at
>> org.apache.hadoop.mapred.FileInputFormat.getSplits(FileInputFormat.java:277)
>> [info]   at
>> org.apache.spark.rdd.HadoopRDD.getPartitions(HadoopRDD.scala:199)
>> [info]   ...
>>
>> The only tests failed, which I believe led to this exception are the ones
>> where my functions call the SparkContext's function textFile(). I tried to
>> debug this, and found that the exception seems to take place within the
>> textFile() function. Does anybody know what is the issue and how to fix it?
>> I used the local host for the SparkContext, does it have anything to do
>> with this exception.
>>
>>
>> Thanks,
>>
>> Jianguo
>>
>
>

Reply via email to