In hadoop 0.22, EventWriter ctor has this call:
this.encoder = new JsonEncoder(Event.SCHEMA$, out);
But in Avro 1.5.3 and 1.5.4, JsonEncoder() tor isn't public:
JsonEncoder(Schema sc, OutputStream out) throws IOException {
This explains the failure for Job initialization.
I seemed that JsonEncoder ctor was once public.
Cheers
On Fri, Oct 7, 2011 at 5:09 PM, Ted Yu <yuzhih...@gmail.com> wrote:
> Hi,
> For Avro 1.5.3, I saw the following when running TestHFileOutputFormat
> against hadoop 0.22:
>
> 2011-10-07 16:40:42,549 ERROR [pool-29-thread-1] mapred.JobTracker(3364):
> Job initialization failed:
> java.lang.IllegalAccessError: tried to access method
> org.apache.avro.io.JsonEncoder.<init>(Lorg/apache/avro/Schema;Ljava/io/OutputStream;)V
> from class org.apache.hadoop.mapreduce.jobhistory.EventWriter
> at
> org.apache.hadoop.mapreduce.jobhistory.EventWriter.<init>(EventWriter.java:57)
> at
> org.apache.hadoop.mapreduce.jobhistory.JobHistory.setupEventWriter(JobHistory.java:291)
> at
> org.apache.hadoop.mapred.JobInProgress.logSubmissionToJobHistory(JobInProgress.java:725)
> at
> org.apache.hadoop.mapred.JobInProgress.initTasks(JobInProgress.java:633)
> at org.apache.hadoop.mapred.JobTracker.initJob(JobTracker.java:3329)
> at
> org.apache.hadoop.mapred.EagerTaskInitializationListener$InitJob.run(EagerTaskInitializationListener.java:79)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
> at java.lang.Thread.run(Thread.java:680)
>
> 2011-10-07 16:40:43,432 INFO [main] hbase.HBaseTestingUtility(1200):
> Stopping mini mapreduce cluster...
> 2011-10-07 16:40:43,440 ERROR [Thread-196]
> filecache.TrackerDistributedCacheManager$CleanupThread(946): Exception in
> DistributedCache CleanupThread.
> java.lang.InterruptedException: sleep interrupted
> at java.lang.Thread.sleep(Native Method)
> at
> org.apache.hadoop.mapreduce.filecache.TrackerDistributedCacheManager$CleanupThread.run(TrackerDistributedCacheManager.java:943)
> 2011-10-07 16:40:43,441 INFO [main] log.Slf4jLog(67): Stopped
> SelectChannelConnector@0.0.0.0:0
> 2011-10-07 16:40:44,857 WARN [taskCleanup] mapred.TaskTracker(1977):
> Unknown job job_20111007164040532_0001 being deleted.
> 2011-10-07 16:40:44,857 ERROR [Thread-203] mapred.TaskTracker(1582): Caught
> exception: java.io.IOException: Jetty problem. Jetty didn't bind to a valid
> port
> at
> org.apache.hadoop.mapred.TaskTracker.checkJettyPort(TaskTracker.java:1387)
> at
> org.apache.hadoop.mapred.TaskTracker.offerService(TaskTracker.java:1560)
> at org.apache.hadoop.mapred.TaskTracker.run(TaskTracker.java:2418)
> at
> org.apache.hadoop.mapred.MiniMRCluster$TaskTrackerRunner.run(MiniMRCluster.java:229)
> at java.lang.Thread.run(Thread.java:680)
>
> Please advise which Avro version would allow Job initialization to succeed.
>
> Thanks
>
