And when you added the list's user email you had a typo which I copy pasted and the email never reached the list - So I am just forwarding again to the correct address this time! :-)
On Thu, Oct 23, 2014 at 5:36 PM, Stratos Dimopoulos < [email protected]> wrote: > Yes Jeff correct - this is the way I compile. Here is the stack trace > close to the error on the executor - I don't think the log file on the BSP > master would help you at all but let me know if you need to send you > anything else: > > I1022 17:25:31.828336 22440 exec.cpp:132] Version: 0.20.0 > I1022 17:25:31.838382 22454 exec.cpp:206] Executor registered on slave > 20141019-011259-1257767434-5050-4023-1 > 14/10/22 17:25:31 INFO bsp.MesosExecutor: Executor registered with the > slave > 14/10/22 17:25:31 INFO bsp.MesosExecutor: Launching task : Task_0 > 14/10/22 17:25:32 INFO bsp.GroomServer: groom start > 14/10/22 17:25:32 INFO bsp.GroomServer: > /mnt/mesos-work-dir/slaves/20141019-011259-1257767434-5050-4023-1/frameworks/20141019-011259-1257767434-5050-4023-0031/executors/executor_Task_0/runs/d69f21af-9180-4c0c-9263-82038ae81937/hama-0.7.0-SNAPSHOT/bsp/local > 14/10/22 17:25:32 WARN util.NativeCodeLoader: Unable to load native-hadoop > library for your platform... using builtin-java classes where applicable > 14/10/22 17:25:33 ERROR bsp.MesosExecutor: Caught exception, committing > suicide. > java.util.ServiceConfigurationError: org.apache.hadoop.fs.FileSystem: > Provider org.apache.hadoop.fs.LocalFileSystem not found > at java.util.ServiceLoader.fail(ServiceLoader.java:231) > at java.util.ServiceLoader.access$300(ServiceLoader.java:181) > at > java.util.ServiceLoader$LazyIterator.next(ServiceLoader.java:365) > at java.util.ServiceLoader$1.next(ServiceLoader.java:445) > at > org.apache.hadoop.fs.FileSystem.loadFileSystems(FileSystem.java:2364) > at > org.apache.hadoop.fs.FileSystem.getFileSystemClass(FileSystem.java:2375) > at > org.apache.hadoop.fs.FileSystem.createFileSystem(FileSystem.java:2392) > at org.apache.hadoop.fs.FileSystem.access$200(FileSystem.java:89) > at > org.apache.hadoop.fs.FileSystem$Cache.getInternal(FileSystem.java:2431) > at org.apache.hadoop.fs.FileSystem$Cache.get(FileSystem.java:2413) > at org.apache.hadoop.fs.FileSystem.get(FileSystem.java:368) > at org.apache.hadoop.fs.FileSystem.getLocal(FileSystem.java:339) > at > org.apache.hama.bsp.GroomServer.deleteLocalFiles(GroomServer.java:483) > at org.apache.hama.bsp.GroomServer.initialize(GroomServer.java:321) > at org.apache.hama.bsp.GroomServer.run(GroomServer.java:860) > at org.apache.hama.bsp.MesosExecutor$1.run(MesosExecutor.java:92) > > thanks, > Stratos > > On Thu, Oct 23, 2014 at 2:51 PM, Jeff Fenchel <[email protected]> wrote: > >> Fyi... Sorry I forgot to add your email to list. Please add >> [email protected] to your reply. >> >> Thanks, >> Jeff >> >> ---------- Forwarded message ---------- >> From: *Jeff Fenchel* <[email protected]> >> Date: Thursday, October 23, 2014 >> Subject: Using hama with Mesos 0.20 and CDH5.1.2 >> To: [email protected] >> >> >> Hey Stratos, >> >> Thanks for giving Hama on Mesos a try. I do not think it has ever been >> tested with hadoop 2, but I see no reason why it shouldn't work. So, I am >> >> guessing you compiled with something like the following? >> >> mvn clean install -Phadoop2 -Dhadoop.version=2.3.0-cdh5.1.2 >> -Dmesos.version=0.20.0 -DskipTests >> >> If so, can you please send us the complete stack trace? It does look >> like this might be a bug in hama. >> >> Thanks, >> -Jeff >> >> >> >> >> On Thu, Oct 23, 2014 at 1:13 AM, Stratos Dimopoulos >> <[email protected]> wrote: >> > Hi, >> > >> > I cloned Hama from Git and installed in a Eucalyptus cloud with Mesos >> 0.20 >> > and CDH5.1.2 HDFS. It is also configured with zookeeper but my >> zookeeper at >> > the moment just runs in one master. I followed the instructions here: >> > http://wiki.apache.org/hama/GettingStartedMesos - that are for an older >> > version and additionally to that I added some more properties to >> overcome >> > some runtime problems and until Mesos was able to recognize Hama as a >> > framework. Overall the properties I have set are the following: >> > >> > bsp.master.address >> > bsp.master.port >> > bsp.master.TaskWorkerManager.class >> > fs.default.name (I don't think it needs to be here but in an effort to >> fix >> > the issues I had) >> > hama.mesos.executor.uri >> > bsp.tasks.maximum.total >> > hama.mesos.master >> > bsp.child.java.opts >> > bsp.system.dir >> > bsp.local.dir >> > bsp.tmp.dir >> > bsp.disk.queue.dir >> > hama.zookeeper.quorum >> > hama.zookeeper.property.clientPort >> > >> > When I run the example the executor is running but the task fails. The >> > error output on the executors log is: ERROR bsp.MesosExecutor: Caught >> > exception, committing suicide. >> > java.util.ServiceConfigurationError: org.apache.hadoop.fs.FileSystem: >> > Provider org.apache.hadoop.fs.LocalFileSystem not found >> > >> > I did some research on this error message but I cannot find something >> that >> > really helps. It seems to me as a configuration issue but I don't know >> how >> > to solve it. >> > Also note that I have verified that my HDFS works properly. I am >> actually >> > already running Spark and Hadoop over Mesos without problems. >> > >> > I hope somebody can help! >> > >> > thanks, >> > Stratos >> >> >
