Hello, I have a Hama-0.6.0 on a Hadoop-1.0.4. I am running in the Pseudo Distributed Mode.
The example SSSP always returns 2147483647 as the shortest distance to each vertex. This is the input file. ===== one\ttwo:1 two\tone:1 ===== This is the command line. ===== <Hama root>/bin/hama jar hama-examples-0.6.0.jar sssp two <path to the input file on the local file system> <path to the output directory on the local file system> 1 ===== This is the shell prompt. ===== 13/06/16 09:41:28 INFO util.NativeCodeLoader: Loaded the native-hadoop library 13/06/16 09:41:28 INFO bsp.FileInputFormat: Total input paths to process : 1 13/06/16 09:41:30 INFO bsp.BSPJobClient: Running job: job_201306160108_0004 13/06/16 09:41:33 INFO bsp.BSPJobClient: Current supersteps number: 0 13/06/16 09:41:36 INFO bsp.BSPJobClient: Current supersteps number: 7 13/06/16 09:41:36 INFO bsp.BSPJobClient: The total number of supersteps: 7 13/06/16 09:41:36 INFO bsp.BSPJobClient: Counters: 11 13/06/16 09:41:36 INFO bsp.BSPJobClient: org.apache.hama.graph.GraphJobRunner$GraphJobCounter 13/06/16 09:41:36 INFO bsp.BSPJobClient: ITERATIONS=2 13/06/16 09:41:36 INFO bsp.BSPJobClient: MULTISTEP_PARTITIONING=1 13/06/16 09:41:36 INFO bsp.BSPJobClient: INPUT_VERTICES=2 13/06/16 09:41:36 INFO bsp.BSPJobClient: org.apache.hama.bsp.JobInProgress$JobCounter 13/06/16 09:41:36 INFO bsp.BSPJobClient: SUPERSTEPS=7 13/06/16 09:41:36 INFO bsp.BSPJobClient: LAUNCHED_TASKS=1 13/06/16 09:41:36 INFO bsp.BSPJobClient: org.apache.hama.bsp.BSPPeerImpl$PeerCounter 13/06/16 09:41:36 INFO bsp.BSPJobClient: SUPERSTEP_SUM=7 13/06/16 09:41:36 INFO bsp.BSPJobClient: TIME_IN_SYNC_MS=261 13/06/16 09:41:36 INFO bsp.BSPJobClient: IO_BYTES_READ=32 13/06/16 09:41:36 INFO bsp.BSPJobClient: TOTAL_MESSAGES_SENT=9 13/06/16 09:41:36 INFO bsp.BSPJobClient: TASK_INPUT_RECORDS=2 13/06/16 09:41:36 INFO bsp.BSPJobClient: TOTAL_MESSAGES_RECEIVED=9 Job Finished in 8.329 seconds ===== This is the output file. ===== one\ttwo:1 2147483647 two\tone:1 2147483647 ===== What may go wrong? Do I miss anything? Many thanks, Chui-hui
