Spark生态之Tachyon学习6---集群版搭建问题之集群无法全部启动
六个worker只启动了3个：
hadoop@Master:~/cloud/alluxio-0.7.1$ ./bin/tachyon format Connecting to Mcnode1 as hadoop... Formatting Tachyon Worker @ Mcnode1 Connection to mcnode1 closed. Connecting to Mcnode2 as hadoop... Formatting Tachyon Worker @ Mcnode2 Connection to mcnode2 closed. Connecting to Mcnode3 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Formatting Tachyon Worker @ Mcnode3 /home/hadoop/cloud/alluxio-0.7.1/bin/tachyon: line 264: /../bin/java: No such file or directory Connection to mcnode3 closed. Connecting to Mcnode4 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Formatting Tachyon Worker @ Mcnode4 /home/hadoop/cloud/alluxio-0.7.1/bin/tachyon: line 264: /../bin/java: No such file or directory Connection to mcnode4 closed. Connecting to Mcnode5 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Formatting Tachyon Worker @ Mcnode5 /home/hadoop/cloud/alluxio-0.7.1/bin/tachyon: line 264: /../bin/java: No such file or directory Connection to mcnode5 closed. Connecting to Mcnode6 as hadoop... Formatting Tachyon Worker @ Mcnode6 Connection to mcnode6 closed. Formatting Tachyon Master @ Master hadoop@Master:~/cloud/alluxio-0.7.1$ ./bin/tachyon-start.sh all Mount Killed 0 processes on Master Killed 0 processes on Master Connecting to Mcnode1 as hadoop... Killed 0 processes on Mcnode1 Connection to mcnode1 closed. Connecting to Mcnode2 as hadoop... Killed 0 processes on Mcnode2 Connection to mcnode2 closed. Connecting to Mcnode3 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode3 Connection to mcnode3 closed. Connecting to Mcnode4 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode4 Connection to mcnode4 closed. Connecting to Mcnode5 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode5 Connection to mcnode5 closed. Connecting to Mcnode6 as hadoop... Killed 0 processes on Mcnode6 Connection to mcnode6 closed. Starting master @ Master Connecting to Mcnode1 as hadoop... Formatting RamFS: /mnt/ramdisk (256mb) umount: only root can do that ERROR: umount RamFS /mnt/ramdisk failed Mount failed, not starting worker Connection to mcnode1 closed. Connecting to Mcnode2 as hadoop... Formatting RamFS: /mnt/ramdisk (256mb) umount: only root can do that ERROR: umount RamFS /mnt/ramdisk failed Mount failed, not starting worker Connection to mcnode2 closed. Connecting to Mcnode3 as hadoop... dirname: missing operand Try 'dirname --help' for more information. TACHYON_LOGS_DIR: /home/hadoop/cloud/alluxio-0.7.1/logs Formatting RamFS: /mnt/ramdisk (256mb) umount: only root can do that ERROR: umount RamFS /mnt/ramdisk failed Mount failed, not starting worker Connection to mcnode3 closed. Connecting to Mcnode4 as hadoop... dirname: missing operand Try 'dirname --help' for more information. TACHYON_LOGS_DIR: /home/hadoop/cloud/alluxio-0.7.1/logs Formatting RamFS: /mnt/ramdisk (256mb) umount: only root can do that ERROR: umount RamFS /mnt/ramdisk failed Mount failed, not starting worker Connection to mcnode4 closed. Connecting to Mcnode5 as hadoop... dirname: missing operand Try 'dirname --help' for more information. TACHYON_LOGS_DIR: /home/hadoop/cloud/alluxio-0.7.1/logs Formatting RamFS: /mnt/ramdisk (256mb) umount: only root can do that ERROR: umount RamFS /mnt/ramdisk failed Mount failed, not starting worker Connection to mcnode5 closed. Connecting to Mcnode6 as hadoop... Formatting RamFS: /mnt/ramdisk (256mb) umount: only root can do that ERROR: umount RamFS /mnt/ramdisk failed Mount failed, not starting worker Connection to mcnode6 closed. hadoop@Master:~/cloud/alluxio-0.7.1$ ./bin/tachyon-start.sh all SudoMount Killed 1 processes on Master Killed 0 processes on Master Connecting to Mcnode1 as hadoop... Killed 0 processes on Mcnode1 Connection to mcnode1 closed. Connecting to Mcnode2 as hadoop... Killed 0 processes on Mcnode2 Connection to mcnode2 closed. Connecting to Mcnode3 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode3 Connection to mcnode3 closed. Connecting to Mcnode4 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode4 Connection to mcnode4 closed. Connecting to Mcnode5 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode5 Connection to mcnode5 closed. Connecting to Mcnode6 as hadoop... Killed 0 processes on Mcnode6 Connection to mcnode6 closed. Starting master @ Master Connecting to Mcnode1 as hadoop... [sudo] password for hadoop: Formatting RamFS: /mnt/ramdisk (256mb) Starting worker @ Mcnode1 Connection to mcnode1 closed. Connecting to Mcnode2 as hadoop... [sudo] password for hadoop: Formatting RamFS: /mnt/ramdisk (256mb) Starting worker @ Mcnode2 Connection to mcnode2 closed. Connecting to Mcnode3 as hadoop... dirname: missing operand Try 'dirname --help' for more information. [sudo] password for hadoop: dirname: missing operand Try 'dirname --help' for more information. Formatting RamFS: /mnt/ramdisk (256mb) Starting worker @ Mcnode3 Connection to mcnode3 closed. Connecting to Mcnode4 as hadoop... dirname: missing operand Try 'dirname --help' for more information. [sudo] password for hadoop: dirname: missing operand Try 'dirname --help' for more information. Formatting RamFS: /mnt/ramdisk (256mb) Starting worker @ Mcnode4 Connection to mcnode4 closed. Connecting to Mcnode5 as hadoop... dirname: missing operand Try 'dirname --help' for more information. [sudo] password for hadoop: dirname: missing operand Try 'dirname --help' for more information. Formatting RamFS: /mnt/ramdisk (256mb) Starting worker @ Mcnode5 Connection to mcnode5 closed. Connecting to Mcnode6 as hadoop... [sudo] password for hadoop: Formatting RamFS: /mnt/ramdisk (256mb) Starting worker @ Mcnode6 Connection to mcnode6 closed. hadoop@Master:~/cloud/alluxio-0.7.1$ hadoop@Master:~/cloud/alluxio-0.7.1$ jps 30101 SparkSubmit 17541 ResourceManager 2526 TachyonMaster 17389 SecondaryNameNode 27677 CoarseGrainedExecutorBackend 2610 Jps 17173 NameNode 17935 Master 18125 Worker 1450 SparkSubmit 27574 SparkSubmit 3528 HistoryServer
解决办法：
在每个节点各自启动，可以将所有节点启动，在master的19999可以看到：
Last login: Thu May 5 20:26:04 2016 from mcnode4 hadoop@Mcnode5:~$ cd cloud/alluxio-0.7.1/bin/ hadoop@Mcnode5:~/cloud/alluxio-0.7.1/bin$ ./tachyon-start.sh local Killed 0 processes on Mcnode5 Killed 0 processes on Mcnode5 Connecting to Mcnode1 as hadoop... Killed 1 processes on Mcnode1 Connection to mcnode1 closed. Connecting to Mcnode2 as hadoop... Killed 1 processes on Mcnode2 Connection to mcnode2 closed. Connecting to Mcnode3 as hadoop... Warning: Permanently added 'mcnode3,219.219.220.131' (ECDSA) to the list of known hosts. dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode3 Connection to mcnode3 closed. Connecting to Mcnode4 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 1 processes on Mcnode4 Connection to mcnode4 closed. Connecting to Mcnode5 as hadoop... Warning: Permanently added 'mcnode5,219.219.220.215' (ECDSA) to the list of known hosts. dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode5 Connection to mcnode5 closed. Connecting to Mcnode6 as hadoop... Killed 1 processes on Mcnode6 Connection to mcnode6 closed. [sudo] password for hadoop: dirname: missing operand Try 'dirname --help' for more information. Formatting RamFS: /mnt/ramdisk (256mb) Starting master @ Master Starting worker @ Mcnode5
但是上传一个文件后又报错：
hadoop@Mcnode3:~/cloud/tachyon-0.7.1-hadoop2.6-bin$ jps Error occurred during initialization of VM java.lang.Error: Properties init: Could not determine current working directory. at java.lang.System.initProperties(Native Method) at java.lang.System.initializeSystemClass(System.java:1119)
关闭集群：
hadoop@Master:~/cloud/tachyon-0.7.1-hadoop2.6-bin$ ./bin/tachyon-stop.sh Killed 1 processes on Master Killed 0 processes on Master Connecting to Mcnode1 as hadoop... Killed 1 processes on Mcnode1 Connection to mcnode1 closed. Connecting to Mcnode2 as hadoop... Killed 1 processes on Mcnode2 Connection to mcnode2 closed. Connecting to Mcnode3 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode3 Connection to mcnode3 closed. Connecting to Mcnode4 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 0 processes on Mcnode4 Connection to mcnode4 closed. Connecting to Mcnode5 as hadoop... dirname: missing operand Try 'dirname --help' for more information. Killed 1 processes on Mcnode5 Connection to mcnode5 closed. Connecting to Mcnode6 as hadoop... Killed 1 processes on Mcnode6 Connection to mcnode6 closed.