[alluxio-2.0-rc2]workers failed to start if config with multiple value to dirs.path

Alluxio Version: alluxio-2.0-rc2

Describe the bug workers failed to start if config with multiple value to dirs.path

To Reproduce

config a worker with below configurations, then try to start worker with alluxio-start.sh script alluxio.worker.tieredstore.level1.dirs.path=/mnt/ssd1,/mnt/ssd2,/mnt/ssd3

Expected behavior

worker should started properly

Urgency high

Additional context

#alluxio-start.sh worker Mount
Killed 0 processes on xxx.xx.x
Ramdisk not detected. Mounting...
Formatting RamFS: /mnt/ramdisk (42949672960)
Starting worker @ xxx.xx.x. Logging to /var/log/alluxio/
--- [ FAILED ] The worker @ xxx.xx.x is not serving requests.
--- Printing the log tail for /var/log/alluxio//worker.log
>>> BEGIN
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:zookeeper.version=3.4.6-1569965, built on 02/20/2014 09:09 GMT
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:host.name=xxx.xx.x
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:java.version=1.8.0_212
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:java.vendor=Oracle Corporation
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:java.home=/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.212.b04-0.el7_6.x86_64/jre
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:java.class.path=/usr/local/alluxio-2.0.0-RC2/conf/::/usr/local/alluxio-2.0.0-RC2/assembly/alluxio-server-2.0.0-RC2.jar
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:java.library.path=:/usr/local/hadoop-2.7.2/lib/native:/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.212.b04-0.el7_6.x86_64/lib/amd64/server:/usr/java/packages/lib/amd64:/usr/lib64:/lib64:/lib:/usr/lib
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:java.io.tmpdir=/tmp
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:java.compiler=<NA>
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:os.name=Linux
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:os.arch=amd64
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:os.version=5.0.10-1.el7.elrepo.x86_64
2019-06-13 11:28:48,014 INFO  ZooKeeper - Client environment:user.name=root
2019-06-13 11:28:48,015 INFO  ZooKeeper - Client environment:user.home=/root
2019-06-13 11:28:48,015 INFO  ZooKeeper - Client environment:user.dir=/root
2019-06-13 11:28:48,015 INFO  ZooKeeper - Initiating client connection, connectString=AA:2181,BB:2181,CC:2181 sessionTimeout=60000 watcher=org.apache.curator.ConnectionState@3c0f93f1
2019-06-13 11:28:48,032 INFO  ClientCnxn - Opening socket connection to server xxx.xx.x/xxx.xx.x:2181. Will not attempt to authenticate using SASL (unknown error)
2019-06-13 11:28:48,033 INFO  CuratorFrameworkImpl - Default schema
2019-06-13 11:28:48,038 INFO  ClientCnxn - Socket connection established to xxx.xx.x/xxx.xx.x:2181, initiating session
2019-06-13 11:28:48,043 INFO  ClientCnxn - Session establishment complete on server xxx.xx.x/xxx.xx.x:2181, sessionid = 0x1000000f43a0030, negotiated timeout = 40000
2019-06-13 11:28:48,050 INFO  ConnectionStateManager - State change: CONNECTED
2019-06-13 11:28:48,505 INFO  ConfigurationUtils - Alluxio client has loaded configuration from meta master xxx.xx.x/xxx.xx.x:19998
2019-06-13 11:28:48,510 INFO  ConfigurationUtils - Alluxio client (version 2.0.0-RC2) is trying to load cluster level configurations
2019-06-13 11:28:48,528 INFO  ConfigurationUtils - Alluxio client has loaded cluster level configurations
2019-06-13 11:28:48,547 INFO  TieredIdentityFactory - Initialized tiered identity TieredIdentity(node=xxx.xx.x, rack=null)
2019-06-13 11:28:48,748 INFO  BlockWorkerFactory - Creating alluxio.worker.block.BlockWorker
2019-06-13 11:28:48,748 INFO  FileSystemWorkerFactory - Creating alluxio.worker.file.FileSystemWorker
2019-06-13 11:28:48,776 INFO  StorageDir - Folder /mnt/ramdisk/alluxioworker was created!
2019-06-13 11:28:48,791 WARN  StorageTier - Failed to verify memory capacity
2019-06-13 11:28:48,793 INFO  StorageDir - Folder /mnt/ssd1/alluxioworker was created!
<<< EOF
--- Printing the log tail for /var/log/alluxio//worker.out
>>> BEGIN
	at alluxio.worker.WorkerProcess$Factory.create(WorkerProcess.java:38)
	at alluxio.worker.AlluxioWorker.main(AlluxioWorker.java:68)
Caused by: java.util.concurrent.ExecutionException: java.lang.ArrayIndexOutOfBoundsException: 1
	at java.util.concurrent.FutureTask.report(FutureTask.java:122)
	at java.util.concurrent.FutureTask.get(FutureTask.java:192)
	at alluxio.util.CommonUtils.invokeAll(CommonUtils.java:493)
	at alluxio.worker.AlluxioWorkerProcess.<init>(AlluxioWorkerProcess.java:118)
	... 3 more
Caused by: java.lang.ArrayIndexOutOfBoundsException: 1
	at alluxio.worker.block.meta.StorageTier.initStorageTier(StorageTier.java:95)
	at alluxio.worker.block.meta.StorageTier.newStorageTier(StorageTier.java:186)
	at alluxio.worker.block.BlockMetadataManager.<init>(BlockMetadataManager.java:68)
	at alluxio.worker.block.BlockMetadataManager.createBlockMetadataManager(BlockMetadataManager.java:83)
	at alluxio.worker.block.TieredBlockStore.<init>(TieredBlockStore.java:126)
	at alluxio.worker.block.DefaultBlockWorker.<init>(DefaultBlockWorker.java:134)
	at alluxio.worker.block.BlockWorkerFactory.create(BlockWorkerFactory.java:43)
	at alluxio.worker.block.BlockWorkerFactory.create(BlockWorkerFactory.java:26)
	at alluxio.worker.AlluxioWorkerProcess.lambda$new$0(AlluxioWorkerProcess.java:110)
	at java.util.concurrent.FutureTask.run(FutureTask.java:266)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
Heap
 par new generation   total 613440K, used 229045K [0x00000005c0000000, 0x00000005e9990000, 0x00000005e9990000)
  eden space 545344K,  42% used [0x00000005c0000000, 0x00000005cdfad660, 0x00000005e1490000)
  from space 68096K,   0% used [0x00000005e1490000, 0x00000005e1490000, 0x00000005e5710000)
  to   space 68096K,   0% used [0x00000005e5710000, 0x00000005e5710000, 0x00000005e9990000)
 concurrent mark-sweep generation total 3512768K, used 0K [0x00000005e9990000, 0x00000006c0000000, 0x00000007c0000000)
 Metaspace       used 21897K, capacity 22156K, committed 22292K, reserved 1069056K
  class space    used 2610K, capacity 2704K, committed 2760K, reserved 1048576K
<<< EOF

[ns1123]

Hi Frank,

Do you have a tiered store configured for level0? Our indices start with 0 so changing the property to alluxio.worker.tieredstore.level0.dirs.path should fix it. To have two tiers, you need to set alluxio.worker.tieredstore.levels=2. Then you could set alluxio.worker.tieredstore.level1.dirs.path to configure your second tier.

[gpang]

This was an issue with worker registration. This was fixed by this PR: https://github.com/Alluxio/alluxio/pull/9517