how to use HBase shell after starting hbase in cluster mode

Question

I have three nodes, a master and two slaves (running as region servers), I initiated the hbase, it says,
starting master...starting slave1 ... starting slvae2... (zookeeper is running in the backend). Now, I did jps on each of the machines and I get:

In master node:

/usr/local/hbase$ jps
19111 HMaster
19338 Jps

In slave1 node:

/usr/local/hbase$ jps
24182 HRegionServer
24277 Jps

In slave2 node:

/usr/local/hbase$ jps
10647 HRegionServer
10696 Jps

Now, my question: Is everything fine, in the sense, are all regionservers up?
When I start the hbase shell , I get the following : what does this mean.. Does this imply any error?? I'm learning hbase pardon me if I'm too trivial in my questions...

/usr/local/hbase$ hbase shell
2018-08-14 12:56:07,482 WARN  [main] util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/usr/local/hbase/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/usr/local/hadoop/share/hadoop/common/lib/slf4j-log4j12-1.7.25.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
HBase Shell; enter 'help<RETURN>' for list of supported commands.
Type "exit<RETURN>" to leave the HBase Shell
Version 1.2.6.1, rUnknown, Sun Jun  3 23:19:26 CDT 2018

hbase(main):001:0>

Am I correct till here.. I don't want to move on ... and then come back to this error if I get stuck later..Can anyone help me say if it is the right output for hbase shell.. I didn't understand the meaning of SLF4J

When I say.. create 'test','cf' it throws me the following error:

ERROR: Can't get master address from ZooKeeper; znode data == null

Hbase is not able to connect zk service. check your zk configuration in hbase. — Shivakumar ss, Aug 14 '18 at 09:29
Are you able to see your zk process running ? it is a process by name "QuorumPeerMain" if yes make sure you are able to connect it via zkCli — Shivakumar ss, Aug 16 '18 at 04:28
No, I can't see QuorumPeerMain running when I give jps in the terminal. — saichand, Aug 16 '18 at 04:39
you have said in your question that zookeeper is running in (zookeeper is running in the backend), so check why those are not running. ? — Shivakumar ss, Aug 16 '18 at 05:12
saichanda@saichanda-OptiPlex-9020:~$ `sudo ./zookeeper-3.4.12/bin/zkServer.sh start` ` ZooKeeper JMX enabled by default` ` Using config: /home/saichanda/zookeeper-3.4.12/bin/../conf/zoo.cfg` `Starting zookeeper ... STARTED` saichanda@saichanda-OptiPlex-9020:~$ jps `29072 Jps` This is how I am running zookeeper, I don't see any QuorumPeerMain — saichand, Aug 16 '18 at 05:20
I don't understand the content of the zookeeper logs, the file named as `log.39d`. I can't read as it is not in english alone, mixed with some encoded content.. ?? I am clueless, I checked the logs, where I saved the zookeeper logs while running zookeeper, as it is set in the zoo.cfg file. — saichand, Aug 16 '18 at 05:51
enable logs in zk. probably you are checking data part of it. — Shivakumar ss, Aug 16 '18 at 05:58
I am checking `/home/saichanda/zookeeper/logs/log.39d`, In zookeeper file, I had `data` and `logs` folders and moreover, I have a related question here, (https://stackoverflow.com/questions/51869678/hmaster-is-not-started-when-run-hbase) — saichand, Aug 16 '18 at 06:17
since you have limited reputation i cannot take you to chat window of stackoverflow. let me know if you are available on skype. if yes than share the id — Shivakumar ss, Aug 16 '18 at 06:22

score -1 · Answer 1 · answered Aug 14 '18 at 07:55

-1

HBase master manage the whole cluster. So you can check your cluster in website: http://master:16010/master-status.

answered Aug 14 '18 at 07:55

D.Eric

50
7

is it 60010? or what you have said?... in fact, using both I can't get the master status, what does this mean? – saichand Aug 14 '18 at 08:27
1

Few things to do: first, check which processes on your master are listening to what ports. I believe some ports have changed in recent HBase versions (with 2.0.0 the port for master web ui is indeed 16010). Second, keep checking via JPS whehter HMaster is up sustainably (chances are it will crash eventually if something is wrong). Third, check which host names all your nodes and master are bound to (make sure that /etc/hosts is in synch with hbase-site.xml and regionserver files). Finally, check the master log in /usr/local/HBase/logs and you might get a clue of what specifically is wrong. – VS_FF Aug 14 '18 at 08:54
@VS_FF, As you said I checked all the details, my HMaster is not up sustainably, In the log folder I have ::: `SLF4J: Class path contains multiple SLF4J bindings. SLF4J: Found binding in [jar:file:/usr/local/hbase/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class] SLF4J: Found binding in [jar:file:/usr/local/hadoop/share/hadoop/common/lib/slf4j-log4j12-1.7.25.jar!/org/slf4j/impl/StaticLoggerBinder.class] SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation. SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory] ` – saichand Aug 14 '18 at 09:40
1

That's not the log output you need to figure out what's going wrong. In the /usr/local/HBase/logs directory, locate specifically the log file that has 'master' in the file name, and scroll all the way to the bottom (it's likely to be annoyingly long) You need to figure out why exactly the master is crashing. My best guess it has to do with host name resolutions/etc. – VS_FF Aug 14 '18 at 10:00
@VS_FF, I found the error to be `Failed to become active: Call From saichanda-OptiPlex-9020/127.0.1.1 to master:16000 failed on connection exception: java.net.ConnectException: Connection refused; For more details see: http://wiki.apache.org/hadoop/ConnectionRefused (since 1sec ago)` How to resolve this? I have looked at the site mentioned but didn't get a clue... – saichand Aug 16 '18 at 06:25
1

There are a lot of data points in that error that you need to investigate. What is the machine listed in the log? Why is something coming from local ip address there? Is that something that you expect? Is your master actually listening on port 16000? You can check all that at the OS level and as said previously, make sure all your host names and IP addresses are stated properly and never use the IP addresses in any HBase config files -- only host names. – VS_FF Aug 16 '18 at 08:49
1

by the way, as an example, i htink the latest 2.0 master configuration is to listen on port 60000. Check your hbase-site.xml to see what it's doing – VS_FF Aug 16 '18 at 10:03

how to use HBase shell after starting hbase in cluster mode

1 Answers1