Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Threaded View
HBase >> mail # user >> HBase client hangs after connecting to Zookeeper in pseudo-mode


Copy link to this message
-
Re: HBase client hangs after connecting to Zookeeper in pseudo-mode
Can you also send us the server logs during that timeframe?
Thanks.

-- Lars

________________________________
 From: Pal Konyves <[EMAIL PROTECTED]>
To: [EMAIL PROTECTED]
Sent: Tuesday, March 19, 2013 5:02 PM
Subject: HBase client hangs after connecting to Zookeeper in pseudo-mode
 
Hi,

I'm new to the list, just registered for this question.

I'm setting up HBase in pseudo-distributed mode. I can run the hbase
shell locally, and create tables, put some rows, they are present in
the HDFS good.
However, when I try to connect either from the same machine or the
host machine (the HBase setup is on a VM), the client hangs after
connecting to ZooKeeper.

The console output is:

13/03/20 00:20:36 INFO zookeeper.ClientCnxn: Opening socket connection to server
debian-vm/192.168.1.102:2181. Will not attempt to authenticate using SASL (unkn
own error)
13/03/20 00:20:36 INFO zookeeper.ClientCnxn: Socket connection established to de
bian-vm/192.168.1.102:2181, initiating session
13/03/20 00:20:36 INFO zookeeper.ClientCnxn: Session establishment complete on s
erver debian-vm/192.168.1.102:2181, sessionid = 0x13d84e02e920005, negotiated ti
meout = 180000
[..then waits forever, then after forever shows:]
org.apache.hadoop.hbase.client.NoServerForRegionException: Unable to find regio
for stops,,99999999999999 after 10 tries.

I checked the following pitfalls:
-- My HBase 94.4 build is compatible with my Hadoop 1.0.4 according to
HBase documentation
-- I put 'vm-debian' (the hostname I want to identify my virtual
machine that runs the HBase setup) into the host and virtual machine's
/etc/host file to map the hostname nicely (and it works, because
browser shows Hadoop and HBase info pages)
-- I copied the hadoop related jars from hadoop root and hadoop/lib
into HBase/lib (although they were the same version, but I could never
know)
-- I set up the time on the virtual machine where the HBase setup runs
properly, no time shifts
-- I checked the logs and web UIs to see every processes are running:
Hadoop, HBase master, Zookeeper

Environment:
HBase install is a Debian 32 bit virtual machine, with JDK 1.7.0_17
Host machine is Windows 7 with JDK 1.7.0_05

but the client program does not work regardless if it's running on the
HBase VM or on the host

The tiny client is:
http://pastebin.com/060Yu2Gz

here are the config files:
http://pastebin.com/ArVvwU59

Please help and tell me what I do wrong, because it has been driving
me crazy for about 3 days, and I'm getting very angry about this whole
Hadoop stuff, that I cannot even get running a very basic use-case
with it.

Thanks,
Pal
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB