Excellent! Glad to hear you got it working, and thanks for sharing the
solution.
On 7/24/14, 1:28 PM, David Snyder wrote:
Great. Thanks Josh. I was focusing on it being a zookeeper issue
before. I have gotten it to work successfully. The change made was
that my conf/masters file was specified as "localhost" and not the
explicit hostname. Once I made sure that all the processes were using
the DNS names everything went through.
On Wed, Jul 23, 2014 at 6:40 PM, Josh Elser <[email protected]
<mailto:[email protected]>> wrote:
Hi David,
Actually, you're speaking just fine to your ZooKeeper. The big hint
here is that ZooKeeper does not use Thrift for communication; Thrift
is what Accumulo is using for its RPCs. All of the ZooKeeper output
at the end of your message all appear normal.
Since everything appears to be working correctly with the shell, I
would try to focus on running these examples:
- Check the classpath, ACCUMULO_HOME and ACCUMULO_CONF_DIR that are
used/set when you invoke the WordCount program.
- Make sure accumulo-env.sh sets the correct environment variables
On 7/23/14, 7:14 PM, David Snyder wrote:
Hello!
I've recently installed Accumulo 1.6 on a 3 node cluster and
have the
processes started and configured correctly to my knowledge, but I am
having difficulty running the wordCount example.
Specifically, when I run:
`./bin/accumulo
org.apache.accumulo.examples.__simple.mapreduce.WordCount
-i accumulo -z zookeeper0:2181 --input /user/jenkins/README -t
wordCount
-u root -p root --debug`
I get the following Thrift error which appears to be from a
failure to
communicate with my Zookeeper server:
https://gist.github.com/__davidsnyder/__b33fad7b92bc6a906809
<https://gist.github.com/davidsnyder/b33fad7b92bc6a906809>
Despite the fact that I can run ListInstances just fine with the
same
"-z" zookeeper arg:
[user@hdp0 accumulo]$ ./bin/accumulo
org.apache.accumulo.server.__util.ListInstances -z zookeeper0:2181
INFO : Using ZooKeepers zookeeper0:2181
Instance Name | Instance ID |
Master
---------------------+--------__------------------------------__+-----------------------------__--
"accumulo" | 4505350e-1f07-48f0-902b-__509722e968f5 |
I also passed the public ip of the zookeeper instance instead of the
DNS, but I also verified that the DNS was resolving:
[user@hdp0 accumulo]$ host zookeeper0
zookeeper0.XXX.XXX has address 10.0.1.18
[user@hdp0 accumulo]$ echo ruok | nc zookeeper0 2181
imok[user@hdp0 accumulo]$
I can create tables in the shell, insert records, scan,
everything just
fine. The monitor GUI shows my two tablet servers and there are no
other errors to speak of. I also attempted to use the helloworld
"org.apache.accumulo.examples.__simple.helloworld.__InsertWithBatchWriter"
jar and received the same result as I did with WordCount. There's
nothing to speak of in the accumulo logs. The zookeeper logs
repeatedly
show the following when I run the wordCount command line
(10.0.0.1 is
the internal ip of hdp0):
014-07-23 22:57:50,670 [myid:] - INFO
[Thread-3079:NIOServerCnxn@__1001]
- Closed socket connection for client /10.0.0.101:44342
<http://10.0.0.101:44342>
<http://10.0.0.101:44342> (no session established for client)
2014-07-23 22:57:55,671 [myid:] - INFO
[NIOServerCxn.Factory:0.0.0.0/__0.0.0.0:2181:__NIOServerCnxnFactory@197
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxnFactory@197>
<http://0.0.0.0/0.0.0.0:2181:__NIOServerCnxnFactory@197
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxnFactory@197>>] - Accepted
socket connection from /10.0.0.101:44363
<http://10.0.0.101:44363> <http://10.0.0.101:44363>
2014-07-23 22:57:55,671 [myid:] - INFO
[NIOServerCxn.Factory:0.0.0.0/__0.0.0.0:2181:NIOServerCnxn@821
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxn@821>
<http://0.0.0.0/0.0.0.0:2181:__NIOServerCnxn@821
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxn@821>>] - Processing stat
command from /10.0.0.101:44363 <http://10.0.0.101:44363>
<http://10.0.0.101:44363>
2014-07-23 22:57:55,672 [myid:] - INFO
[Thread-3081:NIOServerCnxn@__1001]
- Closed socket connection for client /10.0.0.101:44363
<http://10.0.0.101:44363>
<http://10.0.0.101:44363> (no session established for client)
2014-07-23 22:58:00,673 [myid:] - INFO
[NIOServerCxn.Factory:0.0.0.0/__0.0.0.0:2181:__NIOServerCnxnFactory@197
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxnFactory@197>
<http://0.0.0.0/0.0.0.0:2181:__NIOServerCnxnFactory@197
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxnFactory@197>>] - Accepted
socket connection from /10.0.0.101:44380
<http://10.0.0.101:44380> <http://10.0.0.101:44380>
2014-07-23 22:58:00,677 [myid:] - INFO
[NIOServerCxn.Factory:0.0.0.0/__0.0.0.0:2181:NIOServerCnxn@821
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxn@821>
<http://0.0.0.0/0.0.0.0:2181:__NIOServerCnxn@821
<http://0.0.0.0/0.0.0.0:2181:NIOServerCnxn@821>>] - Processing stat
command from /10.0.0.101:44380 <http://10.0.0.101:44380>
<http://10.0.0.101:44380>
2014-07-23 22:58:00,678 [myid:] - INFO
[Thread-3083:NIOServerCnxn@__1001]
- Closed socket connection for client /10.0.0.101:44380
<http://10.0.0.101:44380>
<http://10.0.0.101:44380> (no session established for client)
I'm really not sure what else to try at this point. Any ideas?
Thanks,
David Snyder
