You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@spark.apache.org by Rohith Parameshwara <rp...@couponsinc.com> on 2015/10/19 14:05:51 UTC

Unable to run applications on spark in standalone cluster mode

Hi all,
                I am doing some experiments on spark standalone cluster setup and I am facing the following issue:
I have a 4 node cluster setup. As per http://spark.apache.org/docs/latest/spark-standalone.html#starting-a-cluster-manually I tried to start the cluster with the scripts but, the slaves did not start and gave permission denied error....My conf/slaves had a list of IP addresses of the slaves... But Then I was able to start the worker nodes by going to the respective slave machine and running the start-slave.sh script with the master IP as parameter.... The webui showed the all the 3 worker nodes running as :
[cid:image001.png@01D10A93.B59D7300]

The application when submitted to the cluster will show in all the work nodes in the webui.... But in the command line where is spark-submit command was run, it gives this error periodically and continuously....
ERROR NettyTransport: failed to bind to /172.28.161.138:7077, shutting down Netty transport
15/10/19 11:59:38 WARN Utils: Service 'sparkDriver' could not bind on port 7077. Attempting port 7078.
15/10/19 11:59:38 WARN AppClient$ClientActor: Could not connect to akka.tcp://sparkMaster@localhost:7077: akka.remote.InvalidAssociation: Invalid address: akka.tcp://sparkMaster@localhost:7077....

My conf/spark-env.sh has:

SPARK_MASTER_IP=172.28.161.138
SPARK_MASTER_PORT=7077
SPARK_MASTER_WEBUI_PORT=8080
SPARK_WORKER_WEBUI_PORT=8081
SPARK_WORKER_INSTANCES=1
And I have also put this in all the slave nodes too....

The applications are running fine in -master local mode but in -master spark://masterip:7077, it is not working....
Any type of help would be appreciated..... Thanks in advance

Re: Unable to run applications on spark in standalone cluster mode

Posted by Rohith P <rp...@couponsinc.com>.
The contents of spark-env.sh is :
SPARK_MASTER_IP=marvin.spark.ins-01
SPARK_MASTER_PORT=7077
SPARK_MASTER_WEBUI_PORT=8080
SPARK_WORKER_WEBUI_PORT=8081
SPARK_WORKER_INSTANCES=1
SPARK_LOCAL_IP=marvin.spark.ins-01



The contents of etc/hosts is 
172.28.161.33           marvin.base.ins-01              ip-172-28-161-33

172.28.161.200          marvin.cassandra.ins-01         ip-172-28-161-200
172.28.161.201          marvin.cassandra.ins-02         ip-172-28-161-201

172.28.161.138          marvin.spark.ins-01             ip-172-28-161-138
172.28.161.139          marvin.spark.ins-02             ip-172-28-161-139
172.28.161.140          marvin.spark.ins-03             ip-172-28-161-140
172.28.161.141          marvin.spark.ins-04             ip-172-28-161-141





--
View this message in context: http://apache-spark-developers-list.1001551.n3.nabble.com/Unable-to-run-applications-on-spark-in-standalone-cluster-mode-tp14683p14871.html
Sent from the Apache Spark Developers List mailing list archive at Nabble.com.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@spark.apache.org
For additional commands, e-mail: dev-help@spark.apache.org


Re: Unable to run applications on spark in standalone cluster mode

Posted by Akhil Das <ak...@sigmoidanalytics.com>.
Can you paste the contents of your spark-env.sh file? Also would be good to
have a look at the /etc/hosts file. Cannot bind to the given ip address can
be resolved if you put the hostname instead of the ip address. Also make
sure the configuration (conf directory) across your cluster have the same
contents.

Thanks
Best Regards

On Mon, Oct 26, 2015 at 10:48 AM, Rohith P <rp...@couponsinc.com>
wrote:

> No.. the ./sbin/start-master.sh --ip option did not work... It is still the
> same error
>
>
>
>
> --
> View this message in context:
> http://apache-spark-developers-list.1001551.n3.nabble.com/Unable-to-run-applications-on-spark-in-standalone-cluster-mode-tp14683p14779.html
> Sent from the Apache Spark Developers List mailing list archive at
> Nabble.com.
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@spark.apache.org
> For additional commands, e-mail: dev-help@spark.apache.org
>
>

Re: Unable to run applications on spark in standalone cluster mode

Posted by Rohith P <rp...@couponsinc.com>.
No.. the ./sbin/start-master.sh --ip option did not work... It is still the
same error




--
View this message in context: http://apache-spark-developers-list.1001551.n3.nabble.com/Unable-to-run-applications-on-spark-in-standalone-cluster-mode-tp14683p14779.html
Sent from the Apache Spark Developers List mailing list archive at Nabble.com.

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@spark.apache.org
For additional commands, e-mail: dev-help@spark.apache.org


Re: Unable to run applications on spark in standalone cluster mode

Posted by Jean-Baptiste Onofré <jb...@nanthrax.net>.
Hi Rohith,

Do you have multiple interfaces on the machine hosting the master ?

If so, can you try to force to the public interface using:

sbin/start-master.sh --ip xxx.xxx.xxx.xxx

Regards
JB

On 10/19/2015 02:05 PM, Rohith Parameshwara wrote:
> Hi all,
>
>                  I am doing some experiments on spark standalone cluster
> setup and I am facing the following issue:
>
> I have a 4 node cluster setup. As per
> http://spark.apache.org/docs/latest/spark-standalone.html#starting-a-cluster-manually
> I tried to start the cluster with the scripts but, the slaves did not
> start and gave permission denied error….My conf/slaves had a list of IP
> addresses of the slaves… But Then I was able to start the worker nodes
> by going to the respective slave machine and running the start-slave.sh
> script with the master IP as parameter…. The webui showed the all the 3
> worker nodes running as :
>
> The application when submitted to the cluster will show in all the work
> nodes in the webui…. But in the command line where is spark-submit
> command was run, it gives this error periodically and continuously….
>
> ERROR NettyTransport: failed to bind to /172.28.161.138:7077, shutting
> down Netty transport
>
> 15/10/19 11:59:38 WARN Utils: Service 'sparkDriver' could not bind on
> port 7077. Attempting port 7078.
>
> 15/10/19 11:59:38 WARN AppClient$ClientActor: Could not connect to
> akka.tcp://sparkMaster@localhost:7077: akka.remote.InvalidAssociation:
> Invalid address: akka.tcp://sparkMaster@localhost:7077….
>
> My conf/spark-env.sh has:
>
> SPARK_MASTER_IP=172.28.161.138
>
> SPARK_MASTER_PORT=7077
>
> SPARK_MASTER_WEBUI_PORT=8080
>
> SPARK_WORKER_WEBUI_PORT=8081
>
> SPARK_WORKER_INSTANCES=1
>
> And I have also put this in all the slave nodes too….
>
> The applications are running fine in –master local mode but in –master
> spark://masterip:7077, it is not working….
>
> Any type of help would be appreciated….. Thanks in advance
>

-- 
Jean-Baptiste Onofré
jbonofre@apache.org
http://blog.nanthrax.net
Talend - http://www.talend.com

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@spark.apache.org
For additional commands, e-mail: dev-help@spark.apache.org