I'm trying to setup a hadoop cluster inside a docker swarm with multiple hosts, with a datanode on each docker node with a mounted volume.I made some tests and works fine, but the problem comes when a datanode dies and then return.
I restarted 2 host at the same time and when the containers run again, they get a new ip. The problem is that the namemode give a error because it thinks it is another datanode.
ERROR org.apache.hadoop.hdfs.StateChange: BLOCK* NameSystem.getDatanode: Data node 10.0.0.13:50010 is attempting to report storage ID 3a7b556f-7364-460e-beac-173132d77503. Node 10.0.0.9:50010 is expected to serve this storage.
Is is possible to prevent docker to assign a new ip, and instead keep the last ip after a restart?
Or there are any option for Hadoop config to fix this?
Static DHCP addresses for containers accessing an overlay network are officially not supported for the time being, as told here: https://github.com/moby/moby/issues/31860.
I hope, that docker will provide a solution for this very soon.
Related
I have created a docker swarm configuration which consists of a namenode, datanode, resource manager, and yarn workers. These all work well together and I can run hdfs dfs commands from any container in the swarm. I have also exposed the port 9000 using the ports section of the yaml. In my core-site.xml I use the hostname of the namenode from the swarm configuration.
I am unable to get a client outside of the swarm to access the cluster using the hdfs dfs commands. I have a different core-site.xml which has the address of the host machine for the swarm. When I run commands I get a java.io.EOFException.
Is there any way to get an external client to connect to the hadoop cluster running in docker swarm?
Turns out this is resolved by following the multihomed network instructions. It seems that it appears to the namenode as though it has two network interfaces to work with. One for the overlay network across the swarm and another from exposing the port.
I want to have Hadoop 2.3.0 in a multi bare-metal cluster using Docker. I have a master container and a slave container (in this first setup). When Master and Slave containers are in the same host (and therefore, same Flannel subnet), Hadoop works perfectly. However, if the Master and Slave are in different bare metal nodes (hence, different flannel subnets), it simply does not work (I get a connection refused error). Both containers can ping and ssh one another, so there is no connectivity problem. For some reason, it seems that hadoop needs all the nodes in the cluster to be in the same subnet. Is there a way to circumvent this?
Thanks
I think having the nodes in separate flannel subnets introduces some NAT-related rules which cause such issues.
See the below link which seems to have addressed a similar issue
Re: Networking Problem in creating HDFS cluster.
Hadoop uses a bunch of other ports for communication between the nodes, the above assumes these ports are unblocked.
ssh and ping are not enough. If you have iptables or any other firewalls, either you need to disable or open up the ports. You can set up the cluster, as long as hosts can communicate with each other and ports are open. Run telnet <namenode> <port> to ensure hosts are communicating on desired ports.
This is our first steps using big data stuff like apache spark and hadoop.
We have a installed Cloudera CDH 5.3. From the cloudera manager we choose to install spark. Spark is up and running very well in one of the nodes in the cluster.
From my machine I made a little application that connects to read a text file stored on hadoop HDFS.
I am trying to run the application from Eclipse and it displays these messages
15/02/11 14:44:01 INFO client.AppClient$ClientActor: Connecting to master spark://10.62.82.21:7077...
15/02/11 14:44:02 WARN client.AppClient$ClientActor: Could not connect to akka.tcp://sparkMaster#10.62.82.21:7077: akka.remote.InvalidAssociation: Invalid address: akka.tcp://sparkMaster#10.62.82.21:7077
15/02/11 14:44:02 WARN Remoting: Tried to associate with unreachable remote address [akka.tcp://sparkMaster#10.62.82.21:7077]. Address is now gated for 5000 ms, all messages to this address will be delivered to dead letters. Reason: Connection refused: no further information: /10.62.82.21:7077
The application is has one class the create a context using the following line
JavaSparkContext sc = new JavaSparkContext(new SparkConf().setAppName("Spark Count").setMaster("spark://10.62.82.21:7077"));
where this IP is the IP of the machine spark working on.
Then I try to read a file from HDFS using the following line
sc.textFile("hdfs://10.62.82.21/tmp/words.txt")
When I run the application I got the
Check your Spark master logs, you should see something like:
15/02/11 13:37:14 INFO Remoting: Remoting started; listening on addresses :[akka.tcp://sparkMaster#mymaster:7077]
15/02/11 13:37:14 INFO Remoting: Remoting now listens on addresses: [akka.tcp://sparkMaster#mymaster:7077]
15/02/11 13:37:14 INFO Master: Starting Spark master at spark://mymaster:7077
Then when your connecting to the master, be sure to use exactly the same hostname as found in the logs above (do not use the IP address):
.setMaster("spark://mymaster:7077"));
Spark standalone is a bit picky with this hostname/IP stuff.
When you create your Spark master using the shell command "sbin/start-master.sh". go the the address http://localhost:8080 and check the "URL" row.
I notice no accepted answer, just for info I thought I'd mention a couple things.
First, in the spark-env.sh file in the conf directory, the SPARK_MASTER_IP and SPARK_LOCAL_IP settings can be hostnames. You don't want them to be, but they can be.
As noted in another answer, Spark can be a little picky about hostname vs. IP address, because of this resolved bug/feature: See bug here. The problem is, it's not clear if they "resolved" is simply by telling us to use IP instead of hostname?
Well I am having this same problem right now, and the first thing you do is check the basics.
Can you ping the box where the Spark master is running? Can you ping the worker from the master? More importantly, can you password-less ssh to the worker from the master box? Per 1.5.2 docs you need to be able to do that with a private key AND have the worker entered in the conf/slaves file. I copied the relevant paragraph at the end.
You can get a situation where the worker can contact the master but the master can't get back to the worker so it looks like no connection is being made. Check both directions.
Finally of all the combinations of settings, in a limited experiment just now I only found one that mattered: On the master, in spark-env.sh, set the SPARK_MASTER_IP to the IP address, not hostname. Then connect from the worker with spark://192.168.0.10:7077 and voila it connects! Seemingly none of the other config parameters are needed here.
Here's the paragraph from the docs about ssh and slaves file in conf:
To launch a Spark standalone cluster with the launch scripts, you
should create a file called conf/slaves in your Spark directory, which
must contain the hostnames of all the machines where you intend to
start Spark workers, one per line. If conf/slaves does not exist, the
launch scripts defaults to a single machine (localhost), which is
useful for testing. Note, the master machine accesses each of the
worker machines via ssh. By default, ssh is run in parallel and
requires password-less (using a private key) access to be setup. If
you do not have a password-less setup, you can set the environment
variable SPARK_SSH_FOREGROUND and serially provide a password for each
worker.
Once you have done that, using the IP address should work in your code. Let us know! This can be an annoying problem, and learning that most of the config params don't matter was nice.
I am new to Spark and I am trying to run it on EC2. I follow the tutorial on spark webpage by using spark-ec2 to launch a Spark cluster. Then, I try to use spark-submit to submit the application to the cluster. The command looks like this:
./bin/spark-submit --class org.apache.spark.examples.SparkPi --master spark://ec2-54-88-9-74.compute-1.amazonaws.com:7077 --executor-memory 2G --total-executor-cores 1 ./examples/target/scala-2.10/spark-examples_2.10-1.0.0.jar 100
However, I got the following error:
ERROR SparkDeploySchedulerBackend: Application has been killed. Reason: All masters are unresponsive! Giving up.
Please let me know how to fix it. Thanks.
You're seeing this issue because the master node of your spark-standalone cluster cant open a TCP connection back to the drive (on your machine). The default mode of spark-submit is client which runs the driver on the machine that submitted it.
A new cluster mode was added to spark-deploy that submits the job to the master where it is then run on a client, removing the need for a direct connection. Unfortunately this mode is not supported in standalone mode.
You can vote for the JIRA issue here: https://issues.apache.org/jira/browse/SPARK-2260
Tunneling your connection via SSH is possible but latency would be a big issue since the driver would be running locally on your machine.
I'm curious if you still having this issue ... But in case anyone is asking here is a brief answer. As clarified by jhappoldt, the master node of your spark-standalone cluster cant open a TCP connection back to the drive (on your local machine). Two workarounds are possible, tested and succeeded.
(1) From EC2 Management Console, create a new security group and add rules to enable TCP back and forth from your PC (public IP). (what I did was adding TCP rules inbound and outbound) ... Then add this security group to your master instance. (right click --> Networking --> Change security groups). Note: add it and don't remove the already established security groups.
This solution work well, but in your specific scenario, deploying your application from local machine to EC2 cluster, you will face further problems (resource related) so the next option is the best one
(2) Having your .jar file (or .egg) copy it to the master node using scp. You can check this link http://docs.aws.amazon.com/AWSEC2/latest/UserGuide/AccessingInstancesLinux.html for information about how to do that; and deploy your application from the master node. Note: spark is already pre-insalled so you will do nothing but write the same exact command you write on your local machine from ~/spark/bin. This shall work perfect.
Are you executing the command on your local machine, or on the created EC2 node? If you're doing it locally, make sure port 7077 is open in the security settings, as its closed to the outside by default.
I'm trying to use the Dedoop application that runs using Hadoop and HDFS on Amazon EC2. The Hadoop cluster is set up and the Namenode JobTracker and all other Daemons are running without error.
But the war Dedoop.war application is not able to connect to the Hadoop Namenode after deploying it on tomcat.
I have also checked to see if the ports are open in EC2.
Any help is appreciated.
If you're using Amazon AWS, I highly recommend using Amazon Elastic Map Reduce. Amazon takes care of setting up and provisioning the Hadoop cluster for you, including things like setting up IP addresses, NameNode, etc.
If you're setting up your own cluster on EC2, you have to be careful with public/private IP addresses. Most likely, you are pointing to the external IP addresses - can you replace them with the internal IP addresses and see if that works?
Can you post some lines of the Stacktrace from Tomcat's log files?
Dedoop must etablish an SOCKS proxy server (similar to ssh -D port username#host) to pass connections to Hadoop nodes on EC2. This is mainly because Hadoop resolves puplic IPS to EC2-internal IPs which breaks MR Jobs submission and HDFS management.
To this end Tomcat must be configured to to etablish ssh connections. The setup procedure is described here.