Support Questions

Find answers, ask questions, and share your expertise
Announcements
Celebrating as our community reaches 100,000 members! Thank you!

Can anyone make a step by step tutorial for Installing Apache Atlas .8 or .9 on you laptop

avatar
Expert Contributor

Hi

I tried to install apache atlas on my laptop unfortunately I keep on getting exceptions. Can any one make a tutorial to get atlas running on your laptop with embeded Solr and Hbase setting using the link:

Apache Atlas Installation

How to configure it and what needs to be installed before hand (zookeeper). I just need a basic installation with defaults.

It would be great to mention

  1. Checkout URL
  2. maven commands
  3. additional dependencies (softwares)
  4. configurations file locations and settings

Once you download the repo there are multiple similar folders with the same content which I am confused about, regarding which one to use and where to run the scripts like start_atlas.py etc

Thanks

1 ACCEPTED SOLUTION

avatar
Expert Contributor

Ok here are all the steps required to run Apache Atlas natively with Berkeley DB and Elastic:

  1. Download and install Kafka use the link : https://kafka.apache.org/downloads. Download the binary and extract to your required location.
  2. Kafka and Atlas would also require Zookeeper. By default kafka comes with an instance of zookeeper. If you do not have zookeeper running or installed, you can use this. Navigate to and run : kafkahome/bin/zookeeper-server-start.sh
  3. Once zookeeper has started you can check it using the command: netstat -ant | grep :2181. if everything is fine you should see: tcp6 0 0 :::2181 :::* LISTEN
  4. Now you can start your kafka server using the command: ./kafkaHOME/bin/kafka-server-start.sh /KafkaHome/config/server.properties
  5. To check if kafka is running run the command netstat -ant | grep :9092. You should see a similar result as mentioned above.
  6. Now you are ready to move on with ATLAS. You can either use the link provided on the website or do a branch and tag checkout directly from github. I used the command on their website: git clone https://git-wip-us.apache.org/repos/asf/atlas.git atlas
  7. navigate into the folder : cd atlas
  8. Create new folder called libext using: mkdir libext
  9. You need to download the jar file form this URL. http://download.oracle.com/otn/berkeley-db/je-5.0.73.zip
  10. You will need an oracle account. Create one to download the zip file.
  11. Copy this zip file into your libext folder that you just created.
  12. run command export MAVEN_OPTS="-Xmx1536m -XX:MaxPermSize=512m"
  13. run command mvn clean install -DskipTests (MAKE SURE TO USE SKIP TESTS )
  14. run command: mvn clean package -DskipTests -Pdist,berkeley-elasticsearch
  15. Navigate to the following location: incubator-atlas/distro/target/apache-atlas-0.8-incubating-bin/apache-atlas-0.8-incubating/bin/atlas_start.py
    OR
    /home/arsalan/Development/atlas/distro/target/apache-atlas-0.9-SNAPSHOT-bin/apache-atlas-0.9-SNAPSHOT
    Depending on which repo you have used.
  16. Run the follwoing command python atlas_start.py

You can now navigate to localhost:21000 to check Atlas GUI.

Hope it helps!!!!!

View solution in original post

5 REPLIES 5

avatar
Expert Contributor

If you do a mvn clean install -DskipTests -Pdist,berkeley-elasticsearch (assuming you want to run embedded dependencies) and navigate to distro/target/atlas-<version>-bin, you will be able to run atlas on the local machine using the start script (atlas_start.py) under the bin folder.

If you're want to run against hbase and solr then you would need to have the following running before trying to start Atlas

  1. Zookeeper
  2. Hbase
  3. Kafka
  4. Solr (cloud mode)

Having these running is the first step towards starting Atlas. Once these services are up and running, you'll have to make sure that Zookeeper has registered the HBase servers, Kafka brokers and Solr servers as well. The next step is to update the atlas-application.properties with the correct address/ URLs for the above services. Once that's done then Atlas should be able to start serving requests.

PS: Running embedded mode is the fastest and easiest way to get Atlas up and running on a laptop/desktop.

Hope that helps, if it does please upvote and accept the answer.

avatar
Expert Contributor

@anaik

I was trying with the following command:

mvn clean package _DskipTests-Pdist,embedded-hbase-solr

In this case Solr and Hbase are installed and started automatically when atlas starts. I did not know that I would need to install kafka separately. I do have it installed, I will run it and see if any parameter need to be set in atlas for kafka.

avatar
Expert Contributor

I was not able to run it with hbase and solr. But the installation with berkely db and elastic works by simply following the installation steps on their website

avatar
Expert Contributor

Yeah running against external hbase and solr does require a lot of setup which is mostly done by Ambari in any HDP deployment.

avatar
Expert Contributor

Ok here are all the steps required to run Apache Atlas natively with Berkeley DB and Elastic:

  1. Download and install Kafka use the link : https://kafka.apache.org/downloads. Download the binary and extract to your required location.
  2. Kafka and Atlas would also require Zookeeper. By default kafka comes with an instance of zookeeper. If you do not have zookeeper running or installed, you can use this. Navigate to and run : kafkahome/bin/zookeeper-server-start.sh
  3. Once zookeeper has started you can check it using the command: netstat -ant | grep :2181. if everything is fine you should see: tcp6 0 0 :::2181 :::* LISTEN
  4. Now you can start your kafka server using the command: ./kafkaHOME/bin/kafka-server-start.sh /KafkaHome/config/server.properties
  5. To check if kafka is running run the command netstat -ant | grep :9092. You should see a similar result as mentioned above.
  6. Now you are ready to move on with ATLAS. You can either use the link provided on the website or do a branch and tag checkout directly from github. I used the command on their website: git clone https://git-wip-us.apache.org/repos/asf/atlas.git atlas
  7. navigate into the folder : cd atlas
  8. Create new folder called libext using: mkdir libext
  9. You need to download the jar file form this URL. http://download.oracle.com/otn/berkeley-db/je-5.0.73.zip
  10. You will need an oracle account. Create one to download the zip file.
  11. Copy this zip file into your libext folder that you just created.
  12. run command export MAVEN_OPTS="-Xmx1536m -XX:MaxPermSize=512m"
  13. run command mvn clean install -DskipTests (MAKE SURE TO USE SKIP TESTS )
  14. run command: mvn clean package -DskipTests -Pdist,berkeley-elasticsearch
  15. Navigate to the following location: incubator-atlas/distro/target/apache-atlas-0.8-incubating-bin/apache-atlas-0.8-incubating/bin/atlas_start.py
    OR
    /home/arsalan/Development/atlas/distro/target/apache-atlas-0.9-SNAPSHOT-bin/apache-atlas-0.9-SNAPSHOT
    Depending on which repo you have used.
  16. Run the follwoing command python atlas_start.py

You can now navigate to localhost:21000 to check Atlas GUI.

Hope it helps!!!!!