Connection refused for apache atlas server

Question

I have setup apache atlas on EC2. After setup , I started the apache atlas using command python2 atlas_start.py and it successfully started with below log:

configured for local hbase.
hbase started.
configured for local solr.
solr started.
setting up solr collections...
starting atlas on host localhost
starting atlas on port 21000
..............................................................................................................................................................................................................................................................................................................
Apache Atlas Server started!!!

However, when I try to hit the Server URL to verify if apache atlas is up and running. It gave me connection refused error. Below is the command I used to hit the server: curl -u username:password http://localhost:21000/api/atlas/admin/version

Also, while setup it didnt ask me for username and password SO, I am using admin:admin as username password.

`netstat -a` should be able to tell you if something is listening on port 21000 at all. — MatsLindh, Sep 29 '19 at 09:53
Would you please see the answer I have posted? I have explained the issue completely! Thanks. — Mostafa Ghadimi, Apr 03 '21 at 17:07

score 0 · Answer 1 · answered Dec 24 '20 at 18:42

you have to build the server using "mvn clean install -Pdist,", once this is done you need to navigate to the distro/target/ folder and then run the atlas_start.py from there.

Here solution

https://community.cloudera.com/t5/Support-Questions/Apache-Atlas-Started-but-with-Errors-and-Warnings/td-p/134183

score 0 · Answer 2 · answered Mar 28 '21 at 02:01

The problem with this script is that starting atlas using atlas_start.py takes about 10 minutes. While its process is not completed, you face to connection refused error.

In order to see whether it is completed or not, you can check the application.log file in logs directory using tail -n 10 application.log or cat application.log. In 2.1.0 version the last 10 line of application.log is as follows:

2021-03-28 06:00:06,536 INFO  - [main:] ~ Starting service org.apache.atlas.web.service.ActiveInstanceElectorService (Services:68)
2021-03-28 06:00:06,540 INFO  - [NotificationHookConsumer thread-0:] ~ [atlas-hook-consumer-thread]: Starting (Logging$class:66)
2021-03-28 06:00:06,542 INFO  - [main:] ~ HA is not enabled, no need to start leader election service (ActiveInstanceElectorService:103)
2021-03-28 06:00:06,542 INFO  - [NotificationHookConsumer thread-0:] ~ ==> HookConsumer doWork() (NotificationHookConsumer$HookConsumer:530)
2021-03-28 06:00:06,544 INFO  - [NotificationHookConsumer thread-0:] ~ Atlas Server is ready, can start reading Kafka events. (NotificationHookConsumer$HookConsumer:936)
2021-03-28 06:00:06,628 WARN  - [NotificationHookConsumer thread-0:] ~ [Consumer clientId=consumer-1, groupId=atlas] Error while fetching metadata with correlation id 2 : {ATLAS_HOOK=LEADER_NOT_AVAILABLE} (NetworkClient$DefaultMetadataUpdater:968)
2021-03-28 06:00:06,757 WARN  - [NotificationHookConsumer thread-0:] ~ [Consumer clientId=consumer-1, groupId=atlas] Error while fetching metadata with correlation id 4 : {ATLAS_HOOK=LEADER_NOT_AVAILABLE} (NetworkClient$DefaultMetadataUpdater:968)
2021-03-28 06:00:06,932 WARN  - [NotificationHookConsumer thread-0:] ~ [Consumer clientId=consumer-1, groupId=atlas] Error while fetching metadata with correlation id 6 : {ATLAS_HOOK=LEADER_NOT_AVAILABLE} (NetworkClient$DefaultMetadataUpdater:968)
2021-03-28 06:00:07,585 INFO  - [main:] ~ AuditFilter initialization started (AuditFilter:64)
2021-03-28 06:00:07,585 INFO  - [main:] ~ REST_API_ENABLE_DELETE_TYPE_OVERRIDE=false (AuditFilter:69)

After that you can see the login page http://localhost:21000:

or using curl command:

~ ➤ curl -u admin:admin http://localhost:21000/api/atlas/admin/version         

{"Description":"Metadata Management and Data Governance Platform over Hadoop","Revision":"release","Version":"2.1.0","Name":"apache-atlas"}%   
default username and password are **admin**.

score -1 · Accepted Answer · edited Jun 20 '20 at 09:12

There would be many reasons - do the following checks

Please check the logs which would be in the logs directory based on how you have it setup. (if you just ran it through mvn)
You need hbase(atleast) and solr/elastic/cassandra (atlas backends). Check if you are running embedded mode - where atlas will start its own backend. It is usually set as an environment variable. ie: MANAGE_LOCAL_HBASE
Make sure atlas-application.properties have the right configurations before you start it.

Hope this helps!

EDIT: I noticed you mentioned ec2. Make sure you have configured the right security groups.

Connection refused for apache atlas server

3 Answers3