If you set up the
datahub CLI tool (see here), you can use the built-in check utility:
You can list all Docker containers in your local by running
docker container ls. You should expect to see a log similar to the below:
Also you can check individual Docker container logs by running
docker logs <<container_name>>. For
datahub-gms, you should see a log similar to this at the end of the initialization:
datahub-frontend-react, you should see a log similar to this at the end of the initialization:
If you're seeing errors like below, chances are you didn't give enough resource to docker. Please make sure to allocate at least 8GB of RAM + 2GB swap space.
You can use a utility like kafkacat to list all topics. You can run below command to see the Kafka topics created in your Kafka broker.
MetadataAuditEvent topics exist besides the default ones. Example response as below:
You can run below command to see the search indices created in your Elasticsearch.
corpuserinfodocument indices exist besides the default ones. Example response as below:
Once the mysql container is up and running, you should be able to connect to it dirctly on
localhost:3306 using tools such as MySQL Workbench. You can also run the following command to invoke MySQL Command-Line Client inside the mysql container.
Inspect the content of
metadata_aspect table, which contains the ingested aspects for all entities.
There can be different reasons why a container fails during initialization. Below are the most common reasons:
bind: address already in use#
This error means that the network port (which is supposed to be used by the failed container) is already in use on your system. You need to find and kill the process which is using this specific port before starting the corresponding Docker container. If you don't want to kill the process which is using that port, another option is to change the port number for the Docker container. You need to find and change the ports parameter for the specific Docker container in the
docker-compose.yml configuration file.
OCI runtime create failed#
If you see an error message like below, please make sure to git update your local repo to HEAD.
failed to register layer: devmapper: Unknown device#
This most means that you're out of disk space (see #1879).
ERROR: for kafka-rest-proxy Get https://registry-1.docker.io/v2/confluentinc/cp-kafka-rest/manifests/5.4.0: EOF#
This is most likely a transient issue with Docker Registry. Retry again later.
Try the following
More discussions on the same issue https://github.com/docker/hub-feedback/issues/1250
Table 'datahub.metadata_aspect' doesn't exist error when logging in#
This means the database wasn't properly initialized as part of the quickstart processs (see #1816). Please run the following command to manually initialize it.
Run the following script to remove all the containers and volumes created during the quickstart tutorial. Note that you'll also lose all the data as a result.