Troubleshooting

Quick links

After installing Opster Metricbeat, when will I begin to see events in AutoOps?

AutoOps usually begins showing stats and generating events around 15 minutes after the initial data is received from your cluster.

After connecting Opster Metricbeat, I pressed the “I have run the command” button and nothing happened.

Check the Metricbeat logs to see whether stats are being sent.
If not, contact Opster’s support team for assistance.

To check whether stats are being sent, run the following command or check the log path:

  • Docker:
docker logs opster-metricbeat
  • Kubernetes
kubectl logs -l k8s-app/opster.io=metricbeat-opster
  • RPM Package:
cat /var/log/opster/metricbeat OR journalctl -u metricbeat-opster.service
  • Debian Package:
cat /var/log/opster/metricbeat OR journalctl  -u metricbeat-opster.service
  • Docker Swarm:
docker logs opster-metricbeat
  • Docker Compose:
docker compose logs opster-metricbeat
  • Windows:
c:\ProgramData\metricbeat\logs\metricbeat

Can I use AutoOps if my cluster is not directly connected to the Internet?

Yes. Only Opster Metricbeat needs to have the ability to connect to the internet for AutoOps to work.

What URL should I use if my cluster is running on a localhost?

In most cases, using https://localhost:9200 as the URL is sufficient.

For some networking configurations of the node, it may be necessary to use the configured IP address. To use localhost as your cluster host, ensure that MetricBeat agent machine can resolve and reach that URL.

In the metricbeat-opster logs, I see the following error: “HTTP error 401 in : 401 Unauthorized”. What does this mean?

This error is usually a result of incorrect or missing permissions.
Use the ‘Basic Authentication’ method and verify that your user name and password are correct. 

There may be a need to provide additional permissions for the user that is used to run the Agent. Check the AutoOps installation doc for more details.

If the issue continues, contact Opster’s support team for assistance.

I’m seeing the following error: “Unable to retrieve version information from Elasticsearch nodes. connect ECONNREFUSED 127.0.0.1:9200″. What does this mean?

Make sure that your Elasticsearch cluster is running on localhost and port 9200. If it isn’t, update the ES_URL in your environment file. You should also check whether you are using HTTP or HTTPS. If you are testing with local docker, the network must be set to:

http://host.docker.internal:9200

to reach the local Elasticsearch cluster.

How can I change the name of my cluster/s in AutoOps?

If your clusters have the same name, AutoOps provides an option to set an alias for any of your clusters so you can easily manage and visualize the analyzed data.

Which port should I open in my firewall to send metrics to AutoOps?

If you’re required to open an IP/port on your firewall to ship the data, please add the following address:

  • 100.24.140.28 – receiver.opster.io
  • 3.230.83.185 – rcvs.opster.io

Port: 80, 433 or 5044 (less recommended)