Title: Log Management for UCS Tools Applications#

URL Source: https://docs.nvidia.com/ucf/2.10.0/text/UCS_logging.html

Published Time: Thu, 30 Oct 2025 07:23:04 GMT

Markdown Content:
Telemetry consists of reading logs from various components. It plays a vital role in checking the health of constituent microservices. You must ship logs to a log management system where logs can be stored, analyzed, and alerts generated for any issues.

Prerequisites[#](https://docs.nvidia.com/ucf/2.10.0/text/UCS_logging.html#prerequisites "Link to this heading")
---------------------------------------------------------------------------------------------------------------

*   Microservices used in the application must log to stdout / stderr.

Enable Logging in MicroK8s Cluster[#](https://docs.nvidia.com/ucf/2.10.0/text/UCS_logging.html#enable-logging-in-microk8s-cluster "Link to this heading")
---------------------------------------------------------------------------------------------------------------------------------------------------------

Enable log management easily in a MicroK8s Cluster with the [Elasticsearch, Fluentd and Kibana (the EFK stack)](https://microk8s.io/docs/addon-fluentd).

To demonstrate logging, we will use the application from the tutorial [Sharing Storage between Pod Replicas](https://docs.nvidia.com/ucf/2.10.0/text/tutorials/04_Sharing_Storage_Replicas.html#tutorial-4) and assume it has been already deployed.

### Enable the EFK stack[#](https://docs.nvidia.com/ucf/2.10.0/text/UCS_logging.html#enable-the-efk-stack "Link to this heading")

First enable the `community` add-on and then the `fluentd` add-on.

$ microk8s enable community
...
Community repository is now enabled

$ microk8s enable fluentd
...
Fluentd-Elasticsearch is enabled

Next, forward the Kibana service port to access the Kibana service from any host.

$ microk8s kubectl port-forward -n kube-system service/kibana-logging --address 0.0.0.0 8181:5601
Forwarding from 0.0.0.0:8181 -> 5601

### Create a new index pattern[#](https://docs.nvidia.com/ucf/2.10.0/text/UCS_logging.html#create-a-new-index-pattern "Link to this heading")

Now open `http://<NODE_IP>:8181/app/management/kibana/indexPatterns/create` in a browser where `NODE_IP` is the IP address of the machine where the `kubectl port-forward` command is running.

Enter `*` or `logstash-*` as the index pattern and click on **Next step**.

![Image 1: MicroK8S Logging - Kibana - Index Pattern](https://docs.nvidia.com/ucf/2.10.0/_images/ucf_microk8s_logging_kibana_indexpattern.png)
Next, select `@timestamp` as the timestamp field and click on **Create index pattern**.

![Image 2: MicroK8S Logging - Kibana - Timestamp field](https://docs.nvidia.com/ucf/2.10.0/_images/ucf_microk8s_logging_kibana_timestamp_field.png)
### View logs[#](https://docs.nvidia.com/ucf/2.10.0/text/UCS_logging.html#view-logs "Link to this heading")

Now go to `http://<NODE_IP>:8181/app/discover`. The application logs along with other Kubernetes and system logs should be visible here.

Filter the data to only show `curl-client` logs and only show a few selected fields.

![Image 3: MicroK8S Logging - Kibana - Logs](https://docs.nvidia.com/ucf/2.10.0/_images/ucf_microk8s_logging_kibana_logs.png)
Thus we can see logs from our application in the EFK stack.

Links/Buttons:
- [#](https://docs.nvidia.com/ucf/2.10.0/text/UCS_logging.html#view-logs)
- [Elasticsearch, Fluentd and Kibana (the EFK stack)](https://microk8s.io/docs/addon-fluentd)
- [Sharing Storage between Pod Replicas](https://docs.nvidia.com/ucf/2.10.0/text/tutorials/04_Sharing_Storage_Replicas.html#tutorial-4)