Summary
A free-of-charge technical enablement series to help you get started with Instana.
Join us for an 8-week enablement series to help you understand the fundamentals of Instana.
We will deliver one 1-hour session each week dedicated to a key learning area of the product. The sessions are designed so you can drop in at any point into the curriculum.
During this session, you will learn
- Everything you need to know about the Instana Agent.
- How does it work, architecture, design, installation, and configuration options.
- Ask questions and get familiar with Instana features and capabilities to start your observability journey.
Join the community and be informed of the next time we run Instana Essentials. There is no on-demand recording available publicly for this series.
Please share any questions by clicking on the Reply button.
We have collected your questions and our answers from the live session.
How much CPU/Memory is required to install and run the Agent and How can it be calculated?
The required CPU/Memory are dependent on processes and applications running, number of sensors loaded and collected traces by the agent. In the Instana UI, after the Installation of the Agent - you will have the option to see exactly what resources the Agent are using. In addition, the CPU and Memory can be limited by the configuration file of the Agent. You can refer to our docs to see how to limit the CPU and Memory. https://www.ibm.com/docs/en/instana-observability/current?topic=agents-configuring-host#limiting-host-agent-cpu-and-memory
How to instrument Node.js applications integrated with OpenTelemetry?
Some information can be found in our documentation: https://www.ibm.com/docs/en/instana-observability/current?topic=apis-opentelemetry . There are different option to send data from the Node.js application to the Instana Backend with OTel protocol or using the Instana Agent.
Does agent support profiling for memory resources?
Yes, please refer to our documentation to see what is supported and how to analyse it: https://www.ibm.com/docs/en/instana-observability/current?topic=processes-configuring-profiling and https://www.ibm.com/docs/en/instana-observability/current?topic=processes-analyzing-profiles .
How can the Instana Agent be installed on an air-gaped environment?
That depends on the environment. But for a Linux host you can download the static Agent that contains all the sensors from the Instana UI, transfer them to the target host and install it there.
Is it recommended to install the Agent on control plane/master nodes on K8s cluster?
Yes, the advantage to install on control plane/master nodes on K8s cluster will be to provide Infra and process monitoring - with Dashboards, alerts for the plane/master nodes.
For monitoring some of the technologies we have to write the username and password in the configuration file, is there a way to hash or encrypt the password in the file?
Yes, there is a way to use HashiCorp Vault to manage secrets: https://www.ibm.com/docs/en/instana-observability/current?topic=agents-configuring-host#integration-with-secret-managers
If you are monitoring a physical server, is the agent and sensor the same thing in that specific context or are the agent and the sensor still separate entities?
A key ingredient to the Instana Dynamic APM solution is our agent architecture, and specifically, our use of sensors. Sensors are mini agents – small programs that are specifically designed to attach and monitor one thing. They are automatically managed by our single agent (one per host), which is deployed either as a stand-alone process on the host, or as a container via the container scheduler. A list of supported technologies can be found here: https://www.ibm.com/docs/en/instana-observability/current?topic=requirements-instana-sensor
Is it possible disable specific sensor?
Yes, it can be done in the /opt/instana-agent/etc/instana/configuation.yaml file, just uncomment the sensor and enabled: false part, e.g.,
com.instana.plugin.statsd:
enabled: false
How to install agent for K8S deployment on air gap environment?
There are different ways to do it. You can use proxys, mirror the agent images and helm chart into your own repository (e.g., with Artifactory) or pull the image plus helm chart and transfer them to your air-gapped environment to use them without a connection to the internet. But we haven't documented all the steps in the moment.
Is agent installation for Redhat OpenShift the same as Kubertnetes?
Yes, please refer to: https://www.ibm.com/docs/en/instana-observability/current?topic=agents-installing-host-agent-openshift
How about AWS Fargate support; is it the same as for Kubernetes?
The configuration is covered at https://www.ibm.com/docs/en/instana-observability/current?topic=agents-aws-fargate-agent . But keep in mind that we don't support AWS Fargate on EKS.
In a large distributed K8S environment, can there be more than one Agent? Is agent to agent communication possible?
There is one Agent needed per K8s Node. There is no communication between the Agents, as all the telemetry data are put into context in the Instana Backend. More information can be found in our documentation here: https://www.ibm.com/docs/en/instana-observability/current?topic=instana-monitoring-kubernetes
How can we monitor HMC and VSphere? Do we need some sort of relay server where the agent will be installed and then configure the configuration.yaml for HMC and VSPhere?
On the machine where you have installed INSTANA (/opt/instana-agent/etc/instana/configuration.yaml). No relay servers are needed.
The configuration to remotely monitor VSphere is documented here: https://www.ibm.com/docs/en/instana-observability/current?topic=instana-monitoring-vsphere .
Monitoring IBM Power HMC is covered here: https://www.ibm.com/docs/en/instana-observability/current?topic=technologies-monitoring-power-hmc.
And monitoring IBM Z HMC is covered here: https://www.ibm.com/docs/en/instana-observability/current?topic=technologies-monitoring-z-hmc
------------------------------
Viktor Manweiler
------------------------------