DaemonSet mirroring

Traceable provides you with an option to mirror the traffic for Kubernetes pods by running the Traceable agent in a daemonSet. As shown in the diagram below, each worker node has a Traceable Platform agent pod running. The Traceable agent pod has two containers, the Traceable Platform agent container and the mirroring container.

Before You Begin

Keep the TA_REFRESH_TOKEN handy. It will be required while installing the Traceable Platform Agent. To fetch the token, log into Traceable Platform and navigate to Administration () >  Account > Agent Token.

Download

There is no download required for DaemonSet agentless traffic mirroring. The installation is completed using the Helm chart or Terraform template.

Installation

The installation process consists of the following steps:

1. Create traceableai namespace
2. Install Platform Agent
3. Enable mirroring
4. [optional] Set mirroring mode

Step 1 - Create traceableai namespace

Create a traceableai namespace. Enter the following command: 

kubectl create namespace traceableai

Step 2 - Install Platform agent

You can install Traceable platform agent in Kubernetes environment using either Helm chart or Terraform template.

Install using Helm chart

1. Create a values.yamlfile with the following content:
   YAML

   token: <ACCESS_TOKEN>
   environment: <ENVIRONMENT_NAME>
   runAsDaemonSet: true
   daemonSetMirroringEnabled: true

2. Enter the following commands to install Traceable Platform Agent into your Kubernetes cluster:
   ActionScript

   helm repo add traceableai https://helm.traceable.ai
   helm repo update
   helm install --namespace traceableai traceable-agent traceableai/traceable-agent --values values.yaml

Install using Terraform

1. Create a main.tf file with the following content:
   ActionScript

   module "traceable-agent" {
     source                       = "https://downloads.traceable.ai/install/traceable-agent/terraform/kubernetes/latest/traceable-agent-tf-k8s.tar.gz"
     token                        = "<ACCESS_TOKEN>"
     environment                  = "<ENVIRONMENT_NAME>"
     run_as_daemon_set            = true
     daemon_set_mirroring_enabled = true
   }

2. Initialize the module by entering the following command:
   ActionScript

   terraform init

3. Review the setup that will be installed by entering the following command:
   ActionScript
   ActionScript

   terraform plan

4. Apply the module by entering the following command:
   ActionScript
   ActionScript

   terraform apply

   Terraform creates a terraform.tfstate file. The file stores the current state of the infrastructure. It is recommended to keep track of this file.

Step 3 - Enable mirroring

To configure mirroring, go through the following points:

• Enable mirroring for all namespaces - Mirroring is disabled by default. To enable mirroring for all namespaces, use the following configuration:
  • If you are using Helm, then in values.yaml, set - daemonSetMirrorAllNamespaces: true 
  • If you are using Terraform, then in main.tf, set - daemon_set_mirror_all_namespaces = true
• Enable mirroring for a namespace - set the namespace label traceableai-mirror to enabled.
• Disable mirroring for a namespace - set the namespace label traceableai-mirror to disabled.
• Disable mirroring for a pod - set the pod annotation mirror.traceable.ai/enabled to false.

Step 4 [optional] - Set mirroring mode

By default, only ingress traffic is captured. However, you can capture only egress traffic, or both ingress and egress traffic, by configuring correct annotations.

Capture egress traffic

To capture the egress traffic for a deployment or namespace, set the following annotations.

Deployment

To capture the egress traffic, set the deployment annotation mirror.traceable.ai/mode to egress. Enter the following command:

kubectl patch deployment <deployment> -n <namespace> -p '{"spec": {"template":{"metadata":{"annotations":{"mirror.traceable.ai/mode":"egress"}}}} }'

Namespace

To capture egress traffic at the namespace level, set the annotation mirror.traceable.ai/defaultMode to egress. Enter the following:

kubectl annotate namespace <NAMESPACE> mirror.traceable.ai/defaultMode=egress

Capture ingress and egress traffic

To capture both ingress and egress traffic for a deployment or namespace, set the following annotations.

Deployment

To capture ingress and egress traffic for a deployment, set the deployment annotation mirror.traceable.ai/mode toingress_and_egress. Enter the following command:

kubectl patch deployment <deployment> -n <namespace> -p '{"spec": {"template":{"metadata":{"annotations":{"mirror.traceable.ai/mode":"ingress_and_egress"}}}} }'

Namespace

To capture the ingress and egress traffic at a namespace level, set the annotation mirror.traceable.ai/defaultMode to ingress_and_egress. Enter the following command:

kubectl annotate namespace <NAMESPACE> mirror.traceable.ai/defaultMode=ingress_and_egress

Verification

• To verify a successful installation of traceable-agent, enter the following command:

kubectl -ntraceableai get pods

For example,

NAME                    READY   STATUS    RESTARTS   AGE
traceable-agent-dwclc   2/2     Running   0          26m

• You can also check whether traces are showing in the Traceable Platform by navigating to Traces.

Troubleshooting

If traces are not showing in Traceable Platform, then:

1. Enter the following command to fetch the Traceable agent pod name:
   ActionScript
   ActionScript

   kubectl -ntraceableai get pods 

2. Enter the following command to get the logs of pods being monitored:
   ActionScript
   ActionScript

   kubectl -ntraceableai logs -f <TRACEABLE_AGENT_POD> -c traceable-agent | grep "Added pod to maps"

   For example,
   ActionScript
   ActionScript

   {"level":"info","time":"2022-01-12T08:46:42.218Z","message":"Added pod to maps. {hello-world-deployment-1.default true 0 {default true 0}}","service":"ext_cap","name":"hello-world-deployment-1-58749467bf-d24dr","ip":"10.1.1.221"}
   {"level":"info","time":"2022-01-12T08:46:42.221Z","message":"Added pod to maps. {product-deployment.default true 0 {default true 0}}","service":"ext_cap","name":"product-deployment-7bd74db6dd-kkbl5","ip":"10.1.1.218"}
   {"level":"info","time":"2022-01-12T08:46:42.224Z","message":"Added pod to maps. {product-deployment.default true 0 {default true 0}}","service":"ext_cap","name":"product-deployment-7bd74db6dd-wtwrt","ip":"10.1.1.217"}
   {"level":"info","time":"2022-01-12T08:46:42.234Z","message":"Added pod to maps. {traceable-agent.traceableai true 0 {traceableai true 0}}","service":"ext_cap","name":"traceable-agent-hm7jp","ip":"192.168.65.4"}

   
   

Upgrade

You can upgrade the Traceable agent in Kubernetes using either Helm chart or Terraform. 

Upgrade using Helm

Enter the following commands:

1. Update helm charts by entering the following command:
   ActionScript
   ActionScript

   helm repo update traceableai

2. Enter the following command to upgrade Traceable agent to the latest version:
   ActionScript
   ActionScript

   helm upgrade traceable-agent --namespace traceableai traceableai/traceable-agent

Upgrade using Terraform

To upgrade using terraform, you would need to uninstall and reinstall the Platform agent. 

Uninstall

You can uninstall the agent using either Helm chart or Terraform. 

Uninstall using Helm

Enter the following command to uninstall the Platform agent using Helm:

helm uninstall traceable-agent --namespace traceableai

Uninstall using Terraform

Enter the following command from the directory containing main.tf template file:

terraform destroy