Trace the Path of a Packet with Antrea Traceflow

You can start a Traceflow session in NSX Manager to trace the path of packet in an Antrea container cluster. Antrea Traceflow currently supports tracing the path of only Unicast traffic. Broadcast and Multicast traffic are not supported.

The source of a Traceflow session must be a Pod, whereas the destination can be a Pod or a Service in the same container cluster. You can trace the path of a packet for the following types of traffic in an Antrea container cluster:

• Pod to Pod traffic on the same node (intra-node traffic)
• Pod to Pod traffic between nodes (inter-node traffic)
• Pod to Service traffic on the same node
• Pod to Service traffic between nodes
• Pod to an arbitrary IP address

Traceflow injects a test packet into the Antrea container cluster network and monitors the flow of the packet. As the packet flows from the source to destination, observations are collected from various components along the path of the packet. These observations are displayed in the Traceflow output, which shows the various components in the path of the packet.

Prerequisites

Antrea container cluster is registered to NSX.

Procedure

1. From your browser, log in to an NSX Manager at https://nsx-manager-ip-address.
2. Navigate to Plan & Troubleshoot > Traffic Analysis.
   Important: NSX Manager UI fetches the information about registered Antrea container clusters when you start the NSX Manager application in the browser. If the application UI is already open, it does not fetch the Antrea container cluster registration information automatically. This behavior is expected and per the current UI design. If you have registered the first Antrea container cluster after the NSX Manager application is opened, ensure that you refresh the browser after navigating to the Traceflow Analysis page. A manual refresh ensures that the Antrea Traceflow tab is visible in the UI when you reach step 4 of this procedure.

   This manual browser refresh is required only once, and not every time after a new Antrea container cluster is registered to NSX.

3. In the Traceflow card, click Get Started.
4. Click the Antrea Traceflow tab.
   This tab is available only when at least one Antrea container cluster is registered to NSX. If this tab is not visible, refresh the browser.
5. Specify the configuration settings of the Traceflow session.

   Field	Description
   Cluster	Select an Antrea container cluster from the drop-down menu. Alternatively, enter the first few characters of the cluster name to filter the list, and then select the container cluster.
   IP Address	Select either IPv4 or IPv6.
   Protocol Type	Select any one protocol type: ICMP, TCP, UDP. For ICMP: Optional: Enter the ICMP identifier. Default is 0. Optional: Enter the ICMP sequence number. Default is 0. Optional: Enter the TTL. Default is 64. For TCP, UDP: Optional: Enter the source port number. Default is 0. Optional: Enter the destination port number. Default is 0. (Only for TCP): Add TCP flags, if required. SYN flag is set by default. This flag is required for the Antrea Traceflow. Optional: Enter the TTL. Default is 64.
   Source	Only Pod is supported as the source of an Antrea Traceflow. Note: Pods that use the host network are currently not supported as the source of an Antrea Traceflow. This limitation is a known behavior. For example, Antrea Agent Pods are not supported in the source of an Antrea Traceflow. If you know the Pod name, select it directly from the Pod drop-down menu. Otherwise, filter the list of Pods by doing these steps: Enter the first few characters of the Node name, or select a value from the Node drop-down menu. Enter the first few characters of the Namespace, or select a value from the Namespace drop-down menu.
   Destination	Select the type of destination: Pod, Service, IP Address. If you select Pod or Service as the destination, use the Node and Namespace drop-down menus to filter the list of Pods or Services. If you select IP Address as the destination, enter an IP address.

6. Click Trace.

Results

The Traceflow observations are displayed in a tabular format. For each observation, the table shows the following information.

Observation Type

This column takes the following values.

Observation Type	Description
Delivered	The packet is delivered to destination Pod or Service properly.
Dropped	The packed is dropped by a network policy.
Received	The packet is received from another node in the container cluster.
Forwarded	The packet is forwarded to the next logical node or a container cluster object.

Component

This column shows the components that the test packet had encountered on its path from the source to the destination. Sample component values are: IngressRule, EgressRule, SpoofGuard, Classification, Output, and so on.

Click the component name to view more information in a pop-up window.

Timestamp

The date and time for each observation.