-
Bug
-
Resolution: Done
-
Normal
-
netobserv-1.8-candidate
-
Quality / Stability / Reliability
-
False
-
-
None
-
Moderate
-
None
-
None
-
NetObserv - Sprint 265, NetObserv - Sprint 266
-
None
-
None
-
None
Description of problem:
Some flows with MAC address as 0s are seen on console with Byte and Packets == 0. This are considered as Partial flows, even though they're not as many, they can still occur.
Steps to Reproduce:
1. Set up client and server pods. 2. Set up Ingress policy to deny all traffic for server pod NS. 3. On Network Traffic page, see flows between client and server pods.
Actual results:
{ "AgentIP": "10.0.51.102", "Bytes": 74, "DnsErrno": 0, "Dscp": 0, "DstAddr": "10.129.2.20", "DstK8S_HostIP": "10.0.51.102", "DstK8S_HostName": "ip-10-0-51-102.us-east-2.compute.internal", "DstK8S_Name": "nginx-7b7b55f46f-shfv4", "DstK8S_Namespace": "test-server", "DstK8S_NetworkName": "primary", "DstK8S_OwnerName": "nginx", "DstK8S_OwnerType": "Deployment", "DstK8S_Type": "Pod", "DstMac": "0A:58:64:58:00:07", "DstPort": 8080, "DstSubnetLabel": "Pods", "Etype": 2048, "Flags": 2, "FlowDirection": "0", "IfDirections": [ 1, 0 ], "Interfaces": [ "genev_sys_6081", "genev_sys_6081" ], "K8S_FlowLayer": "app", "Packets": 1, "Proto": 6, "SrcAddr": "10.131.0.10", "SrcK8S_HostIP": "10.0.15.192", "SrcK8S_HostName": "ip-10-0-15-192.us-east-2.compute.internal", "SrcK8S_Name": "client", "SrcK8S_Namespace": "test-client", "SrcK8S_NetworkName": "primary", "SrcK8S_OwnerName": "client", "SrcK8S_OwnerType": "Pod", "SrcK8S_Type": "Pod", "SrcMac": "0A:58:64:58:00:05", "SrcPort": 39708, "SrcSubnetLabel": "Pods", "TimeFlowEndMs": 1732211289311, "TimeFlowStartMs": 1732211289311, "TimeReceived": 1732211292, "app": "netobserv-flowcollector" }
Expected results:
See comment
- is caused by
-
NETOBSERV-1951 Duplicate flows across ebpf-agent nodes render poor user experience
-
- Closed
-
- links to