Alert "NSX-T Edge Node Tunnel Status is 'Degraded'" is triggered due to VM vMotion
search cancel

Alert "NSX-T Edge Node Tunnel Status is 'Degraded'" is triggered due to VM vMotion

book

Article ID: 416703

calendar_today

Updated On:

Products

VCF Operations for Networks

Issue/Introduction

Symptoms:

  • Alert titled ""NSX-T Edge Node Tunnel Status is 'Degraded" appears in the VCF Operations for Networks(formerly vRealize Network Insight) GUI. However, the Edge tunnels are confirmed to be operational when checked via NSX manager GUI.
  • The application team reports no traffic disruption within Guest OS.
  • VM vMotion occur just before alert timestamp.
    • ESXi host log /var/log/vmkernel.log confirms VM vMotion to another host

YYYY-MM-DDTHH:MM:SS In(182) vmkernel: cpu66:4989626 opID=886c5da)Migrate: 314: vmotion: Source vmmLeaderID = 4989634, ts = xxxxxx, srcIP = <Source ESXi IP> dstIP = <Destination ESXi IP> Dest wid = 7312747 using SHARED
 
YYYY-MM-DDTHH:MM:SS In(182) vmkernel: cpu47:4989626)Hbr: 4213: Migration end received (worldID=4989634) (migrateType=1) (event=1) (isSource=1) (sharedConfig=1)

  • Subsequently NSX Edge log /var/log/syslog shows tunnel change and tunnel summary

YYYY-MM-DDTHH:MM:SS <Edge name> NSX 1 FABRIC [nsx@6876 comp="nsx-edge" subcomp="nsxa" s2comp="tunnel" level="INFO"] Tunnel <Edge vTEP IP>:<ESXi vTEP IP>(geneve) state updated from up to down


YYYY-MM-DDTHH:MM:SS <Edge name> NSX 9649 FABRIC [nsx@6876 comp="nsx-edge" subcomp="datapathd" s2comp="dpc-pb" tname="dp-ipc31" level="INFO"] Remove tunnel xxxxxx-xxxxxx-xxxxxx(<Edge vTEP IP> -> <ESXi vTEP IP>) from the repl vector of routing domain xxxxxx-xxxxxx-xxxxxx


YYYY-MM-DDTHH:MM:SS <Edge name> NSX 9649 FABRIC [nsx@6876 comp="nsx-edge" subcomp="datapathd" s2comp="dp-aggsvc" tname="dp-ipc31" level="INFO"] Total tunnels: XXX, up: YYY, down: ZZZ, unknown: 0, skipped: 0

    • The subsequent tunnel summary (generated 30 seconds later) usually display that all tunnels are up, for example:

YYYY-MM-DDTHH:MM:SS <Edge name> NSX 9649 FABRIC [nsx@6876 comp="nsx-edge" subcomp="datapathd" s2comp="dp-aggsvc" tname="dp-ipc31" level="INFO"] Total tunnels: XXX, up: XXX, down: 0, unknown: 0, skipped: 0

 

Environment

VMware VCF Operations for Networks 6.x

Cause

When a VM undergoes vMotion to another host and no VTEP (VXLAN Tunnel Endpoint) spans utilize the tunnel, the tunnel will first be transitioned to a "down" state and then deleted.
This sequence explains why tunnel-down records appear in logs—subsequently, VCF Operations for Networks (formerly vRealize Network Insight) captures these information and displays them in the GUI.

Resolution

It is expected behavior.

Customer can either choose to ignore the alert or disable it from VCF Operations for Networks GUI.

Powered by Wolken Software