HCX Underlay PMTU reporting an MTU value of -1

Products

• VMware HCX

Issue/Introduction

• The following error is reporting at random intervals on both the HCX Dashboard and vCenter UI.
  • From HCX Dashboard:
    • HCX Underlay PMTU Discovery mismatch found for the uplink <uplink_name> Previous Discovered MTU: <MTU_configured> Current DiscoveredMTU: -1.
      
  • From VC UI:
    
    • HCX Interconnect Underlay PMTU mismatch found


• When reviewing the </common/logs/admin/app.log> located on source/target HCX Manager the following behavior is observed.
  
  • Source HCX Manager:
    

    <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] INFO  c.v.v.h.s.i.u.PmtuCollectionJob- PmtuCollectionJob Running in state: GET_PMTU_COLLECTION_TEST_RESULT
    
    <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] INFO  c.v.h.i.u.InterconnectUnderlayUplinkUtil- Underlay test result for the appliance : <Service-Mesh_name> Url : https://<IX-I_IP>:9443/uc/ondemand/results_pmtu?taskId=##########-## responseBody : {"taskId":"##########-##","startMeasureTime":1757043947,"endMeasureTime":1757044017,"overallStatus":1,"tests":[{"localAddress":"##.##.##.###","remoteAddress":"##.##.##.###","currentLocalMtu":1390,"currentRemoteMtu":1390,"txPmtu":0,"rxPmtu":0,"status":1,"error":""}]}
    
    <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] INFO  c.c.s.m.j.c.v.h.i.u.InterconnectUnderlayEventUtil- handleUnderlayEventAlerting: eventCode PMTU_DISCOVERY_FAILED
    
    <timestamps> UTC [VCenterAlarmService_EventListener, , , TxId: ] INFO  c.v.v.h.v.alarm.service.AlarmUtils- Event with alert Id ########-####-####-####-######## successfully posted to vCenter ########-####-####-####-########
    
    <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] INFO  c.c.s.m.j.c.v.h.i.u.InterconnectUnderlayEventUtil- handleUnderlayEventAlerting: eventCode PMTU_MISMATCH
    
    <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] INFO  c.c.s.m.j.c.v.h.i.u.InterconnectUnderlayEventUtil- handleUnderlayEventAlerting: eventCode PMTU_DISCOVERY_FAILED
    
    <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] ERROR c.v.v.h.s.i.u.PmtuCollectionJob- Failed to get PMTU test results {}
    org.codehaus.jettison.json.JSONException: JSONObject["compute"] not found.

Environment

VMware HCX

Cause

• There is a timeout of approx. 60 secs when PMTU is triggered by Manager and executed by Appliances. If no report is given within that time period, PMTU collection workflow timeout occurs and prints below message:
  

  <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] INFO  c.c.s.m.j.c.v.h.i.u.InterconnectUnderlayEventUtil- handleUnderlayEventAlerting: eventCode PMTU_DISCOVERY_FAILED

  <timestamps> UTC [InterconnectService_SvcThread-367, IX:########-####-####-####-########, J:########, , TxId: ########-####-####-####-########] INFO  c.c.s.m.j.c.v.h.i.u.InterconnectUnderlayEventUtil- handleUnderlayEventAlerting: eventCode PMTU_MISMATCH

• PMTU mismatch error should not be printing as there is no actual MTU change detected (issue is with test collection workflow). 

Resolution

• The PMTU workflow will be optimized in an upcoming release of HCX. This KB article will be updated when that occurs. 
• It is safe to ignore this message.