Clients are receiving random HTTP 400 errors or invalid DNS queries after upgrading to version 22.1.7-2p2 or 2p3.

Products

• VMware Avi Load Balancer

Issue/Introduction

1.Clients are receiving random HTTP 400 errors after upgrading to the 22.1.7-2p2 patch version .

Significance:

• • Request abnormal event: Client sent an invalid HTTP/1.x request.
  • Request ended abnormally: response code 4xx

2. In case of DNS VS ,requests can fail with this error:(After upgrading to 22.1.7-2p3)

Significance: DNS query failed: invalid DNS query

 

Environment

VMware and NSX-T

Cause

• On LRO enabled L2 scaleout systems, flows punted to secondary SE may not work properly.
• For such flows, if the packet is lesser than 60 bytes, ethernet padding done by the switch ends up baffling punt path leading to wrong packets being punted to secondary.

Resolution

Workaround:

Disable LRO from Service Engine Group Properties and reboot the SE

How to check if LRO enabled on the SE interface

Check LRO
  
  [admin:10-80-xx-xx]: > show serviceengine <SE-Name> interface | grep lro
  
disable LRO for SEG
  
Disable LRO
  
  > configure serviceenginegroup <SE-Group-Name>
  > no se_lro
  > save
  
  [admin:10-80-xx-xx]: > show serviceenginegroup Default-Group | grep lro
  | se_lro                                  | False                                                   |

Please note that LRO is enabled on serviceenginegroup by default for supported environments, such as, VMware, NSX-T Cloud.

As It's SE-Group configurable property, The SEs should be rebooted for this configuration to work after disable the LRO

KB - https://docs.vmware.com/en/VMware-Avi-Load-Balancer/30.2/Configuration-Guide/GUID-42EAC98B-4DA3-4D84-BF28-634491D5D888.html

Fix Version: 22.1.7-2p4