Forum Discussion

Thiyagu_343098's avatar
Thiyagu_343098
Icon for Nimbostratus rankNimbostratus
Feb 19, 2018

unequal load balancing of UDP based traffic across the pool members.

Hello All, I'm experiancing strance issues in UDP based traffic load balancing across all the pool members. Inspite of having round robing as load balancing method I'm observing one server in the pool taking more connection compared to the others.

 

I do not have any sticky persistance configuration for the application VIP.

 

Could you please guide me some steps in troubleshooting the cause of this unequal load balancing issue?

 

Do I need to clear the connection table and try?

 

Regards, Thiyagu

 

application delivery
BIG-IP
LTM
  • Only1masterbla1's avatar
    Only1masterbla1
    Icon for Cirrus rankCirrus
    Feb 19, 2018

    Please follow instruction as per article: K3605: Configuring the BIG-IP system to load balance UDP packets individually https://support.f5.com/csp/article/K3605

     

    -- other related details:

     

    K10430: Causes of uneven traffic distribution across BIG-IP pool members https://support.f5.com/csp/article/K10430 UDP traffic

     

    When multiple User Datagram Protocol (UDP) datagrams are sent from the same IP address and port to a UDP virtual server, the BIG-IP system, by default, sends all datagrams that arrive within the Idle Timeout period configured in the associated UDP profile to the same pool member to which the preceding datagram from the same source was sent. This can result in uneven load balancing statistics across pool members.

     

    Note: For more information, refer to K7535: Overview of the UDP profile. For information about modifying the default behavior, refer to K3605: Configuring the BIG-IP system to load balance UDP packets individually.