For more information regarding the security incident at F5, the actions we are taking to address it, and our ongoing efforts to protect our customers, click here.

Forum Discussion

kartheeswaran_1's avatar
kartheeswaran_1
Icon for Cirrus rankCirrus
Feb 02, 2015

Roundrobin Load balancing not happen

VIP:10.134.34.101 POOL MEMBERS(SERVER-A:10.134.34.177, SERVER-B: 10.134.34.202 )

 

IF I disable any one pool member round robin load balancing working fine.But if both are up it is not working properly. while pinging web Some times server A opens and once login goes inside server B and not able to login. kindly suggest

 

application delivery
availability
deployment
LTM
TMOS

14 Replies

  • JRahm's avatar
    JRahm
    Icon for Admin rankAdmin
    Feb 02, 2015

    Use a tool like ab (you can launch it from the BIG-IP cli) to send multiple concurrent connections and you should see RR kick in.

     

    ab -n 10000 -c 50 http://10.134.34.101/

     

    If you still aren't seeing load balancing, do you have persistence enabled?

     

  • kartheeswaran_1's avatar
    kartheeswaran_1
    Icon for Cirrus rankCirrus
    Feb 02, 2015

    hi jason thanks for your comments, Persistence not enabled also i suspect due to VIP,Poolmembers in same segment or due to Automap configured. and also there is no vlan configured in my SLB. All are local

     

  • JRahm's avatar
    JRahm
    Icon for Admin rankAdmin
    Feb 02, 2015

    surely the self IPs and the VIP are available on at least one vlan (even if it is shared)? are both pool members available?

     

  • NikhilB's avatar
    NikhilB
    Icon for Employee rankEmployee
    Feb 02, 2015

    Have you tried another LB algo? (least connections member/node)

     

    Let us know the results.

     

  • shaggy's avatar
    shaggy
    Icon for Nimbostratus rankNimbostratus
    Feb 02, 2015
    are you saying that round-robin is not sending even traffic to all pool members, or are you saying that the application breaks when you enable more than one server in a pool with round-robin enabled?
  • swo0sh_gt_13163's avatar
    swo0sh_gt_13163
    Icon for Altostratus rankAltostratus
    Feb 03, 2015
    Hello kartheeswaran, You've mentioned that you have only 2 pool members. i.e. SERVER-A:10.134.34.177, SERVER-B: 10.134.34.202, now when you disable one pool member, round robin works, didn't get that. Minimum requirement to measure any LB algorithm is to have 2 pool members. If you have only one member available in the pool, we cannot say LB works fine. Can you shed more light here?
  • kartheeswaran_1's avatar
    kartheeswaran_1
    Icon for Cirrus rankCirrus
    Feb 03, 2015
    if i disable one pool member , i am able to login server from internet. If I enable both pool member same facing issues. i dont know what bug
  • kartheeswaran_1's avatar
    kartheeswaran_1
    Icon for Cirrus rankCirrus
    Feb 03, 2015

    Also in this scenario 2 servers are having ssl certificate.Here SSL offloading happens in Server. Not in Load balancer. And given http profile has none

     

  • shaggy's avatar
    shaggy
    Icon for Nimbostratus rankNimbostratus
    Feb 03, 2015

    Try enabling persistence. This isn't an issue with the LB method but is an interoperability issue in which you need a user to continue using the same pool member once a pool member has been selected. Try source_addr persistence at first to see if it resolves your issue. Cookie persistence would probably be better, but the F5 would need to terminate SSL for the application (SSL offload).