Teamed Cards communication issue

[JasonHatch]

All;

I have a load balancer that I am using in my environment. I make a web service request from the server to a site controlled by the Load Balancer. Sometimes it communicates successfully while other times it does not. If I go directly to the webserver I don't have any communication issues. After capturing network traffic we have found out that when communication is successful the packets illustrate the communication is going to the MAC address of the Teamed cards which is also the MAC of the first card. When communication fails, captured packets illustrate that the communication is trying to go to the MAC of the second card. I was thinking I would modify the MAC on the teamed card to be its own MAC so it is neither using the burned in MAC of either of the teamed cards. I'm concerned that the issue may be worse however. Has anyone encountered this issue?

[Lmm_1]

Jason,
I´ve seen similar issues where devices with non-traditional TCP/IP stack(NetApps, IP Biometric access, Radiofrequency controllers...etc ) resolves an IP address to the Secondary (TX only) NIC MAC Address on a TLB Team. In this case you will see failure when trying to communicate with some hosts on the same subnet.

The easier way to get rid of this issue will be changing the mode to NFT instead of TLB, another option will be use those devices on a differente subnet, this way all the communication will be done through a router, so the device will reply to the Router MAC, and the router will talk to the team MAC, not the secondary NIC MAC.

Rgds,
Lmm

[JasonHatch]

Setting the card in NFT mode did resolve this issue. Thanks for your help.

[Lmm_1]

http://forums13.itrc.hp.com/service/forums/helptips.do?#28