Hello,
We have just recently updated FreeBSD 8 to 9. Since then it is not possible to ping our Microsoft NLB Cluster anymore, in 8.2 there were no problems.
In /var/log/messages the line
appears (and the corresponding "last messages repeated 500 times" lines too).
I have found no clue except this http://freebsd.1045724.n5.nabble.co...-source-address-error-messages-td4975944.html, which indicates that there are two "culprits": MS is way too lax concerning the RFC and FreeSBD way too strict.
We are running our terminal server NLB on this MAC, the whole thing runs on VMWare ESXi 5.0, our FreeBSD is on a standalone machine. The corresponding ESX/Windows admins told me that they could try to change the network config back to unicast, but this would involve severe changes in their field.
To cut this short: Is there a way to reenable the FreeBSD pre 9 behavior, allowing MAC Multicast?
(We have fixed this problem temporarily by routing the NLB-Clusters IP via another machine (FreeBSD 4(!)), which is really just a temporary solution due to the age of that gateway.)
We have just recently updated FreeBSD 8 to 9. Since then it is not possible to ping our Microsoft NLB Cluster anymore, in 8.2 there were no problems.
In /var/log/messages the line
Code:
kernel: in_arp: 03:bf:##:##:##:## is multicastappears (and the corresponding "last messages repeated 500 times" lines too).
I have found no clue except this http://freebsd.1045724.n5.nabble.co...-source-address-error-messages-td4975944.html, which indicates that there are two "culprits": MS is way too lax concerning the RFC and FreeSBD way too strict.
We are running our terminal server NLB on this MAC, the whole thing runs on VMWare ESXi 5.0, our FreeBSD is on a standalone machine. The corresponding ESX/Windows admins told me that they could try to change the network config back to unicast, but this would involve severe changes in their field.
To cut this short: Is there a way to reenable the FreeBSD pre 9 behavior, allowing MAC Multicast?
(We have fixed this problem temporarily by routing the NLB-Clusters IP via another machine (FreeBSD 4(!)), which is really just a temporary solution due to the age of that gateway.)