Having issues with new core on 7710's upgraded from 6.2(8a) to 6.2(10) running ACL's on all SVI's, Netflow, and DHCP Snooping with one VDC added along side the admin VDC with 2 F348 line cards and dual-sups. The 6.2.10 upgrade was due to TCAM errors before finishing the entry of all ACL's. After the upgrade I entered the command for hardware access-list resource feature bank-mapping. Remaining ACL entries entered without issue, Netflow and DHCP snooping started.
During redundancy testing I can reload vdc 2 and while everything performs flawlessly when the vdc returns back online module 1 cannot be allocated back into the vdc due to tcam allocation error. Checking the tcam allocation resources I see tcam 1 bank0 is half full, tcam 1 bank 1 is empty as is all others.
Can this be reallocated, re-shuffled around for better distribution ?? Do I have to REMOVE everything and reconfigure to do this ?? Dont like having to reset the entire switch to recover modules if a vdc reloads but at least the constant errors seem to have stopped unless trying to configure or unconfigure the bank-mapping to try resource pooling, even with atomic updates disabled.
Please advise, dont have time to call Tac right now, hope someone has been thru this already,
thanks in advance.
dave
