I've been fighting this problem for a while so I don't recall the exact sequence of events, but from what I do recall, I rebooted my chassis which was running TCS 4.0. After it was rebooted it would no longer establish an OSPF session with one of our Cisco routers. It was able to establish a session with our other Cisco, our Foundry and our Lucent PM3 without a problem. Prior to this reboot I've been running OSPF sucessfully for many years. I did some research, talked to Cisco, searched the archives, searched 3Com knowledge base and everything came back with the same advice. It's a known issue with the Hiper Arc, simply reboot it. I rebooted it again, it still refused to talk to the one Cisco. The same article suggesting a reboot of the ARC also said that bug was fixed in a later version of the ARC software so I upgraded to TCS 4.5 and rebooted once again. The problem still persists. I have the priority set to 0 on the Arc, the PM3, and the Foundry. The Cisco's are 200 and 100. All routers exchange fine with each other except the Cisco with the priority of 200 and the Arc. Here's what the Cisco is reporting: (208.186.96.11 is the ARC) 5d01h: %OSPF-5-ADJCHG: Process 1, Nbr 208.186.96.11 on FastEthernet0/0 from EXST ART to DOWN, Neighbor Down: Too many retransmitionsig 5d01h: %OSPF-5-ADJCHG: Process 1, Nbr 208.186.96.11 on FastEthernet0/0 from DOWN to DOWN, Neighbor Down: Ignore timer expired DEBUG RESULTS: 5d01h: OSPF: Send DBD to 208.186.96.11 on FastEthernet0/0 seq 0xAA opt 0x42 flag 0x7 len 32 5d01h: OSPF: Retransmitting DBD to 208.186.96.11 on FastEthernet0/0 [5] 5d01h: OSPF: Send DBD to 208.186.96.11 on FastEthernet0/0 seq 0xAA opt 0x42 flag 0x7 len 32 5d01h: OSPF: Retransmitting DBD to 208.186.96.11 on FastEthernet0/0 [6] 5d01h: OSPF: Send DBD to 208.186.96.11 on FastEthernet0/0 seq 0xAA opt 0x42 flag 0x7 len 32 5d01h: OSPF: Retransmitting DBD to 208.186.96.11 on FastEthernet0/0 [7] 5d01h: OSPF: Send DBD to 208.186.96.11 on FastEthernet0/0 seq 0xAA opt 0x42 flag 0x7 len 32 5d01h: OSPF: Retransmitting DBD to 208.186.96.11 on FastEthernet0/0 [8] 5d01h: OSPF: Send DBD to 208.186.96.11 on FastEthernet0/0 seq 0xAA opt 0x42 flag 0x7 len 32 5d01h: OSPF: Retransmitting DBD to 208.186.96.11 on FastEthernet0/0 [9] You can see where the other OSPF sessions are fine. The ARC cycles from a EXSTART state to a DOWN state indefinitely. edge1.slc#sh ip ospf neig Neighbor ID Pri State Dead Time Address Interface 208.186.96.11 0 EXSTART/DROTHER 00:00:31 208.186.96.11 FastEthernet0/ 0 208.186.96.10 0 FULL/DROTHER 00:00:32 208.186.96.10 FastEthernet0/ 0 192.168.55.1 2 FULL/DROTHER 00:00:31 208.186.96.21 FastEthernet0/ 0 208.186.100.73 100 FULL/DR 00:00:36 208.186.96.15 FastEthernet0/ 0 Any advice would be greatly appreciated.