In Customer Network, one day, there was LCAS Switching unsuccessfully at OSN3500 equipment, which led the BFD of OSPF protocol interrupted.
The network topology is attached.
When Customer cut out the fiber of B link separately, the services were interrupted for few seconds.
The version info:
BIOS 08.09.46T02 20060521 18:25:29 inactive
ExtBios 09.09.46T02 20060810 11:58:38 active
NeSoft(D) 126.96.36.199 20070416 14:15:18
Platform(T) 5.00.07.C03T13 20070328 22:13:59
AosSoft(T) 5.99.02.10T26 20070228 09:35:38
PCB Version : SSN1EMS4 VER.C
BIOS Version : 2.25
ExtBIOS Version : 2.25
Software Version : 4.13
Logic Version : (U82)130(U63)150
We do a serveral test in the lab and Confirm it.
The LCAS switching time may be more than 150ms
There are two solution for this issue:
1. Modify the BFD detection time of OSPF protocol at router side, the specific time can be test from the real network.
2. Use the SNCP protection instead of LCAS protection and use the EFT board instead of EMS4 board at same time.
Please find the attachment for details.
According to analysis of datacom experts, Customer configure the BFD detection time of OSPF protocol is five times and interval is 30 ms, i.e. the switching time
must be less than 150 ms for the transmission network, otherwise the OSPF protocol running on the router will be down. After the OSPF down, it will take several
seconds to up again. The services will be interrupted during the term.
But the switching time of LCAS may be more than 150 ms.
On the scenario of add and delete services on the VCTRUNK, the LCAS function can be hitless for the services (The bandwidth should no more than the real
bandwidth provided by the VC channels after delete the services);but not hitless for the fiber cut down.