One side of a node on RPR ring was doing planning work, and the other side of the node, customer engineer performed a error operation which caused the whole RPR ring melt down, customer experience loss of service.
From the logfile, it has been discovered that all the nodes on the RPR ring are suffering for link up and down very frequently hence customer experience loss of service.
1.Pulled out the fibre we found the interface was 'Down' which means the interface was working.
2.checked the config, WTR was configured.
3.Changed the fibre and the problem still occurs.
4.Customer use their spare parts and changed the chasis, there is still a problem.
5.Shut the peer interface down mannually to make WTR timer working, and the problem disappeared.
1. Fibre was faulty.
2. interface was affected due to error manoeuvre.
3.network was attacked so RPR could not be established.
4.As RPR is not Wraped, maybe WTR timer was not set properly.
It had found out that the fibre and transmission between the nodes has problem, some RPR topology message was not able to be sent fuuly, so the wrong message running across the whole network. so shut down the peer interface had made the WTR timer working hence cured this problem.
PS: WTR timer applies when the topology establishes.