Weird IP/TCP/ISIS issue

[Chidinma Iwe]

Hi guys,

I just ran into an issue between two cisco ios-xr boxes, where apparently due to TCP retransmission issue, ISIS, BGP & LDP all went down.

Prevalent in my logs were these errors: %IP-TCP_NSR-5-DISABLED, %ROUTING-LDP-5-NSR_SYNC_PEER_LOST

Has anyone experienced this scenario before?

Thanks

Chris

[Nagendra Kumar Nainar]

Chris,

While any issue with TCP might bring down the BGP and LDP session, I dont think it should affect or brign down ISIS session. It appears that the control plane protocols lost the packet and so the session.

What error did you see for BGP/ISIS and LDP while bringing the session to neighbor(s) down?.

-Nagendra

[Chidinma Iwe]

Hi Nagendra,

Correct, I would expect  only p2p reachability /physical /L2 issues that could have affected the ISIS session.

Below are some logs from one of the IOS-XR boxes (P router in my MPLS environment):

RP/0/RP0/CPU0:Sep  1 08:11:41.076 WAT: isis[1001]: %ROUTING-ISIS-5-ADJCHANGE : Adjacency to NG-IKJ-A9K-PE-01 (Bundle-Ether13) (L2) Down, Holdtime expired
RP/0/RP0/CPU0:Sep  1 08:12:05.347 WAT: isis[1001]: %ROUTING-ISIS-5-ADJCHANGE : Adjacency to NG-IKJ-A9K-PE-01 (Bundle-Ether13) (L2) Up, New adjacency
RP/0/RP0/CPU0:Sep  1 08:12:35.349 WAT: isis[1001]: %ROUTING-ISIS-5-ADJCHANGE : Adjacency to NG-IKJ-A9K-PE-01 (Bundle-Ether13) (L2) Down, Holdtime expired
RP/0/RP0/CPU0:Sep  1 08:14:33.883 WAT: tcp[354]: %IP-TCP_NSR-5-DISABLED : 41.75.94.176:646 <-> 41.75.94.181:15183:: NSR disabled for TCP connection because Retransmission threshold exceeded
RP/0/RP0/CPU0:Sep  1 08:14:33.886 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NSR_SYNC_PEER_LOST : Peer 41.75.94.181:0 synchronization lost
RP/0/RP0/CPU0:Sep  1 08:14:50.990 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 1 peers
RP/0/RP0/CPU0:Sep  1 08:14:52.433 WAT: isis[1001]: %ROUTING-ISIS-5-ADJCHANGE : Adjacency to NG-IKJ-A9K-PE-01 (Bundle-Ether13) (L2) Up, New adjacency
RP/0/RP0/CPU0:Sep  1 08:15:20.994 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 1 peers
RP/0/RP0/CPU0:Sep  1 08:15:22.143 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-GR : GR session 41.75.94.181:0 (instance 1) disconnected
RP/0/RP0/CPU0:Sep  1 08:15:22.147 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NBR_CHANGE : Neighbor 41.75.94.181:0, DOWN (Received Notification message from peer: KeepAlive Timer Expired)
RP/0/RP0/CPU0:Sep  1 08:15:24.967 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-GR : GR session 41.75.94.181:0 (instance 2) reconnected
RP/0/RP0/CPU0:Sep  1 08:15:24.969 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NBR_CHANGE : Neighbor 41.75.94.181:0, UP
RP/0/RP0/CPU0:Sep  1 08:15:50.999 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 1 peers
RP/0/RP0/CPU0:Sep  1 08:15:52.018 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NSR_SYNC_PEER_DONE : Peer 41.75.94.181:0 initial synchronization done
RP/0/RP0/CPU0:Sep  1 08:15:52.019 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-NSR_SYNC_DONE : Initial synchronization successfully done for 1 peers
RP/0/RP0/CPU0:Sep  1 08:22:58.902 WAT: mpls_ldp[1038]: %ROUTING-LDP-5-SESSION_PROTECTION : Session hold up initiated for peer 41.75.94.181:0
RP/0/RP0/CPU0:Sep  1 08:23:01.955 WAT: isis[1001]: %ROUTING-ISIS-5-ADJCHANGE : Adjacency to NG-IKJ-A9K-PE-01 (Bundle-Ether13) (L2) Down, Holdtime expired

- Chris

[ปลาวาฬทราย RMUTT CPE IX]

What is the log?

RP/0/RSP0/CPU0:Jan 4 17:50:50 BKK: mpls_ldp[1048]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 1 peers

RP/0/RSP0/CPU0:Jan 5 12:10:47 BKK: mpls_ldp[1048]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 8 peers

RP/0/RSP0/CPU0:Jan 6 09:22:17 BKK: mpls_ldp[1048]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 2 peers

RP/0/RSP0/CPU0:Jan 6 09:22:23 BKK: mpls_ldp[1048]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 1 peers

RP/0/RSP0/CPU0:Jan 6 09:22:53 BKK: mpls_ldp[1048]: %ROUTING-LDP-5-NSR_SYNC_START : Initial synchronization started for 5 peers

Does it impact to any service?