04-04-2012 02:56 AM
Hello,
I'm having an intermittent issue with LDP sessions dropping between one specific router and it's two peers.
Solution background:
We have 3 x 7201 (NPE-G2) running c7200p-spservicesk9-mz.150-1.M5.bin.
The routers are arranged in a "triangle" with a link to each router utilising G0/2 + 3 on each router, IGP is OSPF with all routers + loopbacks being in OSPF Area 0. OSPF RID + MPLS RID is Lo0.
Routers are called ABC-CORE-1, CCC-CORE-1, CBC-CORE1.
During the "outage" the DLP session between ABC->CBC remains constant without issue, however the LDP sessions with CCC-CORE1 both drop stating the following:
ABC-CORE-1#show log
Apr 3 17:30:40.236: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.2:0 (1) is DOWN (Received error notification from peer: Holddown time expired)
Apr 3 17:30:44.180: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.2:0 (1) is UP
CCC-CORE-1#show log
.Apr 3 17:30:39: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.1:0 (1) is DOWN (Discovery Hello Hold Timer expired)
.Apr 3 17:30:43: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.1:0 (3) is UP
Neighbor output:
CCC-CORE-1#sh mpls ldp neigh 10.64.255.1 det
Peer LDP Ident: 10.64.255.1:0; Local LDP Ident 10.64.255.2:0
TCP connection: 10.64.255.1.646 - 10.64.255.2.36718
Password: not required, none, in use
State: Oper; Msgs sent/rcvd: 1191/1192; Downstream; Last TIB rev sent 130
Up time: 16:24:42; UID: 10; Peer Id 2;
LDP discovery sources:
GigabitEthernet0/2; Src IP addr: 10.64.0.1
holdtime: 15000 ms, hello interval: 5000 ms
Addresses bound to peer LDP Ident:
10.64.2.1 192.168.128.126 10.64.0.1 10.64.0.5
10.64.255.1 10.64.254.1
Peer holdtime: 180000 ms; KA interval: 60000 ms; Peer state: estab
Clients: Dir Adj Client
Capabilities Sent:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
Capabilities Received:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
ABC-CORE-1#show mpls ldp neigh 10.64.255.2 det
Peer LDP Ident: 10.64.255.2:0; Local LDP Ident 10.64.255.1:0
TCP connection: 10.64.255.2.36718 - 10.64.255.1.646
Password: not required, none, in use
State: Oper; Msgs sent/rcvd: 1193/1191; Downstream; Last TIB rev sent 130
Up time: 16:25:05; UID: 8; Peer Id 0;
LDP discovery sources:
GigabitEthernet0/2; Src IP addr: 10.64.0.2
holdtime: 15000 ms, hello interval: 5000 ms
Addresses bound to peer LDP Ident:
10.64.2.5 192.168.128.254 10.64.0.2 10.64.0.9
10.64.255.2 10.64.254.2
Peer holdtime: 180000 ms; KA interval: 60000 ms; Peer state: estab
Clients: Dir Adj Client
Capabilities Sent:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
Capabilities Received:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
I've just enabled MPLS IGP sync this morning aftere reading a few articles however I'm not convicnced that's going to make much difference, has anyone ever experieced this before?
Thanks,
Duncan.
04-11-2012 04:59 AM
Duncan Mossop wrote:
Hello,
I'm having an intermittent issue with LDP sessions dropping between one specific router and it's two peers.
Solution background:
We have 3 x 7201 (NPE-G2) running c7200p-spservicesk9-mz.150-1.M5.bin.
The routers are arranged in a "triangle" with a link to each router utilising G0/2 + 3 on each router, IGP is OSPF with all routers + loopbacks being in OSPF Area 0. OSPF RID + MPLS RID is Lo0.
Routers are called ABC-CORE-1, CCC-CORE-1, CBC-CORE1.
During the "outage" the DLP session between ABC->CBC remains constant without issue, however the LDP sessions with CCC-CORE1 both drop stating the following:
ABC-CORE-1#show log
Apr 3 17:30:40.236: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.2:0 (1) is DOWN (Received error notification from peer: Holddown time expired)
Apr 3 17:30:44.180: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.2:0 (1) is UP
CCC-CORE-1#show log
.Apr 3 17:30:39: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.1:0 (1) is DOWN (Discovery Hello Hold Timer expired)
.Apr 3 17:30:43: %LDP-5-NBRCHG: LDP Neighbor 10.64.255.1:0 (3) is UP
Neighbor output:
CCC-CORE-1#sh mpls ldp neigh 10.64.255.1 det
Peer LDP Ident: 10.64.255.1:0; Local LDP Ident 10.64.255.2:0
TCP connection: 10.64.255.1.646 - 10.64.255.2.36718
Password: not required, none, in use
State: Oper; Msgs sent/rcvd: 1191/1192; Downstream; Last TIB rev sent 130
Up time: 16:24:42; UID: 10; Peer Id 2;
LDP discovery sources:
GigabitEthernet0/2; Src IP addr: 10.64.0.1
holdtime: 15000 ms, hello interval: 5000 ms
Addresses bound to peer LDP Ident:
10.64.2.1 192.168.128.126 10.64.0.1 10.64.0.5
10.64.255.1 10.64.254.1
Peer holdtime: 180000 ms; KA interval: 60000 ms; Peer state: estab
Clients: Dir Adj Client
Capabilities Sent:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
Capabilities Received:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
ABC-CORE-1#show mpls ldp neigh 10.64.255.2 det
Peer LDP Ident: 10.64.255.2:0; Local LDP Ident 10.64.255.1:0
TCP connection: 10.64.255.2.36718 - 10.64.255.1.646
Password: not required, none, in use
State: Oper; Msgs sent/rcvd: 1193/1191; Downstream; Last TIB rev sent 130
Up time: 16:25:05; UID: 8; Peer Id 0;
LDP discovery sources:
GigabitEthernet0/2; Src IP addr: 10.64.0.2
holdtime: 15000 ms, hello interval: 5000 ms
Addresses bound to peer LDP Ident:
10.64.2.5 192.168.128.254 10.64.0.2 10.64.0.9
10.64.255.2 10.64.254.2
Peer holdtime: 180000 ms; KA interval: 60000 ms; Peer state: estab
Clients: Dir Adj Client
Capabilities Sent:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
Capabilities Received:
[Dynamic Announcement (0x0506)]
[Typed Wildcard (0x0970)]
I've just enabled MPLS IGP sync this morning aftere reading a few articles however I'm not convicnced that's going to make much difference, has anyone ever experieced this before?
Thanks,
Duncan.
For anyone that's interested I fixed this problem after some in depth troubleshooting.
Turns out the CCC router has CEF disabled.
I followed the through the following process:
Noticed flushes in the inteface counters -> researched SPD (selective packet discard) -> looked into CPU usage -> noticed IP Input process running 90%+, checked CEF with "show ip cef" showed a full adjancency table.
Checked the command "show cef interface"
Boom, "CEF Switching disabled" on every interface.
"no ip cef" "ip cef" , fixed.
For reference the running config did indeed show "ip cef" as entered and turned on.
Only thing I can think is when we upgraded from 12.4 to 15.0(1)M there was a glitch with CEF....
Very strange.
04-12-2012 05:44 AM
Tht's great..I have just started reading the mpls..your question will help me to enhance my skills..thanks for ur question and ans too..I really some help of u..pls reach me at ajaysartaj@gmail.com
Sent from Cisco Technical Support iPhone App
Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: