The figure below illustrates a simple VPN scenario for an ISP. This network consists of a core MPLS VPN with four PE devices belonging to three VPNs: red, blue, and green. From the perspective of device PE1, these VPNs are reachable remotely through BGP next hop devices PE2 (device ID: 10.10.10.5), PE3 (device ID: 10.10.10.7), and PE4 (device ID: 10.10.10.8).
Figure 4. Network Used for LSP Health Monitor Example
The following example shows how to configure operation parameters, proactive threshold monitoring, and scheduling options on PE1 (see the figure above) using the LSP Health Monitor. In this example, the LSP discovery option is enabled for LSP Health Monitor operation 1. Operation 1 is configured to automatically create IP SLAs LSP ping operations for all BGP next hop neighbors (PE2, PE3, and PE4) in use by all VRFs (red, blue, and green) associated with device PE1. The BGP next hop neighbor process is enabled, and the time interval at which routing entries that are no longer valid are removed from the BGP next hop neighbor discovery database is set to 60 seconds. The time interval at which the LSP Health Monitor checks the scan queue for BGP next hop neighbor updates is set to 1 minute. The secondary frequency option is enabled for both connection loss and timeout events, and the secondary frequency is set to 10 seconds. As specified by the proactive threshold monitoring configuration, when three consecutive connection loss or timeout events occur, an SNMP trap notification is sent. Multioperation scheduling and the generation of IP SLAs SNMP system logging messages are enabled.
PE1 Configuration
mpls discovery vpn interval 60
mpls discovery vpn next-hop
!
auto ip sla mpls-lsp-monitor 1
type echo ipsla-vrf-all
timeout 1000
scan-interval 1
secondary-frequency both 10
!
auto ip sla mpls-lsp-monitor reaction-configuration 1 react connectionLoss threshold-type consecutive 3 action-type trapOnly
auto ip sla mpls-lsp-monitor reaction-configuration 1 react timeout threshold-type consecutive 3 action-type trapOnly
ip sla traps
snmp-server enable traps rtr
!
auto ip sla mpls-lsp-monitor schedule 1 schedule-period 60 start-time now
The following is sample output from the
show
ip
sla
mpls-lsp-monitor
configuration command for PE1:
PE1# show ip sla mpls-lsp-monitor configuration 1
Entry Number : 1
Modification time : *12:18:21.830 PDT Fri Aug 19 2005
Operation Type : echo
Vrf Name : ipsla-vrf-all
Tag :
EXP Value : 0
Timeout(ms) : 1000
Threshold(ms) : 5000
Frequency(sec) : Equals schedule period
LSP Selector : 127.0.0.1
ScanInterval(min) : 1
Delete Scan Factor : 1
Operations List : 100001-100003
Schedule Period(sec): 60
Request size : 100
Start Time : Start Time already passed
SNMP RowStatus : Active
TTL value : 255
Reply Mode : ipv4
Reply Dscp Bits :
Secondary Frequency : Enabled on Timeout
Value(sec) : 10
Reaction Configs :
Reaction : connectionLoss
Threshold Type : Consecutive
Threshold Count : 3
Action Type : Trap Only
Reaction : timeout
Threshold Type : Consecutive
Threshold Count : 3
Action Type : Trap Only
The following is sample output from the
show
mpls
discovery
vpn command for PE1:
PE1# show mpls discovery vpn
Refresh interval set to 60 seconds.
Next refresh in 46 seconds
Next hop 10.10.10.5 (Prefix: 10.10.10.5/32)
in use by: red, blue, green
Next hop 10.10.10.7 (Prefix: 10.10.10.7/32)
in use by: red, blue, green
Next hop 10.10.10.8 (Prefix: 10.10.10.8/32)
in use by: red, blue, green
The following is sample output from the
show
ip
sla
mpls-lsp-monitor
neighbors command for PE1:
PE1# show ip sla mpls-lsp-monitor neighbors
IP SLA MPLS LSP Monitor Database : 1
BGP Next hop 10.10.10.5 (Prefix: 10.10.10.5/32) OK
ProbeID: 100001 (red, blue, green)
BGP Next hop 10.10.10.7 (Prefix: 10.10.10.7/32) OK
ProbeID: 100002 (red, blue, green)
BGP Next hop 10.10.10.8 (Prefix: 10.10.10.8/32) OK
ProbeID: 100003 (red, blue, green)
The following is sample output from the
show
ip
sla
mpls-lsp-monitor
scan-queue
1 and
debug
ip
sla
mpls-lsp-monitor commands when IP connectivity from PE1 to PE4 is lost. This output shows that connection loss to each of the VPNs associated with PE4 (red, blue, and green) was detected and that this information was added to the LSP Health Monitor scan queue. Also, since PE4 is no longer a valid BGP next hop neighbor, the IP SLAs operation for PE4 (Probe 10003) is being deleted.
PE1# show ip sla mpls-lsp-monitor scan-queue 1
Next scan Time after: 20 Secs
Next Delete scan Time after: 20 Secs
BGP Next hop Prefix vrf Add/Delete?
10.10.10.8 0.0.0.0/0 red Del(100003)
10.10.10.8 0.0.0.0/0 blue Del(100003)
10.10.10.8 0.0.0.0/0 green Del(100003)
PE1# debug ip sla mpls-lsp-monitor
IP SLAs MPLSLM debugging for all entries is on
*Aug 19 19:48: IP SLAs MPLSLM(1):Next hop 10.10.10.8 added in DeleteQ(1)
*Aug 19 19:49: IP SLAs MPLSLM(1):Removing vrf red from tree entry 10.10.10.8
*Aug 19 19:56: IP SLAs MPLSLM(1):Next hop 10.10.10.8 added in DeleteQ(1)
*Aug 19 19:56: IP SLAs MPLSLM(1):Next hop 10.10.10.8 added in DeleteQ(1)
*Aug 19 19:49: IP SLAs MPLSLM(1):Removing vrf blue from tree entry 10.10.10.8
*Aug 19 19:49: IP SLAs MPLSLM(1):Removing vrf green from tree entry 10.10.10.8
*Aug 19 19:49: IP SLAs MPLSLM(1):Removing Probe 100003
The following is sample output from the
show
ip
sla
mpls-lsp-monitor
scan-queue
1 and
debug
ip
sla
mpls-lsp-monitor commands when IP connectivity from PE1 to PE4 is restored. This output shows that each of the VPNs associated with PE4 (red, blue, and green) were discovered and that this information was added to the LSP Health Monitor scan queue. Also, since PE4 is a newly discovered BGP next hop neighbor, a new IP SLAs operation for PE4 (Probe 100005) is being created and added to the LSP Health Monitor multioperation schedule. Even though PE4 belongs to three VPNs, only one IP SLAs operation is being created.
PE1# show ip sla mpls-lsp-monitor scan-queue 1
Next scan Time after: 23 Secs
Next Delete scan Time after: 23 Secs
BGP Next hop Prefix vrf Add/Delete?
10.10.10.8 10.10.10.8/32 red Add
10.10.10.8 10.10.10.8/32 blue Add
10.10.10.8 10.10.10.8/32 green Add
PE1# debug ip sla mpls-lsp-monitor
IP SLAs MPLSLM debugging for all entries is on
*Aug 19 19:59: IP SLAs MPLSLM(1):Next hop 10.10.10.8 added in AddQ
*Aug 19 19:59: IP SLAs MPLSLM(1):Next hop 10.10.10.8 added in AddQ
*Aug 19 19:59: IP SLAs MPLSLM(1):Next hop 10.10.10.8 added in AddQ
*Aug 19 19:59: IP SLAs MPLSLM(1):Adding vrf red into tree entry 10.10.10.8
*Aug 19 19:59: IP SLAs MPLSLM(1):Adding Probe 100005
*Aug 19 19:59: IP SLAs MPLSLM(1):Adding ProbeID 100005 to tree entry 10.10.10.8 (1)
*Aug 19 19:59: IP SLAs MPLSLM(1):Adding vrf blue into tree entry 10.10.10.8
*Aug 19 19:59: IP SLAs MPLSLM(1):Duplicate in AddQ 10.10.10.8
*Aug 19 19:59: IP SLAs MPLSLM(1):Adding vrf green into tree entry 10.10.10.8
*Aug 19 19:59: IP SLAs MPLSLM(1):Duplicate in AddQ 10.10.10.8
*Aug 19 19:59: IP SLAs MPLSLM(1):Added Probe(s) 100005 will be scheduled after 26 secs over schedule period 60