This article explains troubleshooting and isolating latency issues encountered in networks utilizing Juniper devices.

When pinging device B from device A, we observed significant latency increases on particular hops, suggesting potential bottlenecks:

Topology:

root@A> traceroute 11.11.11.11
traceroute to 11.11.11.11 (11.11.11.11), 30 hops max, 52 byte packets

 1  R1 (1.1.1.1)  19.236 ms  17.832 ms  18.266 ms
 2  R2 (2.2.2.2)  143.103 ms  91.284 ms  88.848 ms
 3  R3 (3.3.3.3)  89.522 ms  88.968 ms  88.943 ms
 4  R4 (4.4.4.4)  91.202 ms  88.297 ms  92.201 ms
 .
 .
 .
 .  
 9  R9 (9.9.9.9)  89.522 ms  88.968 ms  88.943 ms
 10 R10 (10.10.10.10) 88.801 ms  180.676 ms  88.808 ms
 11 B (11.11.11.11)  91.202 ms  88.297 ms  92.201 ms

1. Execute a traceroute command towards the destination (B) to examine the path and the number of hops between the source and destination.
   
   NOTE: Traceroute provides latency information based on ICMP ping replies, which are control plane packets. However, this may not correlate with the data plane latency experienced by actual traffic, as it can be affected by factors like CPU load on the individual devices.
   
   
2. To isolate the latency, measure the ping latency from the second hop (R1) to the destination (B). If high latency is observed, the issue likely lies between R1 and R2.
   
   
3. To further diagnose the issue, log in to the router (R1 and R2) and verify the ingress and egress interfaces for errors/discard/drops. 
   
   
4. Continue this process for each subsequent hop in the path, analyzing ping latency to isolate the problem area.