Re: IP SLA not capturing packet loss

jzhang135 · ‎11-02-2010

Hi All,

Currently my company uses Cisco 1811's for our sites which goes through a provider to our primary router a 3825. The existing IP SLAs we have in place are icmp-jitter types, which isn't capturing packet loss. So I have created a new IP SLA to do just that. IP SLA 10 is the existing one, and 30 is the new one. The 3825 has the ip sla responder activated. Yesterday we experience a site which had a rather high percent packet loss when ping'ed (around 8% loss), so I thought it would be a rather good time to test. But it wasn't detecting any packet loss, did I enter something wrong?

ip sla 10

icmp-jitter 10.210.1.8 num-packets 20 interval 40

frequency 600

ip sla schedule 10 life forever start-time now

ip sla 30

udp-jitter 10.210.1.8 5000

frequency 300

ip sla schedule 30 life forever start-time now

ip sla reaction-configuration 10 react packetLoss threshold-type immediate threshold-value 5 1 action-type trapOnly

ip sla logging traps

snmp-server enable traps syslog

yjdabear · ‎11-02-2010

How did you determine that "it wasn't detecting any packet loss"? Was there no entry in the logging buffer (show log) on the 1811 (IP SLA source), or no SNMP trap received on your SNMP server?

>> ip sla reaction-configuration 10 react packetLoss threshold-type immediate threshold-value 5 1 action-type trapOnly

Shouldn't the 10 above be 30 instead?

jzhang135 · ‎11-02-2010

Well using ping from the 1811 to the 3825, 1000 times, we would get anywhere from 30 to 80 packet loss. I look at IP SLA stat and shows nothing for packet loss. And yes that 10 is 30, sorry (it's configured correctly as 30 in the router).

jzhang135 · ‎11-04-2010

Bump for ANY advice?

ccarring · ‎11-04-2010

Hi,

I think a trace is in order here. The udp jitter operation is using port 5000, what port is used by the cli ping?

Please post a trace of the traffic with the operation running, and running a failing ping operation; alsoi include the cli output showing the ip sla stat showing zero loss.

Thanks,

Curtis

jzhang135 · ‎11-04-2010

Thanks for replying. I have no idea what port the ping command is using (I didn't know you can specify what port ping uses). We have already fixed the packet loss issue (duplex mismatch). But here's the ip sla stat, values under Packet Loss Values were all 0 when we were experiencing packet loss as well.

Round Trip Time (RTT) for Index 30

Latest RTT: 10 milliseconds

Latest operation start time: *20:20:52.002 UTC Thu Nov 4 2010

Latest operation return code: OK

RTT Values:

Number Of RTT: 10 RTT Min/Avg/Max: 10/10/10 milliseconds

Latency one-way time:

Number of Latency one-way Samples: 0

Source to Destination Latency one way Min/Avg/Max: 0/0/0 milliseconds

Destination to Source Latency one way Min/Avg/Max: 0/0/0 milliseconds

Jitter Time:

Number of Jitter Samples: 9

Source to Destination Jitter Min/Avg/Max: 1/1/1 milliseconds

Destination to Source Jitter Min/Avg/Max: 1/1/1 milliseconds

Packet Loss Values:

Loss Source to Destination: 0 Loss Destination to Source: 0

Out Of Sequence: 0 Tail Drop: 0 Packet Late Arrival: 0

Voice Score Values:

Calculated Planning Impairment Factor (ICPIF): 0

Mean Opinion Score (MOS): 0

Number of successes: 10

Number of failures: 0

Operation time to live: Forever

To further my question in my original post, how would one implement packet loss SLA and send trap to SNMP server when there is packet loss?

EDIT: Today I noticed a site dropping packets, not as much as the one I mentioned before, here is the ping result and ip sla stat:

Router#ping 10.210.1.8 r 100

Type escape sequence to abort.

Sending 100, 100-byte ICMP Echos to 10.210.1.8, timeout is 2 seconds:

!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!!!.!!!!!!!!!!!!!!!!!

!!!!!!!!!!!!!!!!!!!!!!.!!!!!!!

Success rate is 96 percent (96/100), round-trip min/avg/max = 24/27/40 ms