07-02-2012 01:45 AM
Hi guys,
Our customer has a WAN network (DCoS) and has voice protected across it. I work for an integrator assisting the customer with their CPE routers end-to-end. Essentially CPE routers extend PSTN lines end-to-end over voice-ports and voip call legs between locations. On CPE routers we're seeing voice traffic being marked/matched end-to-end. We're pretty sure that the ISP is somehow shaping/causing delays on the voice traffic interstate but now we're trying to gather some evidence.
To help isolate, test and provide evidence we were planning on using UDP-Jitter in IP SLA. I've been trying to get this enabled on live routers without success due to the reason "No connection" in the initiators show output. An example of the "show ip sla statistics" on a production device is shown below:
SITE-1#show ip sla statistics
IPSLAs Latest Operation Statistics
IPSLA operation id: 100
Type of operation: udp-jitter
Latest RTT: NoConnection/Busy/Timeout
Latest operation start time: 17:26:47 BNE Mon Jul 2 2012
Latest operation return code: No connection
RTT Values:
Number Of RTT: 0 RTT Min/Avg/Max: 0/0/0 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 SD Jitter Samples: 0
Number of DS Jitter Samples: 0
Source to Destination Jitter Min/Avg/Max: 0/0/0 milliseconds
Destination to Source Jitter Min/Avg/Max: 0/0/0 milliseconds
Packet Loss Values:
Loss Source to Destination: 0
Source to Destination Loss Periods Number: 0
Source to Destination Loss Period Length Min/Max: 0/0
Source to Destination Inter Loss Period Length Min/Max: 0/0
Loss Destination to Source: 0
Destination to Source Loss Periods Number: 0
Destination to Source Loss Period Length Min/Max: 0/0
Destination to Source Inter Loss Period Length Min/Max: 0/0
Out Of Sequence: 0 Tail Drop: 0
Packet Late Arrival: 0 Packet Skipped: 0
Voice Score Values:
Calculated Planning Impairment Factor (ICPIF): 0
Mean Opinion Score (MOS): 0
Number of successes: 0
Number of failures: 1
Operation time to live: Forever
Number of failures continues to increase every 30 seconds or so. The configuration of SITE-1 is as follows:
SITE-1#show run | sec ip sla
ip sla 100
udp-jitter 10.2.2.2 16584 source-ip 10.1.1.1 source-port 16384
tos 30
owner NetFlow
timeout 60000
ip sla schedule 100 life forever start-time now
The configuration of SITE-2 (IP SLA Responder) is as follows:
SITE-2#show run | sec ip sla
ip sla responder
For the sake of privacy I've adjusted the IPs above as well as the router hostnames. The 10.2.2.2 IP lives on SITE-2's router and 10.1.1.1 lives on SITE-1's router.
If I do the above in GNS it works perfectly so I think something else is going on here. Either I've got a bug (doubt it) or something is preventing the UDP from connecting/responding for the IP SLA. Frustratingly, if I do a "show ip sla responder" on the SITE-2 router... it says it received the packets normally without error:
SITE-2#show ip sla responder
General IP SLA Responder on Control port 1967
General IP SLA Responder is: Enabled
Number of control message received: 123 Number of errors: 0
Recent sources:
10.1.1.1 [18:06:35.989 BNE Mon Jul 2 2012]
10.1.1.1 [18:06:30.989 BNE Mon Jul 2 2012]
10.1.1.1 [18:06:25.989 BNE Mon Jul 2 2012]
10.1.1.1 [18:05:35.989 BNE Mon Jul 2 2012]
10.1.1.1 [18:05:30.989 BNE Mon Jul 2 2012]
Recent error sources:
Permanent Port IP SLA Responder
Permanent Port IP SLA Responder is: Enabled
udpEcho Responder:
IP Address Port
If I enable trace debug ("debug ip sla trace") on SITE-1's router I get the following...
Jul 2 08:13:25.993: IPSLA-INFRA_TRACE:OPER:100 slaSchedulerEventWakeup
Jul 2 08:13:25.993: IPSLA-INFRA_TRACE:OPER:100 Starting an operation
Jul 2 08:13:25.993: IPSLA-OPER_TRACE:OPER:100 Starting jitter operation
Jul 2 08:13:25.993: IPSLA-OPER_TRACE:OPER:100 Ctrl msg: id=116, type=1, len=52, dest_ip=10.2.2.2, enablePort=16584, duration=60200
Jul 2 08:13:25.993: IPSLA-OPER_TRACE:OPER:100 table_id=0, topo_id=65535
Jul 2 08:13:30.993: IPSLA-OPER_TRACE:OPER:100 Timeout
Jul 2 08:13:30.993: IPSLA-OPER_TRACE:OPER:100 Ctrl msg: id=117, type=1, len=52, dest_ip=10.2.2.2, enablePort=16584, duration=60200
Jul 2 08:13:30.993: IPSLA-OPER_TRACE:OPER:100 table_id=0, topo_id=65535
Jul 2 08:13:35.993: IPSLA-OPER_TRACE:OPER:100 Timeout
Jul 2 08:13:35.993: IPSLA-OPER_TRACE:OPER:100 Ctrl msg: id=118, type=1, len=52, dest_ip=10.2.2.2, enablePort=16584, duration=60200
Jul 2 08:13:35.993: IPSLA-OPER_TRACE:OPER:100 table_id=0, topo_id=65535
Jul 2 08:13:40.993: IPSLA-OPER_TRACE:OPER:100 Timeout
Jul 2 08:13:40.993: IPSLA-OPER_TRACE:OPER:100 No connection
Jul 2 08:13:40.993: IPSLA-INFRA_TRACE:OPER:100 Updating result
09-03-2012 08:33 AM
I am having the exact same problem with one wierd twist:
Site A is a 3925 running c3900-universalk9-mz.SPA.150-1.M1.bin that cannot connect to any of my other sites ( I get the same error messages as you "NoConnection/Busy/Timeout")
Site B is a 3925 running c3900-universalk9-mz.SPA.152-1.T1.bin that can connect to ANY other site, including site A
Site C is a 2921 running c2900-universalk9-mz.SPA.152-1.T1.bin that can connect to ANY other site, including site A
Site D is a 2911 running c2900-universalk9-mz.SPA.150-1.M3.bin that can connect to ANY other site, including site A
I am in the process of upgrading the IOS on site A to c3900-universalk9-mz.SPA.152-1.T1.bin because this is the only logincal thing I can think of that might be causing this problem. Once my change management ticket is complete I will update this post (probably week of 09/10/12).
09-03-2012 03:30 PM
I actually forgot I lodged this in the first place. The problem was resolved for me by removing the specific ports in the IP SLA:
ITE-1#show run | sec ip sla
ip sla 100
udp-jitter 10.2.2.2 16584 source-ip 10.1.1.1
tos 30
owner NetFlow
timeout 60000
ip sla schedule 100 life forever start-time now
Essentially, my IP SLA was being generated by ManageEngine. ManageEngine was defining a "port". That port configuration though valid was not working. CIsco TAC advised to removed the specific port definition.... this resolved it.
09-03-2012 04:35 PM
Thanks for the update.
I tried doing that, but my VoIP gateway is requiring a port number:
voipgw01(config-ip-sla)# udp-jitter 10.2.2.2 source-ip 10.11.1.1
^
% Invalid input detected at '^' marker.
voipgw01(config-ip-sla)# udp-jitter 10.215.50.5 ?
<0-65535> Port Number. For codec options recommended port range
is an Even Port range between 16384-32767 or 49152-65535
voipgw01(config-ip-sla)# udp-jitter 10.2.2.2 source-ip 10.11.1.1 codec g711ulaw codec-numpackets 100
^
% Invalid input detected at '^' marker.
Cisco 2911 running c2900-universalk9-mz.SPA.150-1.M3
voipgw01#show ip sla appl
IP Service Level Agreement Technologies
IPSLAs Infrastructure version: Engine-II
Supported Operation Types:
802.1ag Echo, 802.1ag Jitter, dhcp, dns, frame relay, ftp
http, icmp-echo, icmp-jitter, path-echo, path-jitter, rtp
tcp-connect, udp-echo, udp-jitter, voip
I think my next step is Cisco TAC.
Thanks again for the reply!
09-03-2012 06:19 PM
David - had this same timeout issue and just fixed my issue 5 min ago, problem was a miss configured source-ip.
Can you post the router configuration ?
R1#show ip sla mon stat
Round trip time (RTT) Index 11
Latest RTT: 112 ms
Latest operation start time: *01:06:19.287 UTC Fri Mar 1 2002
Latest operation return code: OK
Number of successes: 21
Number of failures: 0
Operation time to live: Forever
R1#show debug
IP SLA Monitor:
TRACE debugging for all operations is on
ERROR debugging for all operations is on
*Mar 1 01:06:19.283: IP SLA Monitor(11) Scheduler: Starting an operation
*Mar 1 01:06:19.287: IP SLA Monitor(11) echo operation: Sending an echo operation
*Mar 1 01:06:19.399: IP SLA Monitor(11) echo operation: RTT=112
*Mar 1 01:06:19.399: IP SLA Monitor(11) Scheduler: Updating result
09-03-2012 06:25 PM
I know source IP is ok because there is only one IP on the router and I successfully connect to 4 other voice gateways using the same source IP.
What code are you running on your VoIP gateway? Do you have more than one IP SLA sourcing from that device?
I really think there is something wrong with the responder. I can successfully create IP SLA operations to the other devices, but cannot to this one. Any ideas about the responder configuration?
09-03-2012 06:28 PM
The responder will respond to "anything" it supports. Another trick I remember reading through these comments is that you have to be extremely careful that the responder isn't responding to the same port from different IP SLA tests. i.e. each site must be configured to use a unique port number.
EDIT: This was the reason why Cisco recommended letting IP SLA select the port.
09-03-2012 06:36 PM
Mine is running in GNS3 - disregard the wrong date and time
09-03-2012 06:53 PM
Checked the ports and found I am using different ones on each VoIP gateway. I just added a new operation with a totally unique port that I know is not in use on the source or destination and it still failed:
voipgw02(config)#ip sla 4033
voipgw02(config-ip-sla)# udp-jitter 10.205.50.5 17333 source-ip 10.225.50.5 codec g711ulaw codec-numpackets 100
voipgw02(config-ip-sla-jitter)# tos 184
voipgw02(config-ip-sla-jitter)# timeout 18000
voipgw02(config-ip-sla-jitter)# threshold 1000
voipgw02(config-ip-sla-jitter)# frequency 30
voipgw02(config-ip-sla-jitter)#ip sla schedule 4033 life forever start-time now ageout 3600
voipgw02#sh ip sla stat
IPSLAs Latest Operation Statistics
IPSLA operation id: 4033
Type of operation: udp-jitter
Latest RTT: NoConnection/Busy/Timeout
Latest operation start time: 21:47:43 EDT Mon Sep 3 2012
Latest operation return code: No connection
RTT Values:
Number Of RTT: 0 RTT Min/Avg/Max: 0/0/0 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 SD Jitter Samples: 0
Number of DS Jitter Samples: 0
Source to Destination Jitter Min/Avg/Max: 0/0/0 milliseconds
Destination to Source Jitter Min/Avg/Max: 0/0/0 milliseconds
Packet Loss Values:
Loss Source to Destination: 0
Source to Destination Loss Periods Number: 0
Source to Destination Loss Period Length Min/Max: 0/0
Source to Destination Inter Loss Period Length Min/Max: 0/0
Loss Destination to Source: 0
Destination to Source Loss Periods Number: 0
Destination to Source Loss Period Length Min/Max: 0/0
Destination to Source Inter Loss Period Length Min/Max: 0/0
Out Of Sequence: 0 Tail Drop: 0
Packet Late Arrival: 0 Packet Skipped: 0
Voice Score Values:
Calculated Planning Impairment Factor (ICPIF): 0
Mean Opinion Score (MOS): 0
Number of successes: 0
Number of failures: 1
Operation time to live: Forever
Do you know if there are any licensing issues?
barvoipgw02#show ip sla appl
IP Service Level Agreements
Version: Round Trip Time MIB 2.2.0, Infrastructure Engine-III
Supported Operation Types:
icmpEcho, path-echo, path-jitter, udpEcho, tcpConnect, http
dns, udpJitter, dhcp, ftp, VoIP, rtp, icmpJitter
802.1agEcho VLAN, Port, 802.1agJitter VLAN, Port, udpApp
wspApp
Supported Features:
IPSLAs Event Publisher
IP SLAs low memory water mark: 63126085
Estimated system max number of entries: 46234
Estimated number of configurable operations: 39313
Number of Entries configured : 3
Number of active Entries : 3
Number of pending Entries : 0
Number of inactive Entries : 0
Time of last change in whole IP SLAs: .21:47:43.954 EDT Mon Sep 3 2012
09-03-2012 06:54 PM
Contrary to my original response, I am in fact defining ports on my live routers.
I have a mix of 2911 and 2921 where I am running this on. Mix of universal and old codes. In the example I provide below, both sites are using universal and have "uc" licensing enable. That might your issue... do a "show ver"? Do you have "uc" license enabled to get udp-jitter?
A router can be a responder and tracker at the same time. My head office "test" router has the below config:
HeadOffice#show run | sec ip sla
ip sla responder
ip sla 103
udp-jitter
16584 source-ip source-port 16500 codec g729a advantage-factor 4 tos 46
owner NetFlow
timeout 60000
ip sla schedule 103 life forever start-time 16:13:38 Jul 25 ageout 60
ip sla 105
udp-jitter
16584 source-ip source-port 16404 codec g729a advantage-factor 4 tos 46
owner NetFlow
timeout 60000
ip sla schedule 105 life forever start-time now ageout 60
[...] (ip sla threshold config suppressed)
The remote ends just have:
Remote_2911#show run | sec ip sla
ip sla responder
As an example show ver:
[...]
Technology Package License Information for Module:'c2900'
-----------------------------------------------------------------
Technology Technology-package Technology-package
Current Type Next reboot
------------------------------------------------------------------
ipbase ipbasek9 Permanent ipbasek9
security None None None
uc uck9 Permanent uck9
data None None None
Configuration register is 0x2102
Hopefully that helps. IP sla responder config is extremely simple... I doubt that is the issue. Probably more likely ports/config/licensing. Another useful mechanism to troubleshoot is show udp:
Remote_2911#show udp
Proto Remote Port Local Port In Out Stat TTY OutputIF
17 --listen-- --any-- 1975 0 0 1000001 0
[...]
17
16400 16584 0 0 441 0
09-03-2012 07:05 PM
I could be totally wrong with the license. I seem to remember that being a sticking point back when I looked at this.
09-03-2012 07:57 PM
Router R1 has 4 interfaces, I'm sourcing from 2 Ethernet Interfaces and routing packets from each source out the other 2 Serial ports, but each source goes over specific paths.
I have two ACL's, Route-Maps, and IPSLA's configured. I take packets from host 1 and route them out serial 0/0, and take packets from host 2 and routing them out serial 0/1. Both host 1 and host 2 have ACL's allowing them to talk to host 3 on a distant network. Currently studying for CCNP routing, so my config is much simpler than what is shown in your posts. My "NoConnection/Busy/Timeout" issue was missconfigured source-ip on the R1 router. Sorry couldn't be more help.
09-04-2012 01:21 PM
Thanks for all of the feedback. I am going to open a TAC case and see what they have to say. Once I have their solution I will post it here.
08-07-2014 01:19 PM
The same problem here. And your post helped... in a certain way.
The strangest thing is that, I configed SLA jitter from router A to B, it worked, and then used the exact same config(with necessary changes) from router B to A, it won't work, and gave me the " no connection" error.
I would really appreciate any input, since I was stuck here for too long.
Thanks
08-07-2014 01:54 PM
I never did get it working on the original routers, I used a spare ISR whose only mission was to run IPSLA. As a matter of fact, I only created the operation in one direction because I am able to monitor the "Bidirectional Jitter," with that one operation, which was sufficient for me. I can still see packet loss, MOS, latency, and jitter for all of my operations. If you really need to create an operation in each direction, I would use two different routers or two different source/destination IPs on each router. I think the addition of the second operation fails when the router sees that there is an existing operation established between the same source and destination IPs, regardless of the port numbers that are being used.
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: