QOS Help needed - Tandberg VC dropouts between sites

[carl_townshend]

Hi all

we have an issue with VC across the sites, we have an MPLS wan, with 4 meg between HQ and the remote location, when making VC calls between the sites the connection is dropping, the video seems a little fuzzy too, there is no congestion on the WAN, its prob averaging about 1 Meg.

We have provided 1 meg in the priority queue for H323 and RTP traffic as well as matched on the IP's of the vc units, however its made it no better.

what will be affecting the video traffic ? ,

what should i be looking at first ?

is there any IP SLA monitoring i can do ?

here is the policy map interface stats

Service-policy output: MPLS-QOS

    Class-map: class-default (match-any)
      115164776 packets, 42959774951 bytes
      5 minute offered rate 269000 bps, drop rate 2000 bps
      Match: any
      Queueing
      queue limit 64 packets
      (queue depth/total drops/no-buffer drops) 0/283930/0
      (pkts output/bytes output) 115274416/43233135701
      shape (average) cir 4000000, bc 40000, be 0
      target shape rate 4000000

      Service-policy : MPLS-Voip-Qos

        queue stats for all priority classes:
          Queueing
          queue limit 64 packets
          (queue depth/total drops/no-buffer drops) 0/18316/0
          (pkts output/bytes output) 13072247/3319597462

        Class-map: VOIP (match-any)
          13090563 packets, 3343168212 bytes
          5 minute offered rate 62000 bps, drop rate 0 bps
          Match: access-group name VC-VOIP
            782067 packets, 253130323 bytes
            5 minute rate 0 bps
          Match: protocol sip
            0 packets, 0 bytes
            5 minute rate 0 bps
          Match: protocol rtp audio
            11071037 packets, 2413437159 bytes
            5 minute rate 0 bps
          Match: protocol h323
            1217294 packets, 652993119 bytes
            5 minute rate 62000 bps
          Match: protocol rtp video
            20164 packets, 23607175 bytes
            5 minute rate 0 bps
          Priority: 1024 kbps, burst bytes 25600, b/w exceed drops

cheers

Carl

[Sergey Fer]

What speed does your Tandberg negotiate in VC? Consider that LLQ (priority queue) is policed and all traffic that exceeds 1024 kbps is to be dropped. I'm not sure that IP SLA will help you here.

Another thing to consider is an overall delay, jitter and drops in channel. Here you may use IP SLA, of course.

[Joseph W. Doherty]

Disclaimer

The   Author of this posting offers the information contained within this   posting without consideration and with the reader's understanding that   there's no implied or expressed suitability or fitness for any purpose.   Information provided is for informational purposes only and should not   be construed as rendering professional advice of any kind. Usage of  this  posting's information is solely at reader's own risk.

Liability Disclaimer

In   no event shall Author be liable for any damages whatsoever (including,   without limitation, damages for loss of use, data or profit) arising  out  of the use or inability to use the posting's information even if  Author  has been advised of the possibility of such damage.

Posting

Both HQ and remote have same physical bandwidth links and same logical bandwidth guarantees?  Are there more than just the two sites?  What are the physical interfaces?

[carl_townshend]

I believe they are running at 384k which is not that much!

The remote site has a 4 meg link, the HQ has a 10 meg link, both on MPLS.

They both put this trafic in the LLQ policed at 1 meg.

The remote site is a fast ethernet port and so is the HQ that plugs into the WAN.

Any ideas?  any settings on the Tandberg i can change?

We know that our MPLS has no QOS through the cloud.

what delay is video OK with ?

cheers

Carl

[Joseph W. Doherty]

Disclaimer

The    Author of this posting offers the information contained within this    posting without consideration and with the reader's understanding that    there's no implied or expressed suitability or fitness for any  purpose.   Information provided is for informational purposes only and  should not   be construed as rendering professional advice of any kind.  Usage of  this  posting's information is solely at reader's own risk.

Liability Disclaimer

In    no event shall Author be liable for any damages whatsoever  (including,   without limitation, damages for loss of use, data or  profit) arising  out  of the use or inability to use the posting's  information even if  Author  has been advised of the possibility of such  damage.

Posting

The remote site is a fast ethernet port and so is the HQ that plugs into the WAN.

Running at 100 Mbps or 10 Mbps?

what delay is video OK with ?

For VC, usually about the same as VoIP, i.e. about 150 ms one-way.

Are there just these two sites or more?  If more, how many more and what are their bandwidth connections, both physical and logical?  Also if more, site-to-site possible or all traffic between hub and remote site(s)?

[carl_townshend]

here is the IP SLA stats

IPSLAs Latest Operation Statistics

IPSLA operation id: 10
Type of operation: udp-jitter
        Latest RTT: 232 milliseconds
Latest operation start time: 15:20:26.661 UTC Thu May 31 2012
Latest operation return code: OK
RTT Values:
        Number Of RTT: 958              RTT Min/Avg/Max: 230/232/247 milliseconds
Latency one-way time:
        Number of Latency one-way Samples: 958
        Source to Destination Latency one way Min/Avg/Max: 118/119/128 milliseconds
        Destination to Source Latency one way Min/Avg/Max: 111/112/129 milliseconds
Jitter Time:
        Number of SD Jitter Samples: 947
        Number of DS Jitter Samples: 953
        Source to Destination Jitter Min/Avg/Max: 0/2/7 milliseconds
        Destination to Source Jitter Min/Avg/Max: 0/2/17 milliseconds
Packet Loss Values:
        Loss Source to Destination: 38          Loss Destination to Source: 4
        Out Of Sequence: 0      Tail Drop: 0
        Packet Late Arrival: 0  Packet Skipped: 0
Voice Score Values:
        Calculated Planning Impairment Factor (ICPIF): 29
MOS score: 3.30
Number of successes: 10
Number of failures: 0
Operation time to live: Forever

[Joseph W. Doherty]

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

About 250 ms RTT, how remote is remote?  (I.e. how far apart are these two sites; international?)

Your overall latency is pushing the margins for good quality VC, this also is reflected in the MOS score but from your described symptoms, I would lean toward drops and/or jitter issues.

Unfortunately, difficult to make suggestions without  better understanding your environment, and difficult to understand your  environment without the information I've asked for.

For example, the reason I've asked twice about number of sites, MPLS, if a L3 VPN, is often configured for multipoint - very easy to bump into congestion upon MPLS cloud egress.  For MPLS multipoint, normal solution is to take advantage of MPLS QoS that might be provided by the vendor, although it's possible to micro manage bandwidths between sites.

[paul driver]

hi carl,

Have you tried increasing your tc vlaue from 10m to 125m = bc 500000

res

Paul

[carl_townshend]

Hi there

What do you mean to increase the tc value ?

[paul driver]

Your over all traffic shaping is currently set to 4mb with a tc ( time interval of bc = burst) 0.10 ms giving you a bc of 40000

I am wondering if you increase this time interval (tc) to 0.125ms which will give you a increased burst of 50000 may assist this issue.

res

Paul

[carl_townshend]

Hi

We already police the rtp traffic to 1 meg in the priority queue, so how would that help ?