input rate

Is it '5 minute input rate' value from the show int output equal to the actual traffic (in last 5 minutes) rate that arrives on some interface? I have contracted traffic rates with my both ISP's and show int always indicates that the rate is for about 200 Kbps less then the contracted CIR value should be when I load Internet links to the max. so now I'm wondering if I'm maybe reading those values wrongly or both ISP's are breaking LSA:).

Best Regards, Igor

Reply to
Iggy
Loading thread data ...

In article , Iggy wrote: :Is it '5 minute input rate' value from the show int output equal to the :actual traffic (in last 5 minutes) rate that arrives on some interface?

If I recall correctly, it is a weighted average rather than a literal count.

:I have contracted traffic rates with my both ISP's and show int always :indicates that the rate is for about 200 Kbps less then the contracted CIR :value should be when I load Internet links to the max. so now I'm wondering :if I'm maybe reading those values wrongly or both ISP's are breaking LSA:).

What percentage loss is that 200 Kbps ?

Reply to
Walter Roberson

It is indeed an average

Reply to
<Anthrax>

Weighted average. It's weighted to limit the peaks and valleys from affecting the numbers too much.

Reply to
Hansang Bae

So, how then can you measure/calculate actual transfer rate? I use MRTG and max. value it shows is same as in 'show int' command output...Also, on LAN side I have configured traffic shaping which should shape inbound traffic from the Internet. Shape value is set to 1900000 bps, contracted CIR with ISP is 2000000 bps and while the link should be under heavy load there is only few packets holded in the queue by traffic shaper. When I set lower shaping value (1700 Kbps), shaping is active almost constantly...

Reply to
Iggy

Hello,

Personally I found MRTG as the most useful tool to measure the load (since MIB are constantly updated and keep historical record). Just to wide a little bit more in your previous query ....

"Average number of bits and packets transmitted per second in the last 5 minutes. If the interface is not in promiscuous mode, it senses network traffic that it sends and receives (rather than all network traffic). The 5-minute input and output rates should be used only as an approximation of traffic per second during a given 5-minute period. These rates are exponentially weighted averages with a time constant of 5 minutes. A period of four time constants must pass before the average will be within 2 percent of the instantaneous rate of a uniform stream of traffic over that period.

Note The 5-minute period referenced in this output is a load interval that is configurable under the interface. The default value is

5 minutes. "

formatting link
About the traffic shaping issue, if you configured traffic shaping becn makes sense that the queue will not hold packets unless you receive becn's .

formatting link
what shows up in the show frame-relay pvc and show traffic-shape statistics when is under heavy load?

Reply to
<Anthrax>

Thanks for explanation about measuring traffic rate... I remembered that I saw it somewhere, but it was disapeared from my "bio flash memory":)

Regarding shaping you did'nt understood me well, I'm not using FR traffic shaping, but CBTS and this is configured on LAN side since on this way I'm able to use CBWFW on FE interface for 2 mbps Internet traffic:) since I first shape traffic for a little bit lower then actual CIR (if CIR is 2 mbps, then I shape it on 1.9 Mbps) and then simply nest CBWFQ inside traffic shaping. Without traffic shaping queue will always fill up on ISP side. With shaping, queue fills on my router (interface trough which Internet traffic flows outbound into my LAN) and I'm able to give some more priority to certain Internet applications... And it really works:)

B.R. Igor

formatting link

formatting link

--------------------------------------------------------------------------------

Reply to
Iggy

Cabling-Design.com Forums website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.