Workbook 1 QOS - 10.24 MQC Single-Rate Three-Color Policer
My questions is regarding the Bc and Be values used in the Solution guide:
"Ensure the burst size is large enough to accommodate normal and excess burst durations of 200ms and 300ms at a rate of 128Kbps."
When I did this I configured this as police "128000 3200 1600" but the solution guide uses "police 128000 3200 4800"
If this should be 200ms + 300ms for a total of 500ms then I agree with the SG but if this is 200ms normal burst that can grow to 300ms during inactivity or 300ms initial. Either way Bc and Be are different buckets so they are added together:
Rack1R4#ping 155.1.146.6 rep 2 size 982 time 0
Type escape sequence to abort.
Sending 2, 982-byte ICMP Echos to 155.1.146.6, timeout is 0 seconds:
..
Success rate is 0 percent (0/2)
Rack1R6#sh policy-map int
GigabitEthernet0/0.146
Service-policy input: POLICE
Class-map: ICMP (match-all)
2 packets, 2000 bytes
5 minute offered rate 0 bps, drop rate 0 bps
Match: access-group name ICMP
police:
cir 8000 bps, bc 1000 bytes, be 1000 bytes
conformed 1 packets, 1000 bytes; actions:
drop
exceeded 1 packets, 1000 bytes; actions:
drop
I think it needs to be a bit clearer and state the total burst should not accumulate more than ether 500ms or 300ms of credit.
Nick
Comments
Task 10.26
Use the CIR value of 64Kbps and PIR value of 128Kbps.
Use the values of CIR*400ms and PIR*200ms for normal and excess burst sizes.
Hi Nick,
There is something wrong with your number. Here is the formula for Be.
128K * 300/1000/8bits
= (128 * 300)/8bits
= 38400/8
Be = 4,800 bytes
=======================================
By the way, the task number has changed.
If you are searching the sub-forum for this task, the old task number is QoS Section 10.42
QoS sub-forum
Thanks Joe
My Be is 1600 because I am assuming 300ms burst total and I already have 3200 configured for Bc so 3200 + 1600 = 4800.
I guess its just "wording" or one for the proctor...
Nick
I think the wording for 10.42 is fine. They give us the Tc numbers to use. We just need to plug them into the equation.
As far as the second task, there is a problem with the numbers vs the solution.
Note that INE changed the numbering 10.26 was 10.44 in the forum threads.
Here is one of the threads acknowledging this.
WB-I QoS sub-forum
Re: 10.44 MQC two-rate three-color policer
Awesome thanks again Joe
Yes could be my interpretation at fault. It made me understand things better and feel more confident about it which is all good.
Nick
Yes, I get a little confused with a couple of things (wording) in the equations for shaping vs policing. bits vs bytes shape vs peak vs rate
Your lab date is almost here, right? Have you been doing any of the INE mock labs or troubleshooting labs?
Yes D-Day is Tuesday the 3rd.
Have been hitting the TS-Labs a lot and that part has been going really well. I am also finishing off my second run through Vol 1 + my notes (Got security, QoS, management and services to complete). I have not done my Vol 2 labs lately. My last one was a couple of weeks ago (Lab 17). I do have an INE Mock Lab booked for next week (Lab 6) along with more TS Labs then I will have some time spare for last minute weaknesses/practise/speed drills.
I feel doing a lot of labs at this stage can be hit and miss as it can only tes you on certain things where as redoing Vol 1 you cover a ot of ground. That's the plan anyways.
This will be attempt #2.
Nick