02-03-2022 01:14 PM
I need to configure CBWFQ on a Hub-router with a 250meg link, there are 40 Field offices with 10 offices with10meg each, 10 offices with 20meg each, 10 offices with 30meg links. The hub router is over subscribing some to the remote offices at any giving time. I have been ask to configure QOS to limit the Hub to only deliver/transmit 75% of 10, 20 and 30megs to each corresponding FOs, this way the remote offices are never over subscribes with incoming traffic.
I am working on a cisco ASR1002-X with IOS ver. "asr1002x-universalk9.03.16.06.S.155-3.S6-ext.SPA.bin".
I need a solution as soon as possible.
Thanks in advance.
02-03-2022 11:22 PM
M.
02-11-2022 05:56 AM
Good morning,
Thank you for the link, but I already visited that page a few weeks ago, and I don't think that it helped to accomplish the task discribed abave. I am new in this field of QOS implementation.
02-11-2022 09:58 AM
". . . only deliver/transmit 75% . . ."
Why not 100%?
". . . Hub-router with a 250meg link . . ."
Physical gig hand-off? What about exceeding the 250 Mbps?
". . . Hub-router with a 250meg link . . ."
What about remote sites (aggregate egress) overrunning hub's (ingress) bandwidth?
Unsure your ASR will support 2 level shaping, but if it does, you want something (syntax might be incorrect) like:
policy-map parent
class class-default
shape average 250000000
service policy child
policy-map child
class remoteoffice1
shape average 10000000 (or 20 or 30 Mbps)
class remoteoffice2
shape average 10000000 (or 20 or 30 Mbps)
.
.
class remoteoffice39
shape average 10000000 (or 20 or 30 Mbps)
class remoteoffice40
shape average 10000000 (or 20 or 30 Mbps)
BTW, you note ". . . there are 40 Field offices with 10 offices with10meg each, 10 offices with 20meg each, 10 offices with 30meg links."
10 and 10 and 10 offices equal 30 offices, not 40?!
Is this student project?
02-14-2022 08:54 AM
02-14-2022 03:18 PM
#1 With QoS shaping, you should be able to run at 100%. However, on some platforms I suspect Cisco shapers don't "count" L2 overhead. If they don't, they you do need to shape slower than your nominal speed; usually about 15% slower allows for "typical" L2 overhead (whose percentage varies based on packet size).
#2 Hmm, I'm guessing the 1.5 gig is some measure usage, over some time (often 5 minutes). If so, you can still have microburst, causing transient issues.
BTW, unsure your platform will support a 3 level policy, but even better yet would be something like:
policy-map childFQ
class class-default
fair-queue
policy-map child
class remoteoffice1
shape average 10000000 (or 20 or 30 Mbps)
service policy childFQ
class remoteoffice2
shape average 10000000 (or 20 or 30 Mbps)
service policy childFQ
.
.
class remoteoffice39
shape average 10000000 (or 20 or 30 Mbps)
service policy childFQ
class remoteoffice40
shape average 10000000 (or 20 or 30 Mbps)
service policy childFQ
policy-map parent
class class-default
shape average 250000000
service policy child
02-15-2022 06:50 AM
02-15-2022 08:50 AM - edited 02-15-2022 10:19 AM
You'll want a class for each remote office. Generally, the class-map, for each office will match destination addresses used by the office. (As you noted having 40 offices, I created class names from 1 to 40. However, you could name the classes [and their map] for something more representive for the offices.)
Correct! Your high level shaper should be for 2 Gbps (oops, I missed changing that with your revised information).
Also correct, you apply the high level parent policy on the egress interface. (Which, also, if desired, you can rename the polices too.)
BTW, how this works, the parent policy sets a limit for bandwidth usage for the interface as a whole (to whatever bandwidth you've contracted for - if contracted bandwidth is full interface bandwidth, you don't need the parent policy and its shaper).
Each office class, sets a limit for bandwidth to each remote site. Their shapers insure you don't overrun that office's bandwidth.
Also BTW, the above works great, for a hub network. If sites might send traffic between themselves, doing QoS on just your equipment becomes problematic. Generally, for multipoint WAN networks, your WAN provider needs to allow you to define QoS in their network for your requirement. (I don't know if this has changed, but in the past, most MPLS WAN providers could provide some QoS support based on your requirements. MetroE providers, again, at least in the past, often did not provide such QoS support.)
Lastly, such a QoS policy can be used to further analyze bandwidth needs.
02-15-2022 10:03 AM
02-16-2022 08:41 AM
02-16-2022 02:25 PM
"Sorry, but did you see the sample config I sent you yesterday?"
No, I didn't (and don't).
"Sorry, but did you see the sample config I sent you yesterday?"
Doesn't shape any traffic, which you need, and probably overly complex for your needs.
02-16-2022 04:00 PM
02-17-2022 07:49 AM
I would very, very (very) much recommend, first trying FQ (if supported on your platform - especially if platform supports additional FQ settings), as shown in one of my posts.
If that doesn't work for you, the childFQ policy-map is where you would want to "adjust" bandwidth allocations for different traffic.
02-17-2022 03:12 PM
02-17-2022 03:48 PM
See policy-map child.
Each class matches to a remote site.
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide