27
You can use iperf2 or iperf3 to help generate some traffic. There are quite a few options included that will allow you to accomplish some nice traffic classification.
You might also check out scapy - specifically a packet former utility. Allows you to define values on each field to get really granular with how traffic is being formed and sent.
In my lab, I ...
18
QoS is an umbrella term which covers the use of features such as policing, shaping, traffic classification, and advanced queuing mechanisms.
CoS is a form of QoS limited to layer-2 Ethernet and it uses 3-bits (8 values) of the 802.1Q tag to differentiate your traffic. Therefore no trunking, no CoS.
DSCP is the most commonly acted upon value at layer-3 and ...
17
Cisco has something they call Cisco Validated Designs (CVD). These are network designs from different perspectives like QoS, high availability, WAN etc that Cisco recommends and have tested.
For QoS they have 3 different designs with 4, 8 or 12 classes which can be seen in the picture below.
Starting with the 4 class model:
! This section defines the 4-...
17
Class of Service is a layer 2 identifier, usually in conjunction with an 802.1q tag (You shouldn't see CoS referenced on an access port, only a trunk port). DiffServ would be the corresponding identifier on a layer 3 link. Quality of Service is a superset of both controlling how both layer 2 and layer 3 links classify, police, and queue traffic on ...
15
There are a few examples that can do this.
Cacti is one. It has a weather-map plugin that can be leveraged to produce output such as the following
GigaPop Illinois
Wisconsin University
These are examples of what the weather map plugin can do. When configured correctly you can see minute, hourly, daily, weekly and monthly traffic stats.
You can download ...
14
I think this depends heavily on the platform on which you want to accomplish this. For example, IOS uses NBAR to classify transit traffic, primarily for QoS and security mechanisms. But as far as I know, NBAR QoS configurations are entirely static.
For centralized monitoring, NetFlow is probably your best bet.
13
Try this:
show platform hardware interface ten1/1/31 statistics
That should show you input bytes by CoS.
The 4500 is a very different beast than the 3560/3750 for qos config and monitoring. Some other useful verification commands are:
show platform hardware qos interface foo X/Y
shows queue lengths and flow counts
show interface foo X/Y counter detail
...
13
You definitely should enable MLS QoS. It is also prerequisite for CoPP, which you should add in your TODO list to implement.
Enabling 'mls qos' without other config is extremely bad idea in low-end cats, like 3560/3750, due to unexpected default scheduling and due to heavily reduced buffers causing more microbursting.
On 7600/6500 it's comparatively safe ...
13
Even though dropbox uses AWS, they can be blocked...
Blocking Dropbox
I use an address-based approach for stuff like this, just lookup the address blocks that said company owns and filter it...
Using Robtex information for AS19679 (Dropbox) to block dropbox...
object-group network DROPBOX_AS19679
network-object 108.160.160.0 255.255.240.0
network-...
10
I would use vlan-based policing which works better on these switches. This is an example matching a speed value of 48Mb
mls qos
!
interface GigabitEthernet1/0/2
switchport access vlan 500
switchport mode access
mls qos vlan-based
!
class-map match-all CUSTOMER_1
match input-interface GigabitEthernet1/0/2
!
policy-map VLAN500_POLICE
class CUSTOMER_1
...
10
Your question's pretty broad. There's a lot of different commands you can use to troubleshoot and monitor QoS, so I'll focus on the primary question you have, which is how to reasonably verify your QoS configuration is working and how to read the policy-map interface output.
The only true way to verify that QoS is working is to hook up a traffic generator ...
10
Weighted Fair Queueing (WFQ) is as the name implies a queueing algorithm. Queueing is used when there is congestion on an interface. This is usually detected through that the Transmit Ring (TX-Ring) is full. This means that the interface is busy sending packets. Queuing does not take place unless there is congestion on the interface. In some cases the size ...
10
IPv4 has the DSCP (formerly ToS) field, and IPv6 has the Traffic Class field. Those are just fields in the IP headers. IP itself does not have any QoS. QoS is implemented by network devices, often based on those fields in the IP headers. The fields are to facilitate QoS, but QoS in not part of IP.
For example, a VoIP phone could mark its packets as EF (...
10
The easy way to do it could be:
On your WLC home page you should have something like:
You can reset the stat through the gear icon on the top right corner, selecting "Clear Dashboard Data"
Then the machine should quickly appear in the "Top Client Devices" list, you can click on it and you will get the IP address and the mac address of the device.
Bonus ...
9
There is no way for receiver to affect scheduling of sender. You should just work on getting QoS configured at sender.
Theoretically if the frames are correctly classified and your receiver and sender supports 802.1Qbb (essentially CoS aware pause frame) then you could request pausing of BE traffic without causing pausing of EF traffic.
9
You could create a second match condition in the class-map matching all source IP networks you want to block (with an ACL). Any requests to youtube.com from a source IP not matched by this ACL will not be dropped.
9
The key is the 'match-all' or 'match-any' part of the class-map. You may configure the class-map either way.
class-map {match-any | match-all} *class-map name*
If you do a "match-all" class-map, all of the match conditions must be true in order for the traffic to match. As Jeremy mentioned, creating an ACL matching the particular users and matching that ...
9
It will cause zero disruption. I've done it in the middle of the day before.
Any policies will remain, but ignored
9
Unfortunately, I don't think there's a good way to do what you want without the provider's involvement. Working within that restriction, your best bet may be to implement an outbound policy on your LAN interface. For example, if you are PATing the business network and guest networks separately, so that the return traffic can be identified by destination IP,...
9
For Cisco IP phones, you've definitely got the right configuration:
!
mls qos
!
interface UPLINK
! Unconditionally trust all QoS markings
auto qos voip trust
!
interface ACCESS
! Look for the Cisco Phone's CDP frames, and trust VoIP from the phone...
auto qos voip cisco-phone
Should I put "auto qos voip trust" on network uplinks for both directions ...
8
7600 QoS is not straightforward.
If you enable mls qos on its own you will change the way buffers are allocated on all ports, you may find this gives you unexpected results.
Different line cards will have different queueing capabilities, research the following commands to modify the buffer and queueing config. Some line cards are per port and some are ...
8
I have also used netperf in some situations. It seems to perform better with UDP-tests.
8
This platform specific limit, so it's just arbitrary limit Cisco came up with at one point in time, probably driven by memory restrictions of device of that era.
W=K/(precedence +1)
K= 4096 with Cisco IOS 12.0(4)T and earlier releases, and 32384 with
12.0(5)T and later releases.
From:
http://www.cisco.com/en/US/tech/tk543/tk545/...
8
First of all, don't believe everything you read on the Internet ;-)
Sometimes algorithms (or the way they are physically implemented) don't fit neatly into a theoretical category. What you call it is less important than understanding what it does.
The whole point of WFQ (or any other scheduling algorithm) is to share the limited link bandwidth among the ...
8
1.Would it be better to setup traffic policing?
You're using shaping, which is better in this case. The important thing is to buffer your traffic before sending it to your internet provider. Consider this:
Your ISP asks you to set your physical interface to 100Mbps
Your service from the ISP is 30Mbps
What happens if you're already sending 30Mbps of ...
8
I'd recommend migrating from old-school CAR to modern config of CBWFQ. When you do that, you can nest QoS policies together to accomplish more complicated QoS scenarios. Per this Cisco doc, GRE keepalives are marked with CS6. The keepalive is also a part of the GRE packets, not the inner IP packets that are being sent (source).
class-map CM-CS6
match ...
8
VoIP over the public Internet can be a problem, but it usually works good enough, most of the time, although there can be times where it sucks. Most ISPs have extra cost features where they will honor some of your QoS markings and policies.
(I know Verizon Business, among others, has some specific packages for QoS, and you may need to adjust your policies ...
7
Can you provide some more information on the equipment used and how much bandwidth you have?
Will your voice equipment mark packets itself or do you need to take care of it? Usually CoS 5 and/or DSCP EF is used to carry voice payload. Voice should be sent in some form of LLQ (Low Latency Queue).
If you give us some more specifics we could help with ...
7
There are various tools dependant on platform. For example:
Cisco - NBAR
SonicWALL - AppFlow
Juniper - No specific option in JunOS, though can roughly be achieved using class maps.
For monitoring of traffic Netflow & NFSEN is your friend. The bundle simplifies deployment of the NFDUMP capture, dump and analysis tools and gives you a nice frontend to ...
Only top voted, non community-wiki answers of a minimum length are eligible
