frame-relay
n r3 r4 r5 . all router use subinterface ,r3 use multipoint , r4,r5 use point-to-point
//这里主要留意碰到frame map 0.0.0.0的问题，必须要处理掉。我考的时候没有遇到这个情况，
如果是其他不需要的map，可以用以下方法处理:
sh frame map
记下没用DLCI号
进端口
sh
no enc fram
enc fram
no fram inv
no arp fram
fram map ip x.x.x.x 000 b
no fram inv ip 000(没用的DLCI）
no fram inv ip 000（同上）
。。。。
都NO过一遍后
NO SH
就可以了
//留意题目可能会预配置其他FRAM封装模式，如ANSI等，注意到时接口是否UP了
Frame Relay Traffic Shaping：
记得在接口下要开启：frame-relay traffic-shaping
burst up to the maximum trasmission rate：Be=(AR-CIR)*TC/1000
bursting should not be allowed：Be=0
the Frame Relay notifies the router of congestion it should reduce its sending rate to no lower than 384Kbps：
frame-relay mincir 384000
frame-relay adaptive-shaping becn
Any FECNs received should be reflected as BECNs：
frame-relay fecn-adapt
- 要在物理接口开启（即子接口的主接口）启用FRTS：frame-relay traffic-shaping
- 一般tc不配置，由cir和bc自动计算得出
- 单位：cir：bit/s，mincir：bit/s，Bc：bits，Be：bits，Tc：milliseconds
• To resolve a problem configure your network so that Microsoft SQL traffic is limited to an average rate of 256Kbps on R2’s connection to the Frame Relay cloud.
• Up to 2048 SQL packets in excess of 256Kbps should be queued up by R2 before packet loss occurs.
| ip cef |
To resolve this configure R1’s traffic shaping queue to hold 10 times the default amount of packets.
| map-class frame-relay FRTS |
Rack1R4#show frame-relay pvc 401 | beg Queueing
Queueing strategy: fifo
Output queue 0/400, 0 drop, 0 dequeued
To decrease the serialization delay on the circuit ensure that all the shaping intervals are the smallest possible, and that a single packet cannot take more than one interval to be transmitted.
| map-class frame-relay DLCI_403 Rack1R3#show frame-relay pvc 304 |
| Lowest Link Speed in Path | Recommended Fragmentation Size(for 10 ms Serialization) |
| 56 Kbps | 70 bytes |
| 256 Kbps | 320 bytes |
Congestion Aviodance：random-detect precedence [precedence] [开始随机丢弃] [开始尾部丢弃] [轮询时间]
In order to prevent this behavior configure R1 to start dropping packets with an IP precedence of routine on this link when there are at least 15 packets in the output queue
| interface FastEthernet0/0 |
附：ip优先级：0 routine，1 priority ，2 immediate，3 flash，4 flash-override，5 critical，6 internet，7 network，0优先级最低
• Configure your network so that HTTP traffic is dropped prior to the interface becoming congested.
• This HTTP traffic should not be reserved any bandwidth.
| class-map match-all NOT_HTTP |
• In order to prevent this type of tail drop behavior for voice traffic configure R4 to randomly drop packets on the Frame Relay circuit before congestion occurs.
• In order to ensure that voice traffic gets better service than other traffic configure R4 so that ‘critical’ traffic will not be dropped unless there are 60 packets in the output queue.
• If there are 90 critical packets in the output queue R4 should randomly drop 5 out of every 25 of these packets.
• In the case that there are more than 90 critical packets in the output queue they should all be dropped.
| policy-map WRED |
对输出队列带宽进行整形，还有其他几项也是 Show出来的结果，真正能改动的是这个75的值，默认是 100．
| SW3: |
R2 上配置两个 Class-map,一个叫 backbone2,要求输入的接口是 E0 的数据带宽为128K,shaped 256K;另外一个叫做 precedence,设置优先级为 4 的数据分配带宽为 384K,peak 512K
| r2 |
4.2 DE
打开DE位：Router(config)#frame-relay de-list number protocol procotol
把DE列表应用到接口上：Router(config-if)#frame-relay de-group number
The frame-relay link on R5 is experiencing heavy congestion; configure R5 so that the frame-relay provider doesn’t drop any routing protocol packets during congestion. And if the number of packets in R5’s frame-relay interface queue exceeds 10 then the traffic rate will be reduced to 32000bps
| access-list 192 deny ospf any any |
4.3 Congestion Avoiding
Configure R1 so such that if outbound traffic has utilized 75% of total bandwidth , R1 should
signal that the network is congested and the recipients need to slow down sending packets .
do not use frame-relay BECN and FECN
| RackYYR1 |
NBAR
Clients on vlan_bb1 and vlan_55 access a url located on vlan_bb2 frequently , this url is http://www.thiswebsite.com/directory/ select one suitable router to configure so as to conserve bandwidth meeting the g0/0
Traffic from this url to these clients should not exceed 64000bit/s . if the files download from this url are image files then drop the traffic . you may assume image file name end with the suffix .gif .jpg or .jpeg
| RackYYR6 |
