- 积分
- 0
- 鸿鹄币
- 个
- 好评度
- 点
- 精华
- 注册时间
- 2011-7-1
- 最后登录
- 1970-1-1
- 阅读权限
- 0
- 听众
- 收听
游客
|
楼主 |
发表于 2011-7-15 13:02:57
|
显示全部楼层
154TK中,下面这段描述很有用,我没怎么看,但出来后,发现BGP的CASE变种,跟这个有点相似,提醒大家注意!!!8 z5 M# o$ g8 c7 C! A
. T5 _, p" ~* a) }1 xNew BGP question:
1 y& _# V& u4 K5 O% k h! BFirst question is about BGP, involving only 3 of the routers of the topology7 o8 \# d8 }) O, o
R1 ------ R2 -------R36 K; A1 e: O& \. _, H
Shows u the routing table of R1 and partial BGP configuration. R1 Routing Table has 3* I' x$ y5 Q! J- E
routes, one of them is matched by an access list applied to a community setting for+ K1 j5 K; A# \
neighbor R2. This community is NOT ADVERTISE, which means that the router
: J+ s. i( u9 P& e: ]receiving this route (R2) can use it, but not advertise it to any neighbor. The question" g3 D% F3 m4 E/ S+ G2 K
is, which routes will R3 have on its BGP table after this. I was wrong and choose the2 k% s% K* K9 \9 M w3 r' d
option that it will only have the other 2 routes, because I supposed that the question. D6 Y! Q7 H# D( P9 Y) P% I9 b6 O
was related to communities, but I tell you what I really think; R3 will have any of the
7 A* h2 J# {' J% o. o7 i9 n1 Sroutes since both three are iBGP peers ( no sync is on R1 config and it is supposed to be% v3 `( l5 I: D
configured the same in all the AS) Dont let this confuse u. iBGP peers do not advertise9 W' z: ?& X1 i4 ]9 D K5 y
each other routes even if syncronization is disabled (the only way to do it is with route5 z( T* G, ]* Z/ h9 t# I$ {: J
reflectors or confederations) So, my conclusion for the routing question is that no, I3 H! P4 p j# _: e5 T- a
routes will reach R3 BGP table.) A. o5 S8 r6 P2 B
Second question is about IPV6 multicast.( Over the lan link between two routers)
) v+ W0 J& \$ N6 H2 h9 {# l' i2 Y# OShows you Mroute entry and gives several options. Right answers are that the multicast: `- W F5 g( R1 U4 h3 d
group in the shared tree is FFF0:0002 (or whatever I dont remember but u should% p! P& g+ v& `( r. f
check it on the (*. XXXX)
# h$ {; s8 n I, {) N3 s- `, fand the second answer is that a route to the RP has been pruned, which can be
; k9 u: r9 f6 e lconfirmed by checking the flags SPC (sparse mode, connected, pruned)4 ]9 E* e* _, c0 h8 g& C
Third question is about QOS. Shows u some markings by packet lenght. Options are
, }( k- X& d5 D8 K4 N% Dthat the firrst class is marking packets which size is between 500 and 1500 bytes, which* W2 H) }9 j/ O, A
is RIGHT. Second option is that second class is marking packets with size of 500 with ip
6 q/ j5 v, J1 p) Rprecendence 3 which is not wrong, because altouhg one of the condifions is lenght=3 @9 e& N9 j/ Q7 `2 Y) R
500, thsa class statement is a MATCH ALL and there is also another condition
( E: j. H. ~0 Q2 g2 t& O# @regardiong DSCP, so paket lengh is not enough for marlking here. This option is WRONG.
, A* ^: {% B+ k& RThird option is that one of the clases is marking packets with size = 200 and this is true.* s+ V: o f4 a0 W( F
Fourth question is about ipv6 eigrp. over the same lan segment, shows loopbak! A6 {4 a) H0 H
interfaces running eigrpV3, everythig n is ok and pings dont work. The right option is. B/ E" S) `% q
that u have to configure eigrp process on each router (doing it in only one is not enough)- [$ S9 X- b) `4 S9 r! d
This is true because Eigrpv6 needs a no shut under the process to work! |
11#
2011-7-15 13:02:57
回复(0)
收起回复
|