The following is a list of criteria that BGP uses in order to select 'Best Path'.
Remember that BGP does not perform load balancing. It selects which path is best and then enters that in to the routing table (assuming there isn't an IGP with a better Administrative Distance that has the same route).
The best path is firstly one that is free of loops (AS PATH does not have the same AS listed twice - if it does it must be a loop and the path is discarded). Next there must be a valid next hop, i.e. the router must be able to reach the next hop device. As such there will be a route in the routing table be it connected, a static, dynamically applied via an IGP. Once these 2 points are met then the following factors influence path selection. First match wins and the process stops.
1) Prefer Highest Weight - Cisco proprietary, locally significant, if another router receives the weight flag it is simply ignored.
2) Prefer Highest Local Preference - influences routes in bound to the AS. Use a route-map to set the Local preference (highest wins) then apply it to the nei statement for the routes you which to influence.
3) Prefer routes originated by the local router - these are the routes listed with 0.0.0.0 in the Origin field in #sh ip bgp summary
4) Prefer Shortest AS_Path - least number of hops wins, just like a distance vector routing protocol
5) Prefer lowest Origin Code - where the route originated from be it an IGP > EGP > ? or incomplete which usually indicates the route was received via reoute redistribution from another protocol such as EIGRP.
6) Prefer the Lowest MED - this is effectively the BGP metric and influences how remote AS's view the path of the routes you advertise.
There are many more factors that affect Best Path selection but the BSCI only requires you to know up to point 6 (and I can't be bothered listing the rest as it goes on some what, I think 'Router with tidiest cables in the back of the rack' is the last factor at number 17!? ;o) ) *
*not really
Wednesday, 14 July 2010
Friday, 21 May 2010
BSCI - EIGRP terms
Advertised Distance = Metric reported FROM a neighbor, alos known as 'Reported Distance'
Feasible Distance = The metric of the best path
Successor = Best/Primary path assed in to the routing table
Feasible Successor = Secondary/backup path retained in the topology table, must meet the 'Feasibility Condition'
Feasibility Condition = To be considered a Feasible Successor, the Advertised Distance MUST be lower than the Feasible Distance (full path) to be considered a Feasible Successor
- If the Advertised Distance is larger than the Feasible Distance then by definition the path must longer and therefore not as good.
- The Feasible Successor means that a new path can be selected with out recalculation therefore the protocol is fast
Feasible Distance = The metric of the best path
Successor = Best/Primary path assed in to the routing table
Feasible Successor = Secondary/backup path retained in the topology table, must meet the 'Feasibility Condition'
Feasibility Condition = To be considered a Feasible Successor, the Advertised Distance MUST be lower than the Feasible Distance (full path) to be considered a Feasible Successor
- If the Advertised Distance is larger than the Feasible Distance then by definition the path must longer and therefore not as good.
- The Feasible Successor means that a new path can be selected with out recalculation therefore the protocol is fast
BSCI - EIGRP configuration review
Configure EIGRP:
R1(config)#router eigrp 1
R1(config-router)#no auto-summary
R1(config-router)#network 172.16.30.0 0.0.0.255
Where 1 = The EIGRP AS number which must match with adjacency neighbors
no auto-summary = prevents route summarisation and advertises subnets
Network = states which network to advertise AND which interface to advertise it out of
Verify EIGRP:
#sh ip eigrp neighbors
-displays status of neighbor relationships
#sh ip eigrp topology
- displays the topology of the eigrp AS
Where - P=Passive, network is available and installed in the routing table
A = Active, network is currently unavailable and EIGRP is working to source a new route
U = Update, network is being updated
Q = Query, applied if an outstanding packet query exists for the network i.e. waiting for an ACK
R = Reply, router is generating a reply for this network or waiting on an ACK to a reply packet
SIA = Stuck-In-Active, signifies EIGRP convergence issue.
*SIA status must be reset when one or more queries to a neighbor do not return before the Active timer expires (usually 3 minutes)
R1(config)#router eigrp 1
R1(config-router)#no auto-summary
R1(config-router)#network 172.16.30.0 0.0.0.255
Where 1 = The EIGRP AS number which must match with adjacency neighbors
no auto-summary = prevents route summarisation and advertises subnets
Network = states which network to advertise AND which interface to advertise it out of
Verify EIGRP:
#sh ip eigrp neighbors
-displays status of neighbor relationships
#sh ip eigrp topology
- displays the topology of the eigrp AS
Where - P=Passive, network is available and installed in the routing table
A = Active, network is currently unavailable and EIGRP is working to source a new route
U = Update, network is being updated
Q = Query, applied if an outstanding packet query exists for the network i.e. waiting for an ACK
R = Reply, router is generating a reply for this network or waiting on an ACK to a reply packet
SIA = Stuck-In-Active, signifies EIGRP convergence issue.
*SIA status must be reset when one or more queries to a neighbor do not return before the Active timer expires (usually 3 minutes)
Review Documents
It's been a while since I last posted. Mainly due to personal commitments at home and work but also through a lack of inspiration.
Out of the blue I had a message for an old colleague at the weekend to say that one of my posts got him out of a fix in his new role and that made me think again.
Since my last post I've been covering Redistribution, Distribution Lists, and Route Maps. I'm also about half way through my BGP studies which on the face of it seem fine. That said I haven't got in to the fine tuning of the protocol yet and that doesn't look pretty.
I'm intending on covering of the main configuration and verification commands of the previous topics mainly to keep them fresh in my mind as I look forward to Multicasting, IPv6 and the case studies on my BSCI Networking Academy course.
Out of the blue I had a message for an old colleague at the weekend to say that one of my posts got him out of a fix in his new role and that made me think again.
Since my last post I've been covering Redistribution, Distribution Lists, and Route Maps. I'm also about half way through my BGP studies which on the face of it seem fine. That said I haven't got in to the fine tuning of the protocol yet and that doesn't look pretty.
I'm intending on covering of the main configuration and verification commands of the previous topics mainly to keep them fresh in my mind as I look forward to Multicasting, IPv6 and the case studies on my BSCI Networking Academy course.
Monday, 19 April 2010
BSCI - IS-IS - Differences between IS-IS and OSPF
- OSPF supports 50 routers per area
- IS-IS can support up to 1000 routers per area
- OSPF includes more features, including route tags, stub/nssa configuration, on demand circuits
- OSPF requires many LSA's to advertise and withdraw routes.
- IS-IS only requires one LSP per IS-IS router in a given area.
- IS-IS doesn't support a Backup designated router (BDR).i.e. you only have the primary Designated Intermediate System (DIS).
- IS-IS repeats the DIS election process every time a new router is added to an area.
- If the new router has the same of higher priority it will become the new DIS.
- IS-IS is designed to be deterministic so that the same set of IS-IS routers always produce the same DIS.
- Every time a DIS changes a new set of LSP's are flooded to the neighbours.
- IS-IS form adjacencies with all routers in the area not just the DIS.
- OSPF uses a scaled metric by default where as IS-IS uses a default metric of 10 for all links.
BSCI - IS-IS - High level operations
Routers using IS-IS issue Hello packets out of all interfaces enabled for IS-IS to discover neighbours and establish adjacencies.
Routers will become IS-IS neighbours if their Hello packets contain certin information for the creation of an adjacency.
Routers then build Link-State Packets (LSP's) based on the adjacencies present and then flodd the LSP's to their neighbours.
All routers then build a Link-State Database (LSDB) using the information in the received LSP.
The router then runs the SPF algorithm, the shortest path to each dsestination is then calcultated and added to the Routing Information Base (RIB or OSI routing table).
Next, the Partial Route Calcultion is performed using the IP information contained within one of the Time, Length, Value (TLV) fields in the LSP. This generates shortest paths for IP traffic and in turn is entered in to the IP routing table.
Routers will become IS-IS neighbours if their Hello packets contain certin information for the creation of an adjacency.
Routers then build Link-State Packets (LSP's) based on the adjacencies present and then flodd the LSP's to their neighbours.
All routers then build a Link-State Database (LSDB) using the information in the received LSP.
The router then runs the SPF algorithm, the shortest path to each dsestination is then calcultated and added to the Routing Information Base (RIB or OSI routing table).
Next, the Partial Route Calcultion is performed using the IP information contained within one of the Time, Length, Value (TLV) fields in the LSP. This generates shortest paths for IP traffic and in turn is entered in to the IP routing table.
Wednesday, 24 March 2010
BSCI - OSPF - Default Route propagation
By default OSPF does not propagate a default route, and so you need to manually tell OSPF to distribute one from your Autonomous System Border Router (ASBR).
Depending on the type of Area you have employed in your network the way you propagate a default to all your routers will differ slightly.
In Normal Areas (OSPF areas all connected to Area 0) you can do the following:
1) On your ASBR apply a default route:
R1(config)#ip route 0.0.0.0 0.0.0.0 10.1.1.1
2) Inject the default route in to OSPF:
R1(config)#router ospf 1
R1(config-router)#default-information originate [always]
- The [always] option allows you to advertise a default route from the ASBR even when one doesn't actually exist. This can potentially result in better stability for your network. For example, if a default route is learned from a different routing protocol such as RIP and this route for what ever reason starts to flap then every time the route changes type 5 LSA's will be sent into the OSPF domain from the ASBR.
- The [always] option helps prevent actions outside of the OSPF domain from affecting the routers/routes within the OSPF domain.
For Stub and Totally Stub areas the situation is different.
On an ABR you configure an area to be a stub. This in turn prevents type 5 LSA's (external route information) from being sent in to the stub area and in return a default summary route is propagated.
In a Totally Stubby area, this goes a step further. By configuring an area as a Totally Stubby area on the ABR you prevent Type 5 LSA's (for external routes) plus Type 4 and Type 3 LSA's (for inter-area summary routes) from being propogated. A default summary route replaces these types of routes.
In both cases, as a default route is automatically generated at the ABR, you do not require the default-information originate command.
Finally you have Not-So-Stubby-Area's (NSSA)
There are 2 ways to advertise a default route. NSSA ABR can generate a default route with or without a default route in its own routing table.
1) On the ABR connecting Area 0 to the NSSA area you force the ABR to generate a default route:
R3(config)#router ospf 1
R3(config-router)#area 8 nssa default-information originate
-With this example the ABR generates Type 7 LSA's with a link state ID of 0.0.0.0 this is then advertised within the NSSA area
- NOTE - NSSA ASBR can generate a default only when it has a default route in its routing table
- The default route via the ASBR must be known through non-OSPF protocol
2) You can also use the 'no-summary' option when defining your NSSA area and create 'NSSA Totally Stub area':
R3(config)#router ospf 1
R3(config-router)#area 8 nssa no-summary
-In this example you are replacing the Type 3,4,(inter-area summary routes) and Type 5 LSA's(external summary routes) with a default summary route.This is just as you do for a Totally Stubby area.
Depending on the type of Area you have employed in your network the way you propagate a default to all your routers will differ slightly.
In Normal Areas (OSPF areas all connected to Area 0) you can do the following:
1) On your ASBR apply a default route:
R1(config)#ip route 0.0.0.0 0.0.0.0 10.1.1.1
2) Inject the default route in to OSPF:
R1(config)#router ospf 1
R1(config-router)#default-information originate [always]
- The [always] option allows you to advertise a default route from the ASBR even when one doesn't actually exist. This can potentially result in better stability for your network. For example, if a default route is learned from a different routing protocol such as RIP and this route for what ever reason starts to flap then every time the route changes type 5 LSA's will be sent into the OSPF domain from the ASBR.
- The [always] option helps prevent actions outside of the OSPF domain from affecting the routers/routes within the OSPF domain.
For Stub and Totally Stub areas the situation is different.
On an ABR you configure an area to be a stub. This in turn prevents type 5 LSA's (external route information) from being sent in to the stub area and in return a default summary route is propagated.
In a Totally Stubby area, this goes a step further. By configuring an area as a Totally Stubby area on the ABR you prevent Type 5 LSA's (for external routes) plus Type 4 and Type 3 LSA's (for inter-area summary routes) from being propogated. A default summary route replaces these types of routes.
In both cases, as a default route is automatically generated at the ABR, you do not require the default-information originate command.
Finally you have Not-So-Stubby-Area's (NSSA)
There are 2 ways to advertise a default route. NSSA ABR can generate a default route with or without a default route in its own routing table.
1) On the ABR connecting Area 0 to the NSSA area you force the ABR to generate a default route:
R3(config)#router ospf 1
R3(config-router)#area 8 nssa default-information originate
-With this example the ABR generates Type 7 LSA's with a link state ID of 0.0.0.0 this is then advertised within the NSSA area
- NOTE - NSSA ASBR can generate a default only when it has a default route in its routing table
- The default route via the ASBR must be known through non-OSPF protocol
2) You can also use the 'no-summary' option when defining your NSSA area and create 'NSSA Totally Stub area':
R3(config)#router ospf 1
R3(config-router)#area 8 nssa no-summary
-In this example you are replacing the Type 3,4,(inter-area summary routes) and Type 5 LSA's(external summary routes) with a default summary route.This is just as you do for a Totally Stubby area.
Tuesday, 9 March 2010
EIGRP Summarisation and NULL0
EIGRP Summarisation allows you to stream line the routing table making it more efficient. Fewer Routes listed results in less EIGRP updates being sent out and there fore less resources are consumed (Bandwidth, CPU ultilisation, load etc).
By default EIGRP performs auto-summarisation, that is to say that EIGRP will automatically summarise at a major class boundary during redistribution from EIGRP into a classful routing protocol (eg RIP).
EIGRP will also summarise at the major classful boundary when a route is advertised out of an interface that is on a different major class boundary.
In order to prevent a routing loop when summarisation is in effect (whether manual or automatic) a summary is automatically assigned to the NULL0 interface to prevent routing loops. If the router with a summary route received a packet for an unknown subnet that is part of a summarised range then the longest match ends up being the summary route itself (not a subnet of it) and so this is forwarded to the NULL0 interface and is dropped.
The idea is that the router is then prevented from forwarding the packet on to a default route and potentially creating a loop.
Benefits of summarising in this way include a smaller routing table leading to faster look ups, more specific routes will be hidden so if that specific route goes down the whole network does not need to recalculate the DUAL alogrithm, routing updates will be smaller and so limit the number of EGIRP Queries.
The down side of auto summarisation is that it won't factor in discontiguous networks. As a result you could have networks behind the same 172.16.0.0/16 network advertised out of 2 different interfaces to 2 different networks resulting in 50% of traffic arriving in the wrong place.
To address this issue you can disable automatic summarisation and use manual summarisation.
Disabling automatic summarisation is as simple as this:
R1(config)#router eigrp 1
R1(config-router)#no auto-summary
Now the router will not perform summarisation and all available subnets will be advertised.
Manual summarisation is configured on a per-interface basis, when a summary route is applied a NULL0 summary route entry is immediately created in the routing table to help prevent loops.
To configure a manual summary route you do the following:
R1(config)#router eigrp 1
R1(config-router)#no auto-summary
R1(config)#int s0/0/0
R1(config-if)#ip summary-address eigrp [AS] [IP] [Subnet Mask]
e.g) R1(config)#int s0/0/0
R1(config-if)#ip summary-address eigrp 1 172.16.0.0 255.255.224.0
With a manual summary route, the summary route is advertised only if a more specific entry of the summary is present in the routing table. Otherwise it won't appear in the routing table.
By default EIGRP performs auto-summarisation, that is to say that EIGRP will automatically summarise at a major class boundary during redistribution from EIGRP into a classful routing protocol (eg RIP).
EIGRP will also summarise at the major classful boundary when a route is advertised out of an interface that is on a different major class boundary.
In order to prevent a routing loop when summarisation is in effect (whether manual or automatic) a summary is automatically assigned to the NULL0 interface to prevent routing loops. If the router with a summary route received a packet for an unknown subnet that is part of a summarised range then the longest match ends up being the summary route itself (not a subnet of it) and so this is forwarded to the NULL0 interface and is dropped.
The idea is that the router is then prevented from forwarding the packet on to a default route and potentially creating a loop.
Benefits of summarising in this way include a smaller routing table leading to faster look ups, more specific routes will be hidden so if that specific route goes down the whole network does not need to recalculate the DUAL alogrithm, routing updates will be smaller and so limit the number of EGIRP Queries.
The down side of auto summarisation is that it won't factor in discontiguous networks. As a result you could have networks behind the same 172.16.0.0/16 network advertised out of 2 different interfaces to 2 different networks resulting in 50% of traffic arriving in the wrong place.
To address this issue you can disable automatic summarisation and use manual summarisation.
Disabling automatic summarisation is as simple as this:
R1(config)#router eigrp 1
R1(config-router)#no auto-summary
Now the router will not perform summarisation and all available subnets will be advertised.
Manual summarisation is configured on a per-interface basis, when a summary route is applied a NULL0 summary route entry is immediately created in the routing table to help prevent loops.
To configure a manual summary route you do the following:
R1(config)#router eigrp 1
R1(config-router)#no auto-summary
R1(config)#int s0/0/0
R1(config-if)#ip summary-address eigrp [AS] [IP] [Subnet Mask]
e.g) R1(config)#int s0/0/0
R1(config-if)#ip summary-address eigrp 1 172.16.0.0 255.255.224.0
With a manual summary route, the summary route is advertised only if a more specific entry of the summary is present in the routing table. Otherwise it won't appear in the routing table.
CCNP ROUTE Cert Kit Giveaway
Rofi Neron over at ITDualism.wordpress.com has a Cisco CCNP Route Cert Kit to giveaway.
To enter check out his competition here: http://itdualism.wordpress.com/2010/03/03/giveaway-route-cert-kit/
Well worth the time and an excellent opportunity to bag some Cisco Press material.
Good Luck!!
To enter check out his competition here: http://itdualism.wordpress.com/2010/03/03/giveaway-route-cert-kit/
Well worth the time and an excellent opportunity to bag some Cisco Press material.
Good Luck!!
Wednesday, 3 March 2010
CCNP and Networking Academy students
Since Cisco announced their plans to move to the new 3 exam path back in January there's been much discussion at my local Networking Academy about what that actually means for those who have already passed some of the exams but won't finish the Academy programme until after July31st.
Cisco have posted the following information;
CCNP Exam Combinations:
BSCI+BCMSN+ISCW+ONT = Last day is July 31st 2011
COMP+ISCW+ONT = Last day is July 31st 2011
BSCI+SWITCH+ISCW+ONT = Last day is July 31st 2011
ROUTE+BCMSN+ISCW+ONT = Last day is July 31st 2011
ROUTE+SWITCH+ISCW+ONT = Last day is July 31st 2011
BSCI+BCMSN+TSHOOT = Ongoing
COMP+TSHOOT = Ongoing
BSCI+TSHOOT+SWITCH = Ongoing
ROUTE+BCMSN+TSHOOT = Ongoing
ROUTE+SWITCH+TSHOOT = Ongoing
Cisco have posted the following information;
on the last page it says
Q. Until when will Networking Academy students be able to take the current CCNP certification exams?
A. The four current CCNP certification exams will be available to the general public through July 31, 2010 and to Networking Academy students using a special voucher through July 31, 2011. More information about how to obtain a special, non-discount voucher to enable students to continue to take the retiring exams will be communicated as soon as it is available.
CCNP Exam Combinations:
BSCI+BCMSN+ISCW+ONT = Last day is July 31st 2011
COMP+ISCW+ONT = Last day is July 31st 2011
BSCI+SWITCH+ISCW+ONT = Last day is July 31st 2011
ROUTE+BCMSN+ISCW+ONT = Last day is July 31st 2011
ROUTE+SWITCH+ISCW+ONT = Last day is July 31st 2011
BSCI+BCMSN+TSHOOT = Ongoing
COMP+TSHOOT = Ongoing
BSCI+TSHOOT+SWITCH = Ongoing
ROUTE+BCMSN+TSHOOT = Ongoing
ROUTE+SWITCH+TSHOOT = Ongoing
This should be a weight off anyone's mind who is doing their CCNP through a Networking Academy (including me!)
Subscribe to:
Posts (Atom)
Posts
