2.3.0 Introduction to ISIS Basic Commands and Query Commands, ISIS Features, ISIS and OSPF
This article uses the characteristics of ISIS combined with ISIS configuration examples to describe it, so the space will be very large. It is recommended to watch it on the computer in combination with the catalog to avoid affecting the look and feel.
Table of contents
route penetration
By default, the level-1-2 router will perform the following route penetration to ensure the routing communication between different areas of ISIS:
(1) Level-1 routing will penetrate into level-2 by default.
(2) By default, level-2 routes will not infiltrate into level-1. When the area needs to learn detailed routes of level-2, it can be achieved by importing routes.
1. The reason why L2 does not penetrate into L1
Level-1 (L1) area devices are mainly routers with relatively low device performance. As a special area in ISIS, L1 is the same as a special area in OSPF, and there is no need for too many detailed routes.
Therefore, routers in special areas only need to access other areas through the default routes generated by level-1-2 (L1/2) routers.
2. The advantages and disadvantages of accessing other areas through the default route
Advantages :
(1) It can make L1 and L1/2 converge faster, which can enhance the stability of the network to a certain extent.
(2) Low performance requirements for L1 routers.
Disadvantages :
(1) It will lead to the problem of suboptimal path during access.
(2) As a result, devices in L1 cannot control the selection of L1 routes by modifying the overhead of other areas.
Configure route leaking instance
The ultimate goal : observe 路由表信息the AND of R1 and R4 LSDB信息, and understand the meaning of the above route penetration.
1. Configure the IP address of the device
hint:
1. Some commands are displayed in abbreviated form, and related isis commands are written in full.
2. The interface address is: 10.1. [相邻路由器序号]. [路由器序号]For example, the interface connecting AR1 to AR2: 10.1.12.1.
AR1
<Huawei>sys
Enter system view, return user view with Ctrl+Z.
[Huawei]sys AR1
[AR1]un in en
Info: Information center is disabled.
[AR1]int g0/0/0
[AR1-GigabitEthernet0/0/0]ip add 10.1.12.1 24
[AR1-GigabitEthernet0/0/0]int g0/0/1
[AR1-GigabitEthernet0/0/1]ip add 10.1.13.1 24
[AR1-GigabitEthernet0/0/1]q
AR2
<Huawei>sy
Enter system view, return user view with Ctrl+Z.
[Huawei]sys AR2
[AR2]un in en
Info: Information center is disabled.
[AR2]int g0/0/0
[AR2-GigabitEthernet0/0/0]ip add 10.1.12.2 24
[AR2-GigabitEthernet0/0/0]int g0/0/1
[AR2-GigabitEthernet0/0/1]ip add 10.1.24.2 24
[AR2-GigabitEthernet0/0/1]q
AR3
<Huawei>sys
Enter system view, return user view with Ctrl+Z.
[Huawei]sys AR3
[AR3]un in en
Info: Information center is disabled.
[AR3]int g0/0/0
[AR3-GigabitEthernet0/0/0]ip add 10.1.13.3 24
[AR3-GigabitEthernet0/0/0]int g0/0/1
[AR3-GigabitEthernet0/0/1]ip ad 10.1.35.3 24
[AR3-GigabitEthernet0/0/1]q
AR4
<Huawei>sys
Enter system view, return user view with Ctrl+Z.
[Huawei]sys AR4
[AR4]un in en
Info: Information center is disabled.
[AR4]int g0/0/0
[AR4-GigabitEthernet0/0/0]ip ad 10.1.34.4 24
[AR4-GigabitEthernet0/0/0]int g0/0/1
[AR4-GigabitEthernet0/0/1]ip add 10.1.45.4 24
[AR4-GigabitEthernet0/0/1]q
AR5
<Huawei>sys
Enter system view, return user view with Ctrl+Z.
[Huawei]sys AR5
[AR5]un in en
Info: Information center is disabled.
[AR5]int g0/0/0
[AR5-GigabitEthernet0/0/0]ip add 10.1.35.5 24
[AR5-GigabitEthernet0/0/0]int g0/0/1
[AR5-GigabitEthernet0/0/1]ip add 10.1.45.5 24
[AR5-GigabitEthernet0/0/1]q
2. Configure ISIS
AR1, after configuring the ISIS command for the first time, a command explanation will be attached.
Regarding the NET address : 49.0001.0000.0000.0001.00, 49.0001 is the AreaID area number , and the following 0000.0000.00001 means the SystemID , which can be set according to the serial number of the router if there is no regulation.
# 创建一个ISIS实例
[AR1]isis 1
# 每个ISIS必须设置一个NET地址
[AR1-isis-1]network-entity 49.0001.0000.0000.0001.00
# 设置ISIS路由器级别
[AR1-isis-1]is-level level-1
# 接口启用ISIS功能
[AR1-isis-1]int g0/0/0
[AR1-GigabitEthernet0/0/0]isis enable
[AR1-GigabitEthernet0/0/0]int g0/0/1
[AR1-GigabitEthernet0/0/1]isis enable
AR2, since ISIS is level-1-2 by default, there is no need to specify it manually.
[AR2]isis 1
[AR2-isis-1]network-entity 49.0001.0000.0000.0002.00
[AR2-isis-1]int g0/0/0
[AR2-GigabitEthernet0/0/0]isis enable
[AR2-GigabitEthernet0/0/0]int g0/0/1
[AR2-GigabitEthernet0/0/1]isis enable
[AR2-GigabitEthernet0/0/1]q
AR3
[AR3]isis 1
[AR3-isis-1]network-entity 49.0001.0000.0000.0003.00
[AR3-isis-1]int g0/0/0
[AR3-GigabitEthernet0/0/0]isis enable
[AR3-GigabitEthernet0/0/0]int g0/0/1
[AR3-GigabitEthernet0/0/1]isis enable
[AR3-GigabitEthernet0/0/1]q
AR4
[AR4]isis 1
[AR4-isis-1]network-entity 49.0002.0000.0000.0004.00
[AR4-isis-1]is-level level-2
[AR4-isis-1]int g0/0/0
[AR4-GigabitEthernet0/0/0]isis enable
[AR4-GigabitEthernet0/0/0]int g0/0/1
[AR4-GigabitEthernet0/0/1]isis enable
[AR4-GigabitEthernet0/0/1]q
AR5
[AR5]isis 1
[AR5-isis-1]network-entity 49.0002.0000.0000.0005.00
[AR5-isis-1]is-level level-2
[AR5-isis-1]int g0/0/0
[AR5-GigabitEthernet0/0/0]isis enable
[AR5-GigabitEthernet0/0/0]int g0/0/1
[AR5-GigabitEthernet0/0/1]isis enable
[AR5-GigabitEthernet0/0/1]q
3. Check ISIS
(1) Whether an adjacency relationship has been established
You can see some brief information, such as SystemID, peer level type, and interface priority (for DIS election).
<AR1>display isis peer
Peer information for ISIS(1)
System Id Interface Circuit Id State HoldTime Type PRI
-------------------------------------------------------------------------------
0000.0000.0002 GE0/0/0 0000.0000.0002.01 Up 8s L1 64
0000.0000.0003 GE0/0/1 0000.0000.0003.01 Up 7s L1 64
Total Peer(s): 2
Check who is the DIS of the specified interface. By checking the DIS column, the one that is Yes is the DIS.
<AR1>display isis interface
Interface information for ISIS(1)
---------------------------------
Interface Id IPV4.State IPV6.State MTU Type DIS
GE0/0/0 001 Up Down 1497 L1/L2 No/No
GE0/0/1 002 Up Down 1497 L1/L2 No/No
----------------
<AR2>display isis interface
Interface information for ISIS(1)
---------------------------------
Interface Id IPV4.State IPV6.State MTU Type DIS
GE0/0/0 001 Up Down 1497 L1/L2 Yes/No
GE0/0/1 002 Up Down 1497 L1/L2 No/No
(2) Whether the routing table has learned the relevant routes
AR1
Information viewed from the table:
1. Routing entries : Among the routes learned by AR1, there is no 10.1.45.0/24 network segment between L2, but there is a default route .
Why is there a default route?
This has to be recalled to route penetration:
L2不会渗透入L1路由表中, in order to ensure that L1 can communicate with L2, the optimal DIS in the L1 area generates a default route to guide traffic forwarding.
2. Routing overhead :
In ISIS, all link costs are 10 by default.
From the results in the table, it can be seen that the L1 device only knows the cost of all the routes in the L1 area, but does not know the cost of the routes in the L2 area .
For this reason, there may be a problem of suboptimal paths .
(1) Why is there a suboptimal path?
As shown below:
Assuming that AR1 needs to access the network of AR4, since there is only the default route, the cost of going through AR2 and going through AR3 is both 10.
If you take AR3, the actual cost is 30, while the cost of taking AR2 is only 20, which is caused by the optimal path this time.
(2) How to solve the suboptimal path?
If the route of L2 can be introduced into L1 so that L1 can learn the detailed route of L2, this problem can be solved.
<AR1>display isis route
Route information for ISIS(1)
-----------------------------
ISIS(1) Level-1 Forwarding Table
--------------------------------
IPV4 Destination IntCost ExtCost ExitInterface NextHop Flags
-------------------------------------------------------------------------------
0.0.0.0/0 10 NULL GE0/0/1 10.1.13.3 A/-/-/-
GE0/0/0 10.1.12.2
10.1.24.0/24 20 NULL GE0/0/0 10.1.12.2 A/-/-/-
10.1.13.0/24 10 NULL GE0/0/1 Direct D/-/L/-
10.1.12.0/24 10 NULL GE0/0/0 Direct D/-/L/-
10.1.35.0/24 20 NULL GE0/0/1 10.1.13.3 A/-/-/-
Flags: D-Direct, A-Added to URT, L-Advertised in LSPs, S-IGP Shortcut,
U-Up/Down Bit Set
AR4
Information viewed from the table:
1. Routing information : L2 devices can learn all the routing information in the ISIS network. Because the routes are leaking L1路由可渗透入L2路由表中, they can learn all the routes.
2. Routing overhead : Since it has all the routes in the ISIS network, it can select the optimal routing path forwarding based on the routing overhead.
<AR4>display isis route
Route information for ISIS(1)
-----------------------------
ISIS(1) Level-2 Forwarding Table
--------------------------------
IPV4 Destination IntCost ExtCost ExitInterface NextHop Flags
-------------------------------------------------------------------------------
10.1.24.0/24 10 NULL GE0/0/0 Direct D/-/L/-
10.1.13.0/24 30 NULL GE0/0/1 10.1.45.5 A/-/-/-
GE0/0/0 10.1.24.2
10.1.12.0/24 20 NULL GE0/0/0 10.1.24.2 A/-/-/-
10.1.45.0/24 10 NULL GE0/0/1 Direct D/-/L/-
10.1.35.0/24 20 NULL GE0/0/1 10.1.45.5 A/-/-/-
Flags: D-Direct, A-Added to URT, L-Advertised in LSPs, S-IGP Shortcut,
U-Up/Down Bit Set
(3) Link state database of ISIS
AR1
1. The meaning of the LSP ID in the table : take 0000.0000.00001 . 00 - 00 * as an example.
Pseudo-node ID : When this parameter is not zero, it means that the LSP is generated by a pseudo-node .
Fragment number : The fragment number is used to distinguish different LSP fragments, because multiple LSP fragments will be generated when the packet carries too much information.
2. In the table, it can be clearly seen that the SystemIDs of R4 and R5 do not exist, because they do not infiltrate routes into L1.
<AR1>display isis lsdb
Database information for ISIS(1)
--------------------------------
Level-1 Link State Database
LSPID Seq Num Checksum Holdtime Length ATT/P/OL
-------------------------------------------------------------------------------
0000.0000.0001.00-00* 0x00000008 0xf689 1115 97 0/0/0
0000.0000.0002.00-00 0x00000007 0x9a5e 689 86 1/0/0
0000.0000.0002.01-00 0x00000003 0xa0e7 559 55 0/0/0
0000.0000.0003.00-00 0x00000009 0xf9e0 706 86 1/0/0
0000.0000.0003.01-00 0x00000003 0xa9dc 706 55 0/0/0
Total LSP(s): 5
*(In TLV)-Leaking Route, *(By LSPID)-Self LSP, +-Self LSP(Extended),
ATT-Attached, P-Partition, OL-Overload
AR4
<AR4>display isis lsdb
Database information for ISIS(1)
--------------------------------
Level-2 Link State Database
LSPID Seq Num Checksum Holdtime Length ATT/P/OL
-------------------------------------------------------------------------------
0000.0000.0002.00-00 0x0000000a 0xb1b6 634 110 0/0/0
0000.0000.0003.00-00 0x0000000c 0x4f08 650 110 0/0/0
0000.0000.0003.02-00 0x00000003 0x136e 650 55 0/0/0
0000.0000.0004.00-00* 0x0000000a 0x3bda 714 97 0/0/0
0000.0000.0004.01-00* 0x00000002 0xd0b3 714 55 0/0/0
0000.0000.0004.02-00* 0x00000003 0x1c63 714 55 0/0/0
0000.0000.0005.00-00 0x0000000a 0xf408 709 97 0/0/0
Total LSP(s): 7
*(In TLV)-Leaking Route, *(By LSPID)-Self LSP, +-Self LSP(Extended),
ATT-Attached, P-Partition, OL-Overload
Pseudo-node and non-pseudo-node LSP internal information
Take AR1 as an example:
(1) 0000.0000.0002.01-00Internal information of the pseudo-node:
<AR1>display isis lsdb verbose 0000.0000.0002.01-00
(2) 0000.0000.0002.00-00Internal information of non-pseudo-nodes:
<AR1>display isis lsdb verbose 0000.0000.0002.00-00
4. Supplementary experiments
(1) Solve the suboptimal path problem mentioned in L1
By introducing L2 routes into L1 on L1/2 devices, L1 can learn detailed routes of L2 to solve the problem of suboptimal routing.
AR4, create a loopback port to publish to ISIS
[AR4]int lo 0
[AR4-LoopBack0]ip add 192.168.4.254 24
[AR4-LoopBack0]isis enable
[AR4-LoopBack0]q
AR2
[AR2]isis 1
[AR2-isis-1]import-route isis level-2 into level-1
AR3
[AR3]isis 1
[AR3-isis-1]import-route isis level-2 into level-1
(2) View the routing table and LSDB of AR1 again
At this time, AR1 can learn the routing information between AR45 and the route advertised by AR4, and can correctly select the route.
With detailed routes, it is also possible to control route selection by modifying costs in other areas.
But the default route still exists, because the L2 import to L1 can also use the list to filter the specified route.
import-route isis level-2 into level-1 filter-policy ACL/IP-Prefix/Router-policy
<AR1>display isis route
Route information for ISIS(1)
-----------------------------
ISIS(1) Level-1 Forwarding Table
--------------------------------
IPV4 Destination IntCost ExtCost ExitInterface NextHop Flags
-------------------------------------------------------------------------------
0.0.0.0/0 10 NULL GE0/0/1 10.1.13.3 A/-/-/-
GE0/0/0 10.1.12.2
10.1.24.0/24 20 NULL GE0/0/0 10.1.12.2 A/-/-/-
10.1.13.0/24 10 NULL GE0/0/1 Direct D/-/L/-
10.1.12.0/24 10 NULL GE0/0/0 Direct D/-/L/-
10.1.35.0/24 20 NULL GE0/0/1 10.1.13.3 A/-/-/-
192.168.4.0/24 20 NULL GE0/0/0 10.1.12.2 A/-/-/U
10.1.45.0/24 30 NULL GE0/0/0 10.1.12.2 A/-/-/U
GE0/0/1 10.1.13.3
Flags: D-Direct, A-Added to URT, L-Advertised in LSPs, S-IGP Shortcut,
U-Up/Down Bit Set
AR1's LSDB database information can vary from year to year without significant quantity changes.
<AR1>display isis lsdb
Database information for ISIS(1)
--------------------------------
Level-1 Link State Database
LSPID Seq Num Checksum Holdtime Length ATT/P/OL
-------------------------------------------------------------------------------
0000.0000.0001.00-00* 0x0000000a 0xf28b 753 97 0/0/0
0000.0000.0002.00-00 0x0000000c 0xe82b 937 110 1/0/0
0000.0000.0002.01-00 0x00000006 0x9aea 912 55 0/0/0
0000.0000.0003.00-00 0x0000000e 0x25c6 1128 110 1/0/0
0000.0000.0003.01-00 0x00000006 0xa3df 1128 55 0/0/0
Total LSP(s): 5
*(In TLV)-Leaking Route, *(By LSPID)-Self LSP, +-Self LSP(Extended),
ATT-Attached, P-Partition, OL-Overload
Why can't there be routing information without the LSP information of AR4 and AR5?
Because the routing information is learned through osmosis, the routing information of AR45 is in the LSP of AR23:
5. Expansion
1. How is the default route in L1 generated?
In fact, the default route in L1 is generated by sending the LSP with ATT=1 through the L1/2 device.
So what is the LSP condition for L1/2 to generate ATT=1 ?
When the L1/2 device can reach multiple areas through the L2 database, an LSP with ATT=1 will be generated and sent to the L1 area.
In addition to automatically generating LSPs with ATT=1, the generation of ATT can also be controlled by commands on the L1/2 device :
1. In the ISIS mode of L1/2 routing: attached-bit advertise always, ATT is always set to 1 in the LSP that controls L1/2 to pass to L1.
2. In the ISIS mode of L1/2 routing: attached-bit advertise never, ATT is always set to 0 in the LSP that controls L1/2 to pass to L1.
3. In the ISIS mode of L1 routing: attached-bit avoid-learning, separately control the machine not to generate a default route after receiving an LSP with ATT=1.
Based on the example of route leaking, it is configured on AR1 not to generate a default route.
After the operation, although the entry can still be seen in the isis routing table, it has not been introduced into the routing table.
[AR1]isis
[AR1-isis-1]attached-bit avoid-learning
[AR1-isis-1]q
[AR1]display isis route
IPV4 Destination IntCost ExtCost ExitInterface NextHop Flags
-------------------------------------------------------------------------------
0.0.0.0/0 10 NULL
[AR1]display ip routing-table protocol isis
Destination/Mask Proto Pre Cost Flags NextHop Interface
10.1.24.0/24 ISIS-L1 15 20 D 10.1.12.2 GigabitEthernet0/0/0
10.1.35.0/24 ISIS-L1 15 20 D 10.1.13.3 GigabitEthernet0/0/1
10.1.45.0/24 ISIS-L1 15 30 D 10.1.13.3 GigabitEthernet0/0/1
ISIS-L1 15 30 D 10.1.12.2 GigabitEthernet0/0/0
192.168.4.0/24 ISIS-L1 15 20 D 10.1.12.2 GigabitEthernet0/0/0
2. Link cost of ISIS
The cost of this link is partly based on the original article of CSDN blogger "Forever a Boy"
Introduction to Overhead
》When the IS-IS protocol first came out, the Internet network structure was still very simple, so the early version of IS-IS only used 6 bits to describe the link cost, and the value range of the link cost was 1-63. The cost range of a route is only 10 bits, and the value range is 0-1023.
"Later, as the scale of the computer network continued to expand and the network structure became more complex, the original overhead calculation method could no longer meet the networking needs of large-scale networks.
》Because the range of the cost value is too small, when the routing cost exceeds 1023, the cost value will be 1023 by default, so there will be a suboptimal path problem.
》So ISO developed a set of IS-IS overhead model. The original overhead is called the "narrow" type (also called narrow), and the new overhead mode is called the "wide" type (also called wide).
》In the "wide" mode, the interface overhead is 24bit, and the value range is up to more than 16 million; the routing overhead is 32bit, and the value range is up to more than 3.2 billion, which can fully meet the current complex network topology networking needs.
Overhead and TLVs
》IS-IS routers judge the overhead style adopted by neighboring IS-IS routers by identifying the difference in TLV .
》When the IS-IS overhead type is incompatible with itself, it does not affect the establishment of IS-IS neighbors between each other, and it will normally receive the neighbor's LSP message and store it in its own LSDB, but it does not calculate the neighbor's routing information. Routes are learned but not imported into the routing table .
The narrow style uses the following type of TLV ( the default overhead type ):
1. IP Internal Reachability TLV, TLV No. 128 , used to carry IS-IS routing information in the routing domain.
2. IP External Reachability TLV, TLV No. 130 , is used to carry routing information outside the routing domain.
3. Neighbors TLV, No. 2 TLV , used to carry neighbor information.
The wide style uses the following types of TLVs :
1. Extended IP Reachability TLV, TLV No. 135 , is used to replace the original IP Reachability TLV and expand the range of routing overhead values.
2. IS Extended Neighbors TLV, TLV No. 22 , used to carry neighbor information.
By entering the command in ISIS mode: cost-style wideto modify the cost type of the router, it is recommended that all routers be modified uniformly.
overhead type
"In addition to the common Narrow and Wide types, there are three types: wide-compatible, narrow-compatible and compatible.
》The latter three styles of overhead values are introduced to a large extent to solve the compatibility problem of new and old versions of network equipment.
| overhead style | send overhead style | Received and calculated packet type |
|---|---|---|
| narrow | narrow | narrow |
| wide | wide | wide |
| narrow-compatible | narrow | narrow&wide |
| wide-compatible | wide | narrow&wide |
| compatible | narrow&wide | narrow&wide |
3. ISIS route import
Usually, when importing routes, it import-route 引入协议can be implemented directly, but in ISIS, there are certain requirements. If the operation is not done properly, the import will fail.
The route type imported by default import-route 引入协议is level-2. If you import the route by default on the level-1 device, you will find that the import cannot be successful.
(1) Regular route import
Level-1 introduces instances of RIP by default :
Requirement : Import the route of RIP protocol into the ISIS network. The relevant rip command configuration omits the configuration, mainly see the introduction part.
[AR1]isis 1
[AR1-isis-1]import-route rip 1
# 查看ISIS路由表，并不能看到引入的RIP路由
[AR1-isis-1]display isis route
Improvement : When ISIS imports routes, manually specify the imported route type as level-1.
[AR1-isis-1]import-route rip level-1
# 此时就能在ISIS路由表中看到引入的路由了
[AR1-isis-1]display isis route
ISIS(1) Level-1 Redistribute Table
----------------------------------
Type IPV4 Destination IntCost ExtCost Tag
-------------------------------------------------------------------------------
D 10.1.114.0/24 0 NULL
R 192.168.14.0/24 0 NULL
(2) Route penetration + route import
What happens if AR2 (L1/2) "imports" the loopback port into ISIS after L2 penetration into L1 configuration ?
1. The default import type is level-2. At this time, you would think that AR2 will directly import the loopback port into the L1 database. In fact, it is not. Considering the reason of the loop, AR2 will only send the imported loopback interface to the L2 database and pass it on to the routers in the L2 area.
2. If AR3 fails, AR1 will not be able to learn the loopback port introduced by AR2.
3. Of course, if you modify the type to level-1 when introducing the loopback port, it can be directly passed to the L1 area.
Conclusion : After L1/2 is configured with L2 penetration into L1, the imported level-2 routes cannot be directly imported into the level-1 area.
route overload flag
About OL (LSDB Overload) in ISIS LSP: the overload flag.
OL function : When OL=1, it means that the performance of the device is overloaded, and the data packet is allowed to go another way. During this period, it can be replaced with a new device or other operations.
Phenomenon when OL=1 : L1, L2, and L1/2 routers can all be configured to be overloaded. If the L1/2 router leaks routes, it will no longer leak routes after the router is overloaded. If the L1 and L2 settings are overloaded, the generated routes will not be learned and used.
About the OL overload instance
experimental topology
configuration command
[AR3]isis 1
[AR3-isis-1]set-overload
Warning: The IS-IS process overload state will be set. Continue?[Y/N]y
Capture packets on the g0/0/1 interface of AR1 to check whether OL is set
Observe the next hop information in the routing table on AR1
Although most of the routing entries have switched their next hops, there is still one route that has not changed. What is the reason?
The answer is: For the route sent by the overloaded device, this device will not be used for forwarding, and only the direct route on the device will be calculated.
Therefore, it can learn the directly connected routing information published by AR3 in ISIS (directly connected network segment between AR3 and AR5).
<AR1>dis ip routing-table protocol isis
Route Flags: R - relay, D - download to fib
------------------------------------------------------------------------------
Public routing table : ISIS
Destinations : 5 Routes : 5
ISIS routing table status : <Active>
Destinations : 5 Routes : 5
Destination/Mask Proto Pre Cost Flags NextHop Interface
0.0.0.0/0 ISIS-L1 15 10 D 10.1.12.2 GigabitEthernet0/0/0
10.1.24.0/24 ISIS-L1 15 20 D 10.1.12.2 GigabitEthernet0/0/0
10.1.35.0/24 ISIS-L1 15 20 D 10.1.13.3 GigabitEthernet0/0/1
10.1.45.0/24 ISIS-L1 15 30 D 10.1.12.2 GigabitEthernet0/0/0
192.168.4.0/24 ISIS-L1 15 20 D 10.1.12.2 GigabitEthernet0/0/0
ISIS routing internal priority
By default, when level-1-2 learns two routes with the same destination but different levels, the route at level-1 is preferred.
The default route priority of ISIS is 15, which is called external priority, while the hidden route priority of level-1 is 15, and the hidden priority of level-2 is 18, which is called internal priority.
| Types of Routing Protocols | Internal Precedence for Routing Protocols | Routing Protocol External Precedence |
|---|---|---|
| IS-IS Level-1 | 15 | 15 |
| IS-IS Level-2 | 18 | 15 |
As shown in the figure below, if L1 is better than L2, then when L2 is introduced into L1, for the L1/2 device at the other end, if L1 is better than L2, wouldn’t it result in a suboptimal path?
For this reason, the Distribution field (DU bit for short) in the ISIS TLV plays a role. When the DU bit in L1 is set to 1, the priority order is: L1>L2>L1 with DU set.
Conditions for setting the DU bit : When a non-directly connected level-2 route is imported into the level-1 area, the DU bit of the imported level-2 route is set to 1.
ISIS Hostname Mapping
When viewing the LSDB of ISIS, you will find that only the SystemID is displayed on the LSPID, which is not very intuitive to know which device it is.
To do this, enter the command: in ISIS modeis-name 新名称 to set the alias of the ISIS device. When you check the LSDB again, you will find that the LSPID has changed to the alias you set.
Example of host name mapping modification : Modify all ISIS device aliases to device names.
[AR1]isis 1
[AR1-isis-1]is-name AR1
[AR2]isis 1
[AR2-isis-1]is-name AR2
[AR3]isis 1
[AR3-isis-1]is-name AR2
[AR4]isis 1
[AR4-isis-1]is-name AR4
[AR5]isis 1
[AR5-isis-1]is-name AR5
Go back to AR1 to watch LSDB information
<AR1>display isis lsdb
Database information for ISIS(1)
--------------------------------
Level-1 Link State Database
LSPID Seq Num Checksum Holdtime Length ATT/P/OL
-------------------------------------------------------------------------------
AR1.00-00* 0x0000002a 0x3a96 1128 93 0/0/0
AR1.00-01* 0x00000009 0x65d1 355 45 0/0/0
AR2.00-00 0x0000002e 0xe9d2 1139 98 1/0/0
AR2.00-01 0x00000003 0x42a1 664 37 0/0/0
AR2.01-00 0x00000016 0x1b4a 664 54 0/0/0
AR2.00-00 0x00000031 0xeabb 1155 82 0/0/1
AR2.01-00 0x00000014 0x2342 454 54 0/0/0
ISIS Tag
ISIS marks the Tag, punches in different Tags for different routes, and realizes the management of routes with different Tags.
1. Set Tag for the direct connection port: isis tag-value 100.
2. Set Tag for the imported route: 通过Route-policy匹配路由并设置Tag.
Important : The default Narrow overhead does not carry Tag tags, and needs to be switched to Wide overhead.
It will be frequently used when introducing the two-point two-way of ISIS in the future.
Some differences between ISIS and OSPF
1. OSPF divides the area based on the interface, and ISIS divides the area based on the router level (Level-1, Level-2, Level-1-2)
2. The backbone area in OSPF cannot be separated, and the separated areas need to be connected by virtual links. The backbone area (Level-2) in ISIS can be separated, but Level-1-2 needs to be used for connection before dividing the area
3. Smooth switching of ISIS regional network:
Requirement : Area 1 and Area 2 belong to the same network, but since the AR1 equipment in Area 1 needs to be replaced, new equipment needs to be replaced. However, it is updated under the premise of ensuring uninterrupted access to the network in the main area.
1. Add the ISIS network device in area 2, establish an ISIS adjacency relationship with the ISIS network device in the main area and exchange routing information.
2. Finally, area 2 successfully obtains all the routing information of area 1, and the devices in area 1 can go offline at this time. Because both area 1 and area 2 are in the same network, after area 1 goes offline, the main area can also access the network through area 2.
3. Finally, the equipment in area 2 will replace the equipment in area 1 to work until the equipment in area 1 goes online again.