取消
显示结果 
搜索替代 
您的意思是: 
cancel
23512
查看次数
70
有帮助
33
评论
xupeng
Cisco Employee
Cisco Employee
本帖最后由 xupeng 于 2015-6-1 21:51 编辑
212331of2iu0azizwdi1fi.png
故障描述:
===========

-两组由两台3750X组成的堆叠
-两组堆叠间有2条物理链路, 并且绑2层的EtherChannel
-当重启stack2 master SW3设备时, 在日志中显示stack2 member SW4的物理口G2/0/2 down
问题描述:
=========

为何重启master SW3设备,会导致member SW4的物理口down?
日志:
=====

(由于实验环境限制,系统时间不是完全同步)
stack1#
Jun 1 20:23:52.445: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet1/0/1, changed state to down
Jun 1 20:23:53.468: %LINK-3-UPDOWN: Interface GigabitEthernet1/0/1, changed state to down
Jun 1 20:23:54.886: %LINEPROTO-5-UPDOWN: Line protocol on Interface Vlan1, changed state to down
Jun 1 20:23:55.859: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet2/0/1, changed state to down
Jun 1 20:23:55.859: %LINEPROTO-5-UPDOWN: Line protocol on Interface Port-channel10, changed state to down
Jun 1 20:23:56.866: %LINK-3-UPDOWN: Interface Port-channel10, changed state to down
Jun 1 20:23:56.866: %LINK-3-UPDOWN: Interface GigabitEthernet2/0/1, changed state to down
Jun 1 20:24:00.582: %LINK-3-UPDOWN: Interface GigabitEthernet2/0/1, changed state to up
Jun 1 20:24:06.588: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet2/0/1, changed state to up
Jun 1 20:24:07.578: %LINK-3-UPDOWN: Interface Port-channel10, changed state to up
Jun 1 20:24:08.585: %LINEPROTO-5-UPDOWN: Line protocol on Interface Port-channel10, changed state to up
Jun 1 20:24:35.604: %LINEPROTO-5-UPDOWN: Line protocol on Interface Vlan1, changed state to up
stack2#
Jun 1 20:23:57.168: %LINK-5-CHANGED: Interface GigabitEthernet2/0/2, changed state to administratively down
Jun 1 20:23:57.193: %LINK-3-UPDOWN: Interface Port-channel20, changed state to down
Jun 1 20:23:58.183: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet2/0/2, changed state to down
Jun 1 20:23:59.190: %LINK-3-UPDOWN: Interface GigabitEthernet2/0/2, changed state to up
Jun 1 20:24:06.949: %LINEPROTO-5-UPDOWN: Line protocol on Interface GigabitEthernet2/0/2, changed state to up
Jun 1 20:24:07.939: %LINK-3-UPDOWN: Interface Port-channel20, changed state to up
Jun 1 20:24:08.945: %LINEPROTO-5-UPDOWN: Line protocol on Interface Port-channel20, changed state to up
Jun 1 20:24:35.957: %LINEPROTO-5-UPDOWN: Line protocol on Interface Vlan1, changed state to up
配置手册:
==========

With LACP, the system-id uses the stack MAC address from the stack master, and if the stack master changes, the LACP system-id can change. If the LACP system-id changes, the entire EtherChannel will flap, and there will be an STP reconvergence. Use the stack-mac persistent timer command to control whether or not the stack MAC address changes during a master failover.
//LACP会使用stack master的stack MAC地址作为system-id, 如果LACP的system-id会随着stack master的变化而变化. 如果LACP的system-id改变,整个EtherChannel会flap, STP将重新收敛. 可以使用'stack-mac persistent timer'命令, 控制stack MAC地址在master故障切换时是否改变.
http://www.cisco.com/c/en/us/td/ ... .html#pgfId-1454145
The MAC address of the master determines the stack MAC address. When a master is removed from the stack and a new master takes over, the MAC address of the new master to become the new stack MAC router address. However, you can set the persistent MAC address feature with a time delay before the stack MAC address changes. During this time period, if the previous master rejoins the stack, the stack continues to use that MAC address as the stack MAC address, even if the switch is now a member and not a master. You can also configure stack MAC persistency so that the stack MAC address never changes to the new master MAC address.
//master的MAC地址决定了stack的MAC地址. 当将master从stack中移除, 新的master将会接管, 新master的MAC地址将会成为新stack MAC地址.
然而, 你可以使用persistent MAC address特性去设置stack MAC地址改变的延迟时间. 在这个时间段内, 如果先前的master重新加入stack, stack将会继续使用那个MAC地址作为stack MAC地址, 即便这台设备现在成为member而不是master. 你也可以配置stack MAC persistency , 这样stack MAC地址不会随着新master 的MAC地址而改变.

You can set the time period from 0 to 60 minutes.
//计时器可设置的时间周期是0-60分钟.
• If you enter the command with no value, the default delay is 4 minutes. We recommend that you always enter a value. The time delay appears in the configuration file with an explicit timer value of 4 minutes.
//如果你不输入数值, 默认延迟为4分钟. 我们推荐你输入一个数值.这个延迟计时的数值会在配置文件中显示为4分钟.
• If you enter 0, the stack MAC address of the previous master is used until you enter the no stack-mac persistent timer global configuration command, which changes the stack MAC address to that of the current master. If you do not enter this command, the stack MAC address does not change.
//如果你输入0, 会继续使用先前master的stack MAC地址直到输入' no stack-mac persistent timer ', 它将会使stack MAC地址变成当前master的MAC地址.如果你不输入这条命令,它将不会改变stack MAC地址.
• If you enter a time delay of 1 to 60 minutes, the stack MAC address of the previous master is used until the configured time period expires or until you enter the no stack-mac persistent timer command.
//如果你输入1-60分钟, 先前master的stack MAC地址会保持到计数器超时,或者直到你配置' no stack-mac persistent timer '命令.
http://www.cisco.com/c/en/us/td/ ... .html#pgfId-1206500
实验步骤:
=========

1/
查看stack MAC地址

stack1#show switch
Switch/Stack Mac Address : 000a.000a.000a
H/W Current
Switch# Role Mac Address Priority Version State
----------------------------------------------------------
*1 Master 000a.000a.000a 1 3 Ready
2 Member 000b.000b.000b 1 3 Ready
stack2#show switch
Switch/Stack Mac Address : 000c.000c.000c
H/W Current
Switch# Role Mac Address Priority Version State
----------------------------------------------------------
*1 Master 000c.000c.000c 1 3 Ready
2 Member 000d.000d.000d 1 3 Ready
2/
查看LACP system-id

stack1#show lacp sys-id
32768, 000a.000a.000a
stack2#show lacp sys-id
32768, 000c.000c.000c
3/
查看LACP 邻居

stack1#show lacp neighbor
Flags: S - Device is requesting Slow LACPDUs
F - Device is requesting Fast LACPDUs
A - Device is in Active mode P - Device is in Passive mode
Channel group 10 neighbors
Partner's information:
LACP port Admin Oper Port Port
Port Flags Priority Dev ID Age key Key Number State
Gi1/0/1 SA 32768 000c.000c.000c 16s 0x0 0x14 0x10C 0x3D
Gi2/0/1 SA 32768 000c.000c.000c 6s 0x0 0x14 0x20C 0x3D
stack2#show lacp neighbor
Flags: S - Device is requesting Slow LACPDUs
F - Device is requesting Fast LACPDUs
A - Device is in Active mode P - Device is in Passive mode
Channel group 20 neighbors
Partner's information:
LACP port Admin Oper Port Port
Port Flags Priority Dev ID Age key Key Number State
Gi1/0/2 SA 32768 000a.000a.000a 25s 0x0 0xA 0x102 0x3D
Gi2/0/2 SA 32768 000a.000a.000a 20s 0x0 0xA 0x302 0x3D

4/
在stack1上,长ping stack2的SVI地址,观察重启stack2 master时,ping的超时情况
5/
重启stack2 master SW3
stack2#reload slot 1


6/
在stack1上观察show lacp neighbor的变化, 端口up down日志,以及生成树变化show spanning-tree

分析:
=======

-LACP的sys-id与stack MAC地址一致. 当重启stack master时, 由于默认没有修改过'stack-mac persistent timer', stack member将成为新的stack master, stack MAC地址随之改变. LACP的sys-id改变导致EtherChannel链路翻动
-由于STP重新收敛,需要等待forwarding delay. 整个过程会导致业务受到几十秒的中断.
解决办法:
===========

-在stack2上配置stack-mac persistent timer 0
-将STP改为RSTP(可选,加速收敛)
-将EtherChannel模式由LACP改为on(可选)

评论
xupeng
Cisco Employee
Cisco Employee
fanleimin 发表于 2015-6-2 09:17 back.gif
好分享

:$
xupeng
Cisco Employee
Cisco Employee
linzha3 发表于 2015-6-2 09:15 back.gif
感谢分享!

:$
xupeng
Cisco Employee
Cisco Employee
zhahong 发表于 2015-6-2 09:05 back.gif

:$
cpmld-199
Community Member
学习了,谢谢楼主分享。
xupeng
Cisco Employee
Cisco Employee
cpmld-199 发表于 2015-6-2 19:33 back.gif
学习了,谢谢楼主分享。

:$
xksun725
Level 1
Level 1
赞peng老师,很棒的分享 :)
xupeng
Cisco Employee
Cisco Employee
xksun725 发表于 2015-6-8 11:38 back.gif
赞peng老师,很棒的分享

:$
niweijian2014
Spotlight
Spotlight
谢谢分享,虽然没遇到过这案例,估计我遇着了我也不会
jing zhang
Level 1
Level 1
感谢分享,赞……
jiangygu
Cisco Employee
Cisco Employee
:)赞
xupeng
Cisco Employee
Cisco Employee
niweijian2014 发表于 2015-6-10 11:18 back.gif
谢谢分享,虽然没遇到过这案例,估计我遇着了我也不会

看完这篇文章就会了.
xupeng
Cisco Employee
Cisco Employee
thousandmoun 发表于 2015-6-10 21:45 back.gif
感谢分享,赞……

:$
xupeng
Cisco Employee
Cisco Employee
jiangygu 发表于 2015-6-11 11:36 back.gif

:$
linchen2
Community Member
赞啊
yyc25160760
Level 1
Level 1
赞 感谢分享!
入门指南

使用上面的搜索栏输入关键字、短语或问题,搜索问题的答案。

我们希望您在这里的旅程尽可能顺利,因此这里有一些链接可以帮助您快速熟悉思科社区:









快捷链接