L2/L3 Etherchannel Flashcards

Learning all things L2/L3 Etherchannel for Cisco devices

1
Q

What does PAgP standfor

A

Port AGgregation Protocol

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What does LACP stand for?

A

Link Aggregation Control Protocol

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the maximum number of links you can bundle in PAgP?

A

Max 8 Links

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is the maximum number of links you can bundle with LACP

A

As many as you want. However only 8 can be active at one time. All other links will be in standby mode.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Which Protocol is Cisco proprietary? LACP or PAgP?

A

PAgP

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What IEEE standard was created for LACP? Both original and new.

A

802.3ad is the original and 802.1ax is the new standard.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Which Etherchannel dynamic protocol supports half duplex links?

A

PAgP, this is one of the benefits of this protocol. LACP does not support half duplex links.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What problems is solved by etherchannel?

A

-STP won’t block extra bandwidth/increased performance and link fault tolerance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Which Etherchannel dynamic protocol uses “On”?

A

Trick questions, “on” forces the etherchannel into an up state. Does not use a dynamic protocol. Would be susceptible to loops if etherchannel misconfiguration wasn’t enabled by default.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Which Etherchannel dynamic protocol uses Passive and Active? Describe these commands.

A

LACP. Passive will negotiate LACP if the other end is configured as Active. But will not act as the aggressor. Active acts as the aggressor and will try to establish LACP.

Active/Active or Active/Passive will work but passive/passive will not work.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Which Etherchannel dynamic protocol uses Auto and Desirable? Describe these commands.

A

PAgP. Auto will negotiate PagP if the other end is configured as desirable. But will not act as the aggressor. Desirable acts as the aggressor and will try to establish PAgP.

Desirable/Desirable or Desirable/Auto will work but Auto/Auto will not work.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the default priority of LACP ports?

A

32768 - since STP and LACP are both IEEE standards they correspond in this instance.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the default load-balance setting on L2 etherchannel?

A

Source MAC address - You can check this by using the “show etherchannel load-balance” command.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

In L2 etherchannel, what configuration on the port channel and physical ports must be the same for it to work?

A
Speed
Duplex 
Trunking Protocol 
Native vlan (for 802.1Q)
Vlan permit list
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

In L3 etherchannel, what configuration on the port channel and physical ports must be the same for it to work?

A

Speed
Duplex
All must have “no switchport” command

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What two parts make up an etherchannel?

A

Port-channel interface
- logical interface representing the link bundle

Member interfaces
- Physical links that are apart of the link bundle

17
Q

A switch is connected to another switch running LACP. You preform the “Show etherchannel summary” command. You see the letter H next to one of you ports. Based on this information what is the minimum number of physical ports in the port channel?

A

9 - H is an etherchannel flag meaning a port is in hot-standby. This flag is only available for LACP. LACP can support up to 8 concurrent active ports in a port-channel every other interface after that is put into a Hot-standby mode.

*Remember, LACP is a priority based protocol for you to manage which ports will become active and which on will be standby you must configure the port priority . If you do not and everything is default priority of 32768, then the ports that will become active are the ones with the lowest port number. I.E. Port g1/0/1 has a higher priority than g1/0/2. Think Spanning-Tree.

18
Q

Etherchannel Flags - What does D(Capital) stand for?

A

The etherchannel flag D represents Down

19
Q

Etherchannel Flags - What does P(Capital) stand for?

A

The etherchannel flag P represents that the port is bundled in the port-channel.

20
Q

Etherchannel Flags - What does U(Capital) stand for?

A

The etherchannel flag P represents the port is “in use”.

21
Q

Etherchannel Flags - What does S(Capital) stand for?

A

The etherchannel flag S represents the Port channel is Layer 2

22
Q

Etherchannel Flags - What does s(lower-case) stand for?

A

The etherchannel flag s represents that the port is suspended from the interface.

23
Q

Etherchannel Flags - What does R(Capital) stand for?

A

The etherchannel flag R represents that the port channel is Layer 3

24
Q

What syslog is generated whenever one of the Layer 2 etherchannel requirements is not met? (Same speed, duplex, native vlan.. etc)

Also what etherchannel flag will help indicate this?

A

%EC-5-CANNOT_BUNNLE2: is not compatible with and will be suspended ().

The etherchannel flag will be set to lower-case s for suspended.

25
Q

Why must port speeds match for interfaces in the same port-channel?

A

When you go to hash the traffic it will not end up on the other side at the same rate. You will end up in a weird inter-leaving problem. Your TCP flows will not match up and all of your traffic will have to be re-ordered. For UDP all the packets will arrive out of order and any packet that arrives out of order and will be discarded. Therefore this is a safety mechanism to prevent this.

26
Q

You receive this error whenever you are trying to configure “channel-group 1 mode active” under an interface.

“Command rejected (channel protocol mismatch for interface fa0/19 in group 1): the interface can not be added to the channel group.”

In troubleshooting you see this in the “do sh run int fa0/19” command

Interface FastEthernet0/19
switchport trunk encapsulation dot1q
switchport mod trunk
channel-protocol pagp

What must you do in order to configure this interface? (three options):

A

1- You can configure the channel-group for auto or desirable

2- You can change the “channel-protocol” from PAgP to LACP

3- You can remove the last command with “no channel-protocol pagp” and configure the channel-group for passive and active or on.

The “channel-protocol pagp/lacp” command is a little weird. All it does is keep you from configuring the dynamic etherchannel protocol with something else. This command was create as a safety mechanism to keep you from the wrong protocol command. It’s an administrative limit you can run in your environment. However it has not found much use in Cisco devices. Some people believe it would be better if it remove the syntax for the other dynamic etherchannel protocol. But be aware of this for troubleshooting.

27
Q

What are all the load balancing methods existing on all cisco platforms? Only three answers.

A

Source and Destination Mac address
Source and Destination Ip Address
Source and Destination Layer 4

28
Q

Do you need load-balancing on both sides of the etherchannel to match?

A

No - Balancing method is logically significant and outbound.

29
Q

Is LAG independent of port mode?

A

Yes - The only things that matters is if all ports in a etherchannel match the port mode. You can etherchannel access, trunk, L3 and tunnel interfaces.

30
Q

What is the biggest issue in regards to etherchannel?

A

LAG commonly suffers from order of operation issues.

31
Q

Which set of configurations will result in all ports on both switches successfully bundling into an Etherchannel?

A. Switch 1
channel-group 2 mode active
Switch 2
channel-group 2 mode auto

B. Switch 1
channel-group 1 mode desirable
Switch 2
channel-group 1 mode passive

C. Switch 1
channel-group 1 mode auto
Switch 2
channel-group 1 mode auto

D. Switch 1
channel-group 1 mode desirable
Switch 2
channel-group 2 mode auto

A

D - PAgP uses desirable and auto. The channel-groups do not have to match, they are locally significant.

32
Q

After an Etherchannel is configured between two Cisco switches, interface port channel 1 is in the down/down state. Switch A is configured with channel-group 1 mode active, while switch B is configured with channel-group 1 mode desirable. Why is the Etherchannel bundle not working?

A. The switches are using mismatched Etherchannel negotiation modes.
B. The switch ports are not configured in trunking mode.
C. LACP priority must be configured on both sides.
D. The channel group identifier must be different for Switch A and Switch B.

A

A - LACP is Passive/Active and PAgP is Auto/Desirable

33
Q

Etherchannel bundle has been established between a Cisco switch and a corporate web server. The network administrator noticed that only one of the Etherchannel link is being utilized to reach the web server. What should be done on the Cisco switch to allow for better Etherchannel utilization to the corporate web server? (choose one)

A. Enable Cisco Express Forwarding to allow for more effective traffic sharing over the Etherchannel
B. Adjust the Etherchannel load-balancing method based on destination Ip addresses.
C. Disable spanning tree on all interfaces that are participating in the Etherchannel bundle.
D. Use link-state tracking to allow for improved load balancing of traffic upon link failure to the server.
E. Adjust the Etherchannel load-balancing method based on source Ip address.

A

E - If you adjust the load-balancing method based on destination ip the path will always be the same. Since the destination will be the server. Remember Load-balancing is logical significant. Therefore we cannot change the return traffic only the forwarding. If multiple clients are connecting to the same web server and we are doing load-balancing by source ip then each client could potentially go across different links.

34
Q

An access switch has been configured with an Etherchannel port. After configuring SPAN to monitor this port, the network administrator notices that not all traffic is being replicated to the management server. What is a cause for this issue? (Choose one)

A. VLAN filters are required to ensure traffic mirrors effectively.
B. SPAN encapsulation replication must be enabled to capture Etherchannel destination traffic.
C. The port channel can be used as a SPAN source, but not a destination.
D. RSPAN must be used to capture Etherchannel bidirectional traffic.

A

C. The port channel can be used as a SPAN source, but not as a destination.

35
Q

What is the result of this configuration? (Choose one)

Switch 1 interface gi0/1 connects to switch 2 gi0/1
Switch 1 interface gi0/2 connects to switch 2 gi0/2
XXXXXXXXXXXXXXXXXXXXXXXXXXX
hostname switch 1
! 
port-channel load-balance dst-ip
!
Interface GigabitEthernet0/1
 channel-group 10 mode active
!
Interface GigabitEthernet0/2
 channel-group 10 mode passive
!
XXXXXXXXXXXXXXXXXXXXXXXXXXX
XXXXXXXXXXXXXXXXXXXXXXXXXXX
hostname switch 2
! 
port-channel load-balance src-mac
!
Interface GigabitEthernet0/1
 channel-group 10 mode passive
!
Interface GigabitEthernet0/2
 channel-group 10 mode active
!
XXXXXXXXXXXXXXXXXXXXXXXXXXX

A. The etherchannels would not form because the load-balancing method must match on both sides.
B. The etherchannels would form and function properly even though the load-balancing and Etherchannel modes do not match.
C. The etherchannels would form, but network loops would occur because the the load-balancing methods do not match.
D. The Etherchannels would form and both devices would use the dst-ip load-balancing method because Switch 1 is configured with Etherchannel mode active.

A

B. This configuration would work because one side is active while the other is passive. They both are using LACP for their dynamic protocols.

The load-balancing method is locally significant and do not need to match. This is the reason answers A, C and D are incorrect.

36
Q

A Network Engineer tries to configure storm control on an Etherchannel bundle. What is the result of the configuration? (choose one)

A. The storm control settings will appear on the Etherchannel, but not on the associated physical ports.
B. The configuration will be rejected because storm control is not supported for Etherchannels.
C. The storm control configuration will be accepted but will only be present on the physical interfaces.
D. The settings will be applied to the Etherchannel bundle and all associated physical interfaces.

A

The correct answer is D. Storm control is supported on Etherchannels. This configuration applied on the port channel will populate to the physical members.

37
Q

A network engineer must set the load balance method on an existing port channel in a 3650 switch. Which action must be done to apply a new load balancing method?

A. Configure the new load balancing method globally using the command “port-channel load-balance”.
B. Adjust the switch SDM back to “default”
C. Configure the new load balancing method on the port-channel interface using the command “port-channel load-balance”.
D. Ensure that IP CEF is enabled globally to support all load balancing methods.

A

A is the correct answer. On Cisco 3650 switches load-balancing is a global command, all port-channels on this switch will use the same load-balancing method. Higher module switches will support different load-balancing methods for different line cards. This varies, but no Cisco device can do load-balancing per port-channel, there is some form of grouping. Also the CCNP will not contain higher end devices as it is beyond the scope.

38
Q

A network administrator sees a syslog message on one of their switches. The syslog mentions something about the etherchannel. After doing the show etherchannel summary command he notices these two etherchannel flags (Du). What should his first troubleshooting sets be?

A. To physically swap out the cables.
B. Do nothing, these are the flags meaning everything is up and working.
C. Check for mismatching factors on the physical interfaces.
D. Un-Shutdown both physical interfaces.

A

C - The flags D means down and u means unsuitable for bundling. While there are a few factors majority of these will be do to mismatching factors. Speed, duplex, trunking protocol, native vlan and vlan permit lists.

39
Q

Is order of operation important? What is the order of operation for L2 Etherchannel as apposed to L3 Etherchannel?

A

Yes, it is one of the most common problems.

L2 - Default interfaces, Create Port-channel, Configure member interfaces for that port-channel, configure port-channel interfaces.

L3 - Reverse order: default interfaces, create port-channel, configure port-channel interface, then configure member interfaces for port-channel. Remember “no switchport” needs to be in both PC and member interfaces.