04-16-2013 11:08 AM - edited 03-10-2019 12:21 PM
I have an inquiry about a configuration I deploy in a C2960 switch. I have configured a ether channel with 8 ports, the load balanced method is source mac address. The bundling protocol is LACP.
I have found the ether channel is not balanced as I expect. One of the eight interfaces is congested.
*Detail of congestion interface:
|
… GigabitEthernet1/0/2 is up, line protocol is up (connected) … reliability 255/255, txload 238/255, rxload 32/255 ***congested interface*** …
|
I been reviewing some documents about the configuration of the ether channel and configuration looks and behave as is expected. Port channel all interfaces are active:
switch#sh int po1 Port-channel1 is up, line protocol is up (connected) Hardware is EtherChannel, address is ........... Description: INSIDE-MASTER MTU 1500 bytes, BW 8000000 Kbit, DLY 10 usec, reliability 255/255, txload 34/255, rxload 33/255 Encapsulation ARPA, loopback not set Keepalive set (10 sec) Full-duplex, 1000Mb/s, link type is auto, media type is unknown input flow-control is off, output flow-control is unsupported Members in this channel: Gi1/0/1 Gi1/0/2 Gi2/0/1 Gi2/0/2 Gi3/0/1 Gi3/0/2 Gi4/0/1 Gi4/0/2 ***eight interfaces*** ARP type: ARPA, ARP Timeout 04:00:00 Last input never, output 00:00:00, output hang never Last clearing of "show interface" counters 20w0d Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 1662386899 ***Drops for congested interface GigabitEthernet1/0/2*** … | … Queueing strategy: fifo Output queue: 0/40 (size/max) 30 second input rate 1053373000 bits/sec, 165017 packets/sec 30 second output rate 1082746000 bits/sec, 173079 packets/sec 755178023967 packets input, 628737602889253 bytes, 0 no buffer Received 3681164 broadcasts (3234870 multicasts) 0 runts, 0 giants, 0 throttles 0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored 0 watchdog, 3234870 multicast, 0 pause input 0 input packets with dribble condition detected 852162422509 packets output, 699901291834183 bytes, 0 underruns 0 output errors, 0 collisions, 0 interface resets 0 babbles, 0 late collision, 0 deferred 0 lost carrier, 0 no carrier, 0 PAUSE output 0 output buffer failures, 0 output buffers swapped out … |
Here is the LACP information
switch#sh lacp 1 internal 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 1 LACP port Admin Oper Port Port Port Flags State Priority Key Key Number State Gi1/0/1 SA bndl 32768 0x1 0x1 0x102 0x3D Gi1/0/2 SA bndl 32768 0x1 0x1 0x103 0x3D Gi2/0/1 SA bndl 32768 0x1 0x1 0x202 0x3D Gi2/0/2 SA bndl 32768 0x1 0x1 0x203 0x3D Gi3/0/1 SA bndl 32768 0x1 0x1 0x302 0x3D Gi3/0/2 SA bndl 32768 0x1 0x1 0x303 0x3D Gi4/0/1 SA bndl 32768 0x1 0x1 0x402 0x3D Gi4/0/2 SA bndl 32768 0x1 0x1 0x403 0x3D | switch#sh lacp 1 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 1 neighbors Partner's information: LACP port Admin Oper Port Port Port Flags Priority Dev ID Age key Key Number State Gi1/0/1 SA 255 0009.0fe7.5aed 23s 0x0 0x11 0x1 0x3D Gi1/0/2 SA 255 0009.0fe7.5aed 22s 0x0 0x11 0x2 0x3D Gi2/0/1 SA 255 0009.0fe7.5aed 22s 0x0 0x11 0x3 0x3D Gi2/0/2 SA 255 0009.0fe7.5aed 21s 0x0 0x11 0x4 0x3D Gi3/0/1 SA 255 0009.0fe7.5aed 23s 0x0 0x11 0x5 0x3D Gi3/0/2 SA 255 0009.0fe7.5aed 19s 0x0 0x11 0x6 0x3D Gi4/0/1 SA 255 0009.0fe7.5aed 7s 0x0 0x11 0x7 0x3D Gi4/0/2 SA 255 0009.0fe7.5aed 30s 0x0 0x11 0x8 0x3D |
The only difference I’ve found against the cisco documentation is that when the etherchannel was configured the interfaces where added in UP state.
Since I suspect this would affect the etherchannel hash calculation, I would like to know is if I shutdown the etherchannel (all interfaces goes down) and then “no shutdown” over the etherchannel , it would calculate again the etherchannel hash and redistribute the traffic over the other ports.
I appreciate any help,
Best Regards
04-16-2013 11:19 AM
Hello Eleazar,
I do not know if shut/no shut will make difference. I think that either nothing change or some other interface will become congested.
You should change load balancing method because src-mac method is probably not diversing traffic equally.
Best Regards
Please rate all helpful posts and close solved questions
04-16-2013 01:59 PM
You should look at what traffic is most diverse, is the source mac the most different or is the destination Mac? Maybe it's the destination ip or source ip.
Look at the direction that the majority of your traffic flows to and then find what the biggest distinguished is of that traffic.
Or if you don't want to go through the work, you can just change the load balancing method until you get the desired results
Sent from Cisco Technical Support iPhone App
Discover and save your favorite ideas. Come back to expert answers, step-by-step guides, recent topics, and more.
New here? Get started with these tips. How to use Community New member guide