Cisco ASR 1002 , traffic output is greater than traffic input ??? why ?

Dr.X · ‎10-26-2015

Hi ,

LAN-------ASR----------internet

i have cisco ASR 1002 which has 4 G interfaces .

i have port channel with the internet provider , it include Gi0/0 & Gi0/4

i have 2 other ports going to LAN , it include Gi0/1 & Gi0/2

both ports are loadbalanced using EIGRP protocol , that mean GI0/1 & Gi0/2 has same upload/download traffic .

i have my total BW download as 1.7 Gbps

in the rush hour i have each LAN interfaces about equally traffic as 1.7/2 = 850 Mbps on Gi0/1 and same 850 Mbps on Gi0/2

i have done limit on int gi0/1 as 840Mbps and gi0/2 limit to 840Mbps

now why the total summ of the Gi0/1 & Gi0/2 is greater than the internet of the port channle ????

here is what i mean :

Gateway-ASR1002#sh int gigabitEthernet 0/0/1
GigabitEthernet0/0/1 is up, line protocol is up
Hardware is 4XGE-BUILT-IN, address is d072.dccd.a181 (bia d072.dccd.a181)
Description: ASR-7600-Link1
Internet address is 172.20.40.1/30
MTU 1500 bytes, BW 1000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 214/255, rxload 21/255
Encapsulation ARPA, loopback not set
Keepalive not supported
Full Duplex, 1000Mbps, link type is auto, media type is T
output flow-control is on, input flow-control is on
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:37, output 00:06:37, output hang never
Last clearing of "show interface" counters never
Input queue: 0/375/0/0 (size/max/drops/flushes); Total output drops: 3362565180
Queueing strategy: Class-based queueing
Output queue: 0/40 (size/max)
30 second input rate 83944000 bits/sec, 62166 packets/sec
30 second output rate 840008000 bits/sec, 89672 packets/sec
92234037878 packets input, 13671514423590 bytes, 0 no buffer
Received 3 broadcasts (0 IP multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 1603543 multicast, 0 pause input
130898672180 packets output, 157021256233213 bytes, 0 underruns
0 output errors, 0 collisions, 2 interface resets
36564 unknown protocol drops
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
Gateway-ASR1002#sh int gigabitEthernet 0/0/2
GigabitEthernet0/0/2 is up, line protocol is up
Hardware is 4XGE-BUILT-IN, address is d072.dccd.a182 (bia d072.dccd.a182)
Description: ASR-7600-Link2
Internet address is 176.58.76.2/30
MTU 1500 bytes, BW 1000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 214/255, rxload 46/255
Encapsulation ARPA, loopback not set
Keepalive not supported
Full Duplex, 1000Mbps, link type is auto, media type is T
output flow-control is on, input flow-control is on
ARP type: ARPA, ARP Timeout 04:00:00
Last input 00:00:43, output 00:06:39, output hang never
Last clearing of "show interface" counters never
Input queue: 1/375/0/0 (size/max/drops/flushes); Total output drops: 1059194221
Queueing strategy: Class-based queueing
Output queue: 0/40 (size/max)
30 second input rate 180430000 bits/sec, 127661 packets/sec
30 second output rate 839979000 bits/sec, 136485 packets/sec
164174177430 packets input, 28282695179348 bytes, 0 no buffer
Received 3 broadcasts (0 IP multicasts)
0 runts, 0 giants, 0 throttles
0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored
0 watchdog, 1603792 multicast, 0 pause input
192556562568 packets output, 164348506635962 bytes, 0 underruns
0 output errors, 0 collisions, 2 interface resets
36573 unknown protocol drops
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

Gateway-ASR1002#sh int port-channel 2
Port-channel2 is up, line protocol is up
Hardware is GEChannel, address is d072.dccd.a241 (bia d072.dccd.a241)
Description: ASR-port-channel
Internet address is 213.244.66.78/30
MTU 1500 bytes, BW 2000000 Kbit/sec, DLY 10 usec,
reliability 255/255, txload 21/255, rxload 207/255
Encapsulation ARPA, loopback not set
Keepalive set (10 sec)
ARP type: ARPA, ARP Timeout 04:00:00
No. of active members in this channel: 2
Member 0 : GigabitEthernet0/0/0 , Full-duplex, 1000Mb/s
Member 1 : GigabitEthernet0/0/3 , Full-duplex, 1000Mb/s
No. of PF_JUMBO supported members in this channel : 2
Last input 02:34:09, output never, output hang never
Last clearing of "show interface" counters never
Input queue: 0/750/0/0 (size/max/drops/flushes); Total output drops: 0
Queueing strategy: fifo
Output queue: 0/80 (size/max)
30 second input rate 1629061000 bits/sec, 177162 packets/sec
30 second output rate 165222000 bits/sec, 133120 packets/sec
271425990142 packets input, 319939292858216 bytes, 0 no buffer
Received 1 broadcasts (0 IP multicasts)
0 runts, 0 giants, 0 throttles
31442 input errors, 0 CRC, 0 frame, 31442 overrun, 0 ignored
0 watchdog, 129014 multicast, 0 pause input
195271410284 packets output, 28665406306356 bytes, 0 underruns
0 output errors, 0 collisions, 0 interface resets
0 unknown protocol drops
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
Gateway-ASR1002#

if we look @ the red above

we have internet enters router with port channel interface = download 1629061000 bits/sec =====1.62 Gb/ps

look @ interfaces GI0/1 & Gi0/2 the out is about 840008000 bits/sec ....lets stay 840*2 for 2 interfaces = 1.68 Gbps

the question is being asked here is

why the traffic inter the router not equal exactly the traffic exit the router althoug no other thing connected that may be a possible thing to do that ????

i belive that there is 30 Sec load interval , but so far i dont understand why the traffic not equal ???

plz look @ the red parameters above and see the difference

thanks alot

Joseph W. Doherty · ‎10-26-2015

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

You're assuming the port-channel member interfaces and the port-channel are measuring the same 30 seconds; if not, that could account for the difference.

Dr.X · ‎10-26-2015

Hi josef

all the interfaces in the router has load interval 30 sec

thank you

Joseph W. Doherty · ‎10-26-2015

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

Yes, I saw that. But again, are they counting the same 30 seconds? They may be if you load the router with that config, but, for example, might not if you change the load interval while the router is running.

Dr.X · ‎10-27-2015

i think your question here is logical "are they counting the same 30 seconds " .

great

but assume they are not counting the same 30 seconds.

why the traffic is always higher ?

wasnt supposed to see sometimes the traffic equal in & out?????

why always the out less than the in ??

do you think there is other possibilities such as like the router save some incomming packets as buffer in memory ?

thank you for your kind response

cheers

Joseph W. Doherty · ‎10-27-2015

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

Ah, that's an excellent point, i.e. always higher. If the interfaces are counting against different measurement intervals, I would expect the port-channel to sometimes be lower too.

Ok, so assuming it's not, perhaps Cisco is "counting" some traffic for the port-channel only, such as LACP (if you're using that) or your EIGRP packets, that it also doesn't count across the member ports. Or, it might be some bug in your IOS.

At this point to run something like this down, you would probably need to open a TAC case.

Dr.X · ‎10-27-2015

thanks for you kind response agian .

but let me ask you more .

as i understood , when we do traffic shaping , the router can save some traffic in memory buffer .

if if you imagine it as a water pump that enter the router port-channel

and assume u save some water in a tank and other water go to other side .

(((so i think that traffic shaping behavior is the reason for that ))

i think this is story we can imagine what i have with my router

but agian im not sure of what im saying

maybe in low traffic , we dont notice that , but in traffic like 2Gbps we noticed it

i still not sure

anyway thanks a lot for your kind support and you always were nice man for helping .

cheers

Joseph W. Doherty · ‎10-27-2015

Disclaimer

The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.

Liability Disclaimer

In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.

Posting

A traffic shaper creates an egress point with less bandwidth than it naturally has.

For example, let us assume you have a 10/100 Ethernet port. Your can run the port at 100 Mbps or 10 Mbps, physically. However, on many Cisco routers, you could run the interface physically at 100 Mbps, but shape at 10 Mbps. If you do that, the shaper tries to emulate the traffic bandwidth of the interface physically running at 10 Mbps.

In the prior example, it doesn't make much sense to shape the interface at 10 Mbps, when you might be able to physically run it at 10 Mbps, but the shaper can do similar for other bandwidths when there is no physical option. For example, let's say you want to shape the Ethernet interface at 45 Mbps so it behaves much like a T3 or 34 Mbps so it behaves much like a E3.

The prior, hopefully, will help understand a shaper a little better, but in any case, unless we're dealing with a bug, I wouldn't see how a shaper would change the relationship between a port-channel's active egress bandwidth vs. the aggregate active egress bandwidths of all its active member ports.