How to check wheather Load Balancing working or not?

[muaiqel]

Hello all,

We have two internet lines connection connected to e0/0 and e0/1. And it's configured as load balancing (as the supplier said). But we encounter problem in one line because of all traffic goes to one line and rarely goes to other one.

Please how can I check out that the load balancing work? Note that CEF not configured.

Thanks in advance.

[yusuff]

If i understand you correctly, you have 2 default gateways; one through e0/0 and another e0/1

If you want to see if packets are being load balanced correctly across both the links, do the following;

- clea counters (this clears interface counters)

- show interface e0/0 & show interface e0/1

and you should see equal output packet counts on both the above outputs.

eg

sv3-11#show interfaces fastEthernet 0/0

FastEthernet0/0 is up, line protocol is up

Hardware is DEC21140A, address is 0001.c922.6a00 (bia 0001.c922.6a00)

Internet address is 10.66.79.107/27

MTU 1500 bytes, BW 100000 Kbit, DLY 100 usec,

reliability 255/255, txload 1/255, rxload 1/255

Encapsulation ARPA, loopback not set

Keepalive set (10 sec)

Full-duplex, 100Mb/s, 100BaseTX/FX

ARP type: ARPA, ARP Timeout 04:00:00

Last input 00:00:19, output 00:00:04, output hang never

Last clearing of "show interface" counters 00:00:08

Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0

Queueing strategy: fifo

Output queue :0/40 (size/max)

5 minute input rate 0 bits/sec, 0 packets/sec

5 minute output rate 0 bits/sec, 0 packets/sec

0 packets input, 0 bytes

Received 0 broadcasts, 0 runts, 0 giants, 0 throttles

0 input errors, 0 CRC, 0 frame, 0 overrun, 0 ignored

0 watchdog

0 input packets with dribble condition detected

1 packets output, 60 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 output buffer failures, 0 output buffers swapped out

sv3-11#

in the abvoe example, see the 'packets output' count is 1, this what you should check on both your interfaces.

HTH

R/Yusuf

[Gilles Dufour]

To have loadbalancing you need 2 routes (or more) for each destination.

The simple case is 2 default routes.

So, do a 'sh ip route 0.0.0.0' and make sure you see 2 possible next-hops on 2 different interfaces.

If you only see one, you have a config problem.

If you're using BGP and get the full routing table, this might be trickier.

So, check if you are using static default routes or not.

If using BGP, just let us know and will tell you what to do next.

Hope this help,

Gilles.

[muaiqel]

Hi, thanks for your reply...

We don't use BGP. when running the command "sh ip rout 0.0.0.0" it gives the following:

Known via "static", distance 1, metric 0, candidate default path

Routing Descriptor Blocks:

10.1.0.138

Route metric is 0, traffic share count is 1

* 10.0.0.138

Route metric is 0, traffic share count is 1

But most of our traffic (approx. 98%) goes throwgh e0/0. Any Ideas?

Thanks in advance.

[Gilles Dufour]

asI saidin my previous answer, you need to see 2 next-hops (one for each interface).

The result of the command indicates you have only 1 next-hop.

You need to configure a 2nd one.

Also, could you do a 'sh ip route summ'

If you don't have a lot of routes, capture a 'sh ip route' and send it to us.

Regards,

Gilles.

[muaiqel]

How do u determine that it's only one next-hop? There r two 10.1.0.138 and 10.0.0.138. Am I wrong. If yes, how to configure 2nd one? Please explain as much as u can.

This is the result of "sh ip rout summ":

Route Source Networks Subnets Overhead Memory (bytes)

connected 0 3 192 432

static 2 2 320 576

internal 1 1164

Total 3 5 512 2172

Thanks for your help.

[Gilles Dufour]

ok, you don't have a lot of routes. So, you can send me a complete 'sh ip route'.

You need to see 10.10.138 and 10.1.0.139

Use the command I gave you previously to configure the 2nd next-hop.

'ip route 0.0.0.0 0.0.0.0 10.1.0.139'

[muaiqel]

Hi,

for that command ... it's already there in the run configuration:

ip route 0.0.0.0 0.0.0.0 10.0.0.138

ip route 0.0.0.0 0.0.0.0 10.1.0.138

Do I have to add 'ip route 0.0.0.0 0.0.0.0 10.1.0.139'?

The "sh ip route" for the static route is:

S 10.7.1.0 [1/0] via 10.2.0.2

S 10.6.1.0 [1/0] via 10.2.0.2

S* 0.0.0.0/0 [1/0] via 10.1.0.138

[1/0] via 10.0.0.138

Appretiate your help...

[Gilles Dufour]

You have only 1 next-hop define there.

If .139 is the 2nd next-hop, you need to setup a 2nd route pointing to it.

So, yes, configure 'ip route 0.0.0.0 0.0.0.0 10.1.0.139'.

Then issue a 'sh ip route static' again and you will see the difference.

If you have 2 routes to reach .138, then you don't need the command above.

However, to check that the routing table is ok, do a 'sh ip route 10.1.0.138'

But only if you have 2 routes to reach 10.1.0.138.

[muaiqel]

First .. I want to thank you very much for your detailed answer and I really appretiate that ...

Second ... I did what you said ( clear and show ) but I noticed the following:

- All our traffic goes throwgh e0/0. And this the information about e0/0:

22784 packets output, 4450592 bytes, 0 underruns

5 minute input rate 220000 bits/sec, 27 packets/sec

5 minute output rate 28000 bits/sec, 19 packets/sec

- Nothing or few packets goes throwgh e0/1 and this is the information about e0/1:

932 packets output, 61546 bytes, 0 underruns

5 minute input rate 0 bits/sec, 0 packets/sec

5 minute output rate 0 bits/sec, 0 packets/sec

See the diffrence? Any Ideas?

Thanks in advance.

[srenga]

Try enabling per packet load balancing on u'r incoming interface.

command is "ip load-sharing per-packet"

[muaiqel]

Hi,

How to check the current state of the load sharing? Per Packets ot Per Destinations?

Also, Should I configure it on the two interfaces?

Thanks in advance.