Re: OSPF - Load balancing

From: garry baker <baker.garry_at_gmail.com>
Date: Sat, 31 Jul 2010 12:45:04 +0300

also to follow up with the CEF you can see that pretty easy in this test,
check out the per-destination sharing vs per packet sharing when you send 4
ICMP echo requests from R2 loopback to R1 loopback interface, you can see
the per packet goes over each interface per ICMP packet and then it sticks
to s0/1 in the per-destination CEF setting:

R2#debug ip packet detail 101
IP packet debugging is on (detailed) for access list 101

R2#sh ip cef 100.100.100.100
100.100.100.100/32, version 27, epoch 0, per-packet sharing

R2#ping 100.100.100.100 r 4 source loopback 200
Type escape sequence to abort.
Sending 4, 100-byte ICMP Echos to 100.100.100.100, timeout is 2 seconds:
Packet sent with a source address of 200.200.200.200
!!!!
Success rate is 100 percent (4/4), round-trip min/avg/max = 8/56/80 ms
R2#
*Mar 1 00:22:53.339: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (FastEthernet0/1), routed via FIB
*Mar 1 00:22:53.339: IP: s=200.200.200.200 (local), d=100.100.100.100
(FastEthernet0/1), len 100, sending
*Mar 1 00:22:53.343: ICMP type=8, code=0
*Mar 1 00:22:53.415: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (FastEthernet0/0), routed via FIB
*Mar 1 00:22:53.415: IP: s=200.200.200.200 (local), d=100.100.100.100
(FastEthernet0/0), len 100, sending
*Mar 1 00:22:53.419: ICMP type=8, code=0
*Mar 1 00:22:53.479: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (Serial0/1), routed via FIB
*Mar 1 00:22:53.479: IP: s=200.200.200.200 (local), d=100.100.100.100
(Serial0/1), len 100, sending
*Mar 1 00:22:53.479: ICMP type=8, code=0
*Mar 1 00:22:53.487: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (Serial0/0), routed via FIB
*Mar 1 00:22:53.491: IP: s=200.200.200.200 (local), d=100.100.100.100
(Serial0/0), len 100, sending
*Mar 1 00:22:53.491: ICMP type=8, code=0

R2#conf t
Enter configuration commands, one per line. End with CNTL/Z.
R2(config)#int f0/0
R2(config-if)#ip load-sharing per-destination
R2#clear ip route *
R2#sh ip cef 100.100.100.100
100.100.100.100/32, version 27, epoch 0, per-destination sharing
R2#ping 100.100.100.100 r 4 source loopback 200
Type escape sequence to abort.
Sending 4, 100-byte ICMP Echos to 100.100.100.100, timeout is 2 seconds:
Packet sent with a source address of 200.200.200.200
!!!!
Success rate is 100 percent (4/4), round-trip min/avg/max = 8/49/96 ms
R2#
*Mar 1 00:25:15.807: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (Serl0/1), routed via FIB
*Mar 1 00:25:15.807: IP: s=200.200.200.200 (local), d=100.100.100.100
(Serial0/1), le 100, sending
*Mar 1 00:25:15.811: ICMP type=8, code=0
*Mar 1 00:25:15.843: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (Ser l0/1), routed via FIB
*Mar 1 00:25:15.843: IP: s=200.200.200.200 (local), d=100.100.100.100
(Serial0/1), le 100, sending
*Mar 1 00:25:15.847: ICMP type=8, code=0
*Mar 1 00:25:15.939: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (Ser l0/1), routed via FIB
*Mar 1 00:25:15.943: IP: s=200.200.200.200 (local), d=100.100.100.100
(Serial0/1), le 100, sending
*Mar 1 00:25:15.943: ICMP type=8, code=0
*Mar 1 00:25:15.999: IP: tableid=0, s=200.200.200.200 (local),
d=100.100.100.100 (Ser l0/1), routed via FIB
*Mar 1 00:25:15.999: IP: s=200.200.200.200 (local), d=100.100.100.100
(Serial0/1), le 100, sending

--
Garry L. Baker
"There is no 'patch' for stupidity." - www.sqlsecurity.com
On Sat, Jul 31, 2010 at 12:35 PM, garry baker <baker.garry_at_gmail.com> wrote:
> maybe some of this will get you going in the right direction...
>
> by default OSPF does equal cost load balancing:
>
>
> http://www.cisco.com/en/US/tech/tk365/technologies_q_and_a_item09186a0080094704.shtml#q37
>
> simple example to see it and also change the defaults to get into load
> balance over "unequal cost" paths by making them equal:
> R1#sh ip int br | e una
> Interface                  IP-Address      OK? Method Status
> Protocol
> FastEthernet0/0            1.1.1.1         YES manual up
> up
> Serial0/0                  3.3.3.1         YES manual up
> up
> FastEthernet0/1            2.2.2.1         YES manual up
> up
> Serial0/1                  4.4.4.1         YES manual up
> up
> Loopback100                100.100.100.100 YES manual up
> up
>
> R2#sh ip int brief
> Interface                  IP-Address      OK? Method Status
> Protocol
> FastEthernet0/0            1.1.1.2         YES manual up
> up
> Serial0/0                  3.3.3.2         YES manual up
> up
> FastEthernet0/1            2.2.2.2         YES manual up
> up
> Serial0/1                  4.4.4.2         YES manual up
> up
> Loopback200                200.200.200.200 YES manual up
> up
>
> R2#sh ip route  100.100.100.100
> Routing entry for 100.100.100.100/32
>   Known via "ospf 1", distance 110, metric 11, type intra area
>   Last update from 2.2.2.1 on FastEthernet0/1, 00:03:50 ago
>   Routing Descriptor Blocks:
>     2.2.2.1, from 0.0.1.0, 00:03:50 ago, via FastEthernet0/1
>       Route metric is 11, traffic share count is 1
>   * 1.1.1.1, from 0.0.1.0, 00:03:50 ago, via FastEthernet0/0
>       Route metric is 11, traffic share count is 1
> R2#conf t
> Enter configuration commands, one per line.  End with CNTL/Z.
> R2(config)#int s0/1
> R2(config-if)#ip os
> R2(config-if)#ip ospf co
> R2(config-if)#ip ospf cost 10
> R2(config-if)#int s0/0
> R2(config-if)#ip ospf cost 10
> R2(config-if)#do sh ip route  100.100.100.100
> Routing entry for 100.100.100.100/32
>   Known via "ospf 1", distance 110, metric 11, type intra area
>   Last update from 4.4.4.1 on Serial0/1, 00:00:23 ago
>   Routing Descriptor Blocks:
>     4.4.4.1, from 0.0.1.0, 00:00:23 ago, via Serial0/1
>       Route metric is 11, traffic share count is 1
>     3.3.3.1, from 0.0.1.0, 00:00:23 ago, via Serial0/0
>       Route metric is 11, traffic share count is 1
>     2.2.2.1, from 0.0.1.0, 00:04:40 ago, via FastEthernet0/1
>       Route metric is 11, traffic share count is 1
>   * 1.1.1.1, from 0.0.1.0, 00:04:40 ago, via FastEthernet0/0
>       Route metric is 11, traffic share count is 1
>
> and also a quick look at the CEF table for that route for per-packet vs.
> per-destination:
>
> R2#sh ip cef 100.100.100.100
> 100.100.100.100/32, version 29, epoch 0, per-destination sharing
> 0 packets, 0 bytes
>   via 4.4.4.1, Serial0/1, 0 dependencies
>     traffic share 1
>     next hop 4.4.4.1, Serial0/1
>     valid adjacency
>   via 3.3.3.1, Serial0/0, 0 dependencies
>     traffic share 1
>     next hop 3.3.3.1, Serial0/0
>     valid adjacency
>   via 2.2.2.1, FastEthernet0/1, 0 dependencies
>     traffic share 1
>     next hop 2.2.2.1, FastEthernet0/1
>     valid adjacency
>   via 1.1.1.1, FastEthernet0/0, 0 dependencies
>     traffic share 1
>     next hop 1.1.1.1, FastEthernet0/0
>     valid adjacency
>   0 packets, 0 bytes switched through the prefix
>   tmstats: external 0 packets, 0 bytes
>            internal 0 packets, 0 bytes
> R2(config-if)#int f0/0
> R2(config-if)#ip load-sharing per-packet
>
> R2(config-if)#do clear ip route *
>
> R2(config-if)#do sh ip cef 100.100.100.100
> 100.100.100.100/32, version 27, epoch 0, per-packet sharing
> 0 packets, 0 bytes
>   via 4.4.4.1, Serial0/1, 0 dependencies
>     traffic share 1, current path
>     next hop 4.4.4.1, Serial0/1
>     valid adjacency
>   via 3.3.3.1, Serial0/0, 0 dependencies
>     traffic share 1
>     next hop 3.3.3.1, Serial0/0
>     valid adjacency
>   via 2.2.2.1, FastEthernet0/1, 0 dependencies
>     traffic share 1
>     next hop 2.2.2.1, FastEthernet0/1
>     valid adjacency
>   via 1.1.1.1, FastEthernet0/0, 0 dependencies
>     traffic share 1
>     next hop 1.1.1.1, FastEthernet0/0
>     valid adjacency
>   0 packets, 0 bytes switched through the prefix
>   tmstats: external 0 packets, 0 bytes
>            internal 0 packets, 0 bytes
>
> http://www.cisco.com/en/US/products/hw/modules/ps2033/prod_technical_reference09186a00800afeb7.html
>
> --
> Garry L. Baker
>
> "There is no 'patch' for stupidity." - www.sqlsecurity.com
>
>
>   On Sat, Jul 31, 2010 at 10:26 AM, Mohammed Aftab Memon <
> aftabmemon12_at_gmail.com> wrote:
>
>> Hello All,
>>
>> There was some problem with my yahoo id and from days I was not getting
>> any
>> emails from this group.
>>
>> Please assist me on my below query.
>> *
>> Quote\\
>> *Hello Group,
>>
>> I have completed my preparation for OSPF but still feel that there is
>> something missing.
>>
>> When I compared it with other protocols I realized that the Load balancing
>> over OSPF is not done.
>>
>> I have been though video material of different vendors but stll I did not
>> find in any of them.
>>
>> MIght be I am wrong or might be I did not concentrate if they did cover
>> that
>> topic during their lecture.
>>
>> I did find something related to load balancing which talks about Max -
>> lsa.
>>
>> Can some one please advise me on this.
>>
>>
>> Like if we have 2 routers with 3 redundant link between them, how wil the
>> ospf load balance between them ?
>>
>>
>> Also is there any specific command with which we can see only the flow of
>> LSA's ?
>>
>> *Unquote*
>>
>> Regards
>> Mohammed Aftab Memon
>>
>> --- On *Sun, 25/7/10, Aftab Memon <aftab.memon_at_ymail.com>* wrote:
>>
>>
>> From: Aftab Memon <aftab.memon_at_ymail.com>
>> Subject: Fw: OSPF - Load balancing
>> To: "ccielab" <ccielab_at_groupstudy.com>
>> Date: Sunday, 25 July, 2010, 6:37 AM
>>
>>  FYI Group
>>
>> Regards,
>>
>> Mohd Aftab MEMON
>> Network Analyst
>>
>> --- On *Sat, 24/7/10, Aftab Memon <aftab.memon_at_ymail.com>* wrote:
>>
>>
>> From: Aftab Memon <aftab.memon_at_ymail.com>
>> Subject: OSPF - Load balancing
>> To: ccielab_at_groupstudy.com
>> Date: Saturday, 24 July, 2010, 5:49 AM
>>
>>  Hello Group,
>>
>> I have completed my preparation for OSPF but still feel that there is
>> something missing.
>>
>> When I compared it with other protocols I realized that the Load balancing
>> over OSPF is not done.
>>
>> I have been though video material of different vendors but stll I did not
>> find in any of them.
>>
>> MIght be I am wrong or might be I did not concentrate if they did cover
>> that
>> topic during their lecture.
>>
>> I did find something related to load balancing which talks about Max -
>> lsa.
>>
>> Can some one please advise me on this.
>>
>>
>> Like if we have 2 routers with 3 redundant link between them, how wil the
>> ospf load balance between them ?
>>
>>
>> Also is there any specific command with which we can see only the flow of
>> LSA's ?
>>
>> Regards,
>>
>> Mohd Aftab MEMON
>> Network Analyst
>>
>>
>> Blogs and organic groups at http://www.ccie.net
>>
>> _______________________________________________________________________
>> Subscription information may be found at:
>> http://www.groupstudy.com/list/CCIELab.html
Blogs and organic groups at http://www.ccie.net
Received on Sat Jul 31 2010 - 12:45:04 ART

This archive was generated by hypermail 2.2.0 : Sun Aug 01 2010 - 19:19:15 ART