Do the Cisco 3560-C series L3 switches support OSPF?

[britpgj21]

                   OSPF normally only comes with IPservices image and not IP Base image. The 3560-C series data sheet says that it only suport IP Base image, yet it mentions that support for OSPF in included. Are there any restictions in the OSPF support?.

[Reza Sharifi]

Even though the data sheet says that IP base support OSPF, I have not been able to enable it. In some data sheet there is also support for OSPF version-1.  No idea what is OSPF version-1, and neither have been able to enable it.

I would say you need IP Services if you need OSPF.

HTH

[Leo Laohoo]

No idea what is OSPF version-1

Not alot of people are using OSPFv1.  It's very archaic.   (Hence, it's FREE to use with IP Base feature set.)  v1 is also NOT compatible with v2 and v3.

Majority of OSPF implementation is OSPFv2.

OSPFv3 is IPv6.

[smitesh kharecha]

Hey Leo,

I was under impression that OSPFv1 was never implemented anywhere, and as a matter of fact it didn't even moved from Experimental stage.

Source: Routing TCP/IP Vol1 by Jeff Doyle.

Regards,

Smitesh

[Leo Laohoo]

I was under impression that OSPFv1 was never implemented anywhere,

I'm not sure.  Peter Paluch should be a good authority to this.

I didn't know about this until Peter pointed this out to me. 

[Peter Paluch]

Hello Leo and Smitesh,

Regarding OSPFv1, I have always referred to nobody other than John Moy, the author of OSPF himself. In his book "OSPF: Anatomy of an Internet Routing Protocol" on pages 56-57, he writes about OSPFv1:

http://books.google.sk/books?id=YXUWsqVhx60C&lpg=PP1&hl=sk&pg=PA56#v=onepage&q&f=false

"The IETF's motto is rough consensus and running code. Completing the OSPF design was the first part of the OSPF Working Group's job; now we had to prove that the protocol worked by implementing it.

Two implementations of OSPFv1 were written, one to run on Proteon routers, and another, written by Rob Coltun at the University of Maryland, to run on UNIX workstations. The latter implementation was made available in source code form for a very nominal fee. This implementation later became quite widespread and is now available as part of the GATED distribution".

According to his own words here, these implementations were done primarily to test the concept, and not to be generally deployed. During the tests, several shortcomings of OSPFv1 were uncovered. Even further in the chapter, Mr. Moy indicates:

"All of these issues led us to revise the OSPFv1 specification. A fix to the LSA sequence space could not be backward compatible [ ... cut ... ] Hence we incremented the OSPF version number, producing a specification for OSPFv2, which was eventually published as RFC 1247 in July 1991.

The failure of OSPFv1 and OSPFv2 to interoperate was not an issue, since OSPFv1 was never deployed."

So the OSPFv1 did not exist commercially - its implementations were primarily intended to gain experiences. The shortcomings of OSPFv1 probably caused that OSPFv1 never gained widespread usage. Only after rectifying the OSPF's design in OSPFv2, it began to gain momentum.

So to sum it up, apart from testing implementations of OSPFv1, it was never commercially implemented and deployed. I also do not believe that any Cisco router ever supported OSPFv1 for general deployment.

Best regards,

Peter

[smitesh kharecha]

Thanks Peter.

+5 for deep insight on OSPFv1.

And special thanks for jumping on to topic on request.

Regards,

Smitesh

[Alessio Andreoli]

HI There,

read at the boootm of this page:

http://www.cisco.com/en/US/products/ps11290/index.html

HTH

Alessio

[rasmus.elmholt]

I think OSPF is supported on the device, but routing in general is not supported by the TCAM.

SW-SYG-01#show sdm pref
 The current template is "default" template.
 The selected template optimizes the resources in
 the switch to support this level of features for
 8 routed interfaces and 1024 VLANs. 
  number of unicast mac addresses:                  4K
  number of IPv4 IGMP groups + multicast routes:    0.25K
  number of IPv4 unicast routes:                    0.875k
    number of directly-connected IPv4 hosts:        0.875k
    number of indirect IPv4 routes:                 0
  number of IPv6 multicast groups:                  0.25K
  number of IPv6 unicast routes:                    0.25K
    number of directly-connected IPv6 addresses:    0.25K
    number of indirect IPv6 unicast routes:         0
  number of IPv4 policy based routing aces:         0
  number of IPv4/MAC qos aces:                      0.375k
  number of IPv4/MAC security aces:                 0.375k
  number of IPv6 policy based routing aces:         0
  number of IPv6 qos aces:                          60
  number of IPv6 security aces:                     0.125k
 
SW-SYG-01#conf t
Enter configuration commands, one per line.  End with CNTL/Z.
SW-SYG-01(config)#sdm pref?
% Unrecognized command
SW-SYG-01(config)#sdm pref

Anyone know anything about this?

[Mark Malone]

Is that the GIG version ? looks like only 1 sdm template available but does support routing

The SDM Template

The Catalyst 3560-C Fast Ethernet switches support the same templates as other Catalyst 3560 switches. See the Catalyst 3560 Software Configuration Guide and Catalyst 3560 Command Reference for details on the templates.

The Catalyst 3560-C Gigabit Ethernet switches support only a default Switch Database Management (SDM) template, which includes support for routing and for some IPv6 features. You cannot configure SDM templates, but you can use the show sdm prefer privileged EXEC command to verify supported resources.

[rasmus.elmholt]

Hi Mark

Yes it is the Gbit version.

So is this error relevant:

IPv4 unicast indirectly-connected routes
Max Mask: 0 - Used Mask: 29 (290% utilization)
IPv4 unicast indirectly-connected routes
Max Mask: 0 - Used Mask: 29 (290% utilization)

I have one static default route, witch seems to be more than it can handle?

[Mark Malone]

You only have 1 default route that shouldn't be overkill or its kind of pointless saying it can route , looking at that output though is its not saying its maxed on /29 masks , do you have a large routing table or is it literally just your break out default route , are you using ospf ?

[rasmus.elmholt]

The Mask used is not the prefix but the amount of masks used in the TCAM. I only have one indirect route:

SW-SYG-01#show ip route

Gateway of last resort is 10.1.253.1 to network 0.0.0.0

S* 0.0.0.0/0 [1/0] via 10.1.253.1

10.0.0.0/8 is variably subnetted, 6 subnets, 2 masks

C 10.1.253.0/24 is directly connected, Vlan156

L 10.1.253.2/32 is directly connected, Vlan156

C 10.7.10.0/24 is directly connected, Vlan10

L 10.7.10.1/32 is directly connected, Vlan10

C 10.7.200.0/24 is directly connected, Vlan200

L 10.7.200.1/32 is directly connected, Vlan200

[Mark Malone]

So what exactly is the problem your seeing whats the effect of this problem in terms of your routing is it intermittent , is the  cpu seeing issues when your having problems with the default route

[rasmus.elmholt]

Besides that the CLI analyzer tool gives the error, the only problem I see is high CPU utilization on the switch when there is a lot of traffic on the network.