RE: Does ISL Support...........?

From: Daniel_Steyn@Dell.com
Date: Mon Nov 13 2006 - 18:10:05 ART


Nice find, Venkatesh. That's what we were looking for. Thanks for the
correction.

________________________________

From: Venkatesh Venkatesh [mailto:kvpalani@gmail.com]
Sent: Monday, November 13, 2006 11:57 AM
To: Steyn, Daniel
Cc: acer0001@gmail.com; ccielab@groupstudy.com
Subject: Re: Does ISL Support...........?

All frames should be tagged in ISL bit not the case is 802.1Q

REF cisco.com

"In an ISL trunk port, all received packets are expected to be
encapsulated with an ISL header, and all transmitted packets are sent
with an ISL header. Native (non-tagged) frames received from an ISL
trunk port are dropped.

* <http://www.cisco.com/univercd/illus/images/blank.gif> An IEEE 802.1Q
trunk port supports simultaneous tagged and untagged traffic. An 802.1Q
trunk port is assigned a default Port VLAN ID (PVID), and all untagged
traffic travels on the port default PVID. All untagged traffic and
tagged traffic with a NULL VLAN ID are assumed to belong to the port
default PVID. A packet with a VLAN ID equal to the outgoing port default
PVID is sent untagged. All other traffic is sent with a VLAN tag.

"

- Venkatesh

On 11/13/06, Daniel_Steyn@dell.com <Daniel_Steyn@dell.com> wrote:

        Well...it does LET you configure a native vlan on ISL. Having
said
        that, ISL will ALWAYS tag egress traffic - even the "native"
vlan. My
        belief is that (correct me if I am wrong, please) the native
VLAN will
        mark INGRESS UNTAGGED traffic with the specified native vlan
value. I
        do understand that between 2 switches running ISL, there should
be no
        untagged traffic (they tag all VLANs), however, some NICs do
allow you
        to run ISL in which the possibility of untagged traffic is
present. But
        even if that were the case...traffic would be tagged upon the
return, so
        I'm not sure to be honest. We'll have to fire it up in a lab
and test.

        switch> (enable) show trunk 1/3
        * - indicates vtp domain mismatch
        # - indicates dot1q-all-tagged enabled on the port
        $ - indicates non-default dot1q-ethertype value
        Port Mode Encapsulation Status Native vlan
        -------- ----------- ------------- ------------ -----------
        1/3 on isl trunking* 400

        Port Vlans allowed on trunk
        --------

---------------------------------------------------------------------
        1/3

4-7,9-12,26-27,34-35,39-40,76,87-88,90,93,127,134,187,200,216-217,220,23
        3,251,255,399-402,408,410,415,500,725,802,874,985

        Brian/Scott or some of the other serious gurus may be able to
provide us
        with the answer on this one.

        ________________________________

        From: John Jones [mailto: acer0001@gmail.com
<mailto:acer0001@gmail.com> ]
        Sent: Monday, November 13, 2006 10:05 AM
        To: ccielab@groupstudy.com; Steyn, Daniel
        Subject: Does ISL Support...........?

        I would like info on this as well. I ask because I always
thought that
        the concept of ISL and native VLANs didn't mix. I haven't tested
this
        though. See below:

        ISL also encapsulates the entire frame, increasing the network
overhead.
        Dot1q only places a header on the frame, and in some
circumstances,
        doesn't even do that. There is much less overhead with dot1q as
compared
        to ISL. That leads to the third major difference, the way the
protocols
        work with the native vlan.

        The native vlan is simply the default vlan that switch ports are
placed
        into if they are not expressly placed into another vlan. On
Cisco
        switches, the native vlan is vlan 1. (This can be changed.) If
dot1q is
        running, frames that are going to be sent across the trunk line
don't
        even have a header placed on them; the remote switch will assume
that
        any frame that has no header is destined for the native vlan.

        The problem with ISL is that is doesn't understand what a native
vlan
        is. Every single frame will be encapsulated, regardless of the
vlan it's
        destined for.

http://www.networkliquidators.com/article-cisco-ccna-certification-how-a
        nd-why-switches-trunk.asp

        Also...

http://www.ezinearticles.com/?Cisco-CCNA-/-CCNP-/-BCMSN-Exam-Review:--Tr
        unking-And-Trunking-Protocols&id=195572

        John

        On 11/13/06, Alexei Monastyrnyi <alexeim@orcsoftware.com >
wrote:

               and any details/URL to read on this?

               Daniel_Steyn@Dell.com wrote:
> ISL does support a native VLAN in which ingress
untagged VLANs
        are then
> tagged with the vlan specified.
>
>
> -----Original Message-----
> From: nobody@groupstudy.com
[mailto:nobody@groupstudy.com] On
        Behalf Of
> Alexei Monastyrnyi
> Sent: Monday, November 13, 2006 9:19 AM
> To: Rajiv
> Cc: ccielab@groupstudy.com
> Subject: Re: Does ISL Support...........?
>
> Hi.
>
> As per Q1, since ISL adds its specific L2 header to
each
        frame, all
> frames have to have it, otherwise the frame should be
        considered as
> invalid. In this sense there should be no untagged
frames for
        ISL.
>
> As per Q2, if I remember right, both protocols support
up to
        4095 VLANs.
>
> HTH
> A.
>
> Rajiv wrote:
>
>> Hi all,
>>
>> I want to ask that Does ISL supports the processing
of
        untagged
>>
> frames?
>
>>
>> Also does 802.1q supports fewer VLANs than ISL?
>>
>> Regards,
>> Rajiv
>>
>>
>> ---------------------------------
>> Everyone is raving about the all-new Yahoo! Mail beta.
>>
>>



This archive was generated by hypermail 2.1.4 : Fri Dec 01 2006 - 08:05:46 ART