Antonio has a good example to play with:

http://www.ccie18473.net/dynamips/dynamips.htm#isis-att-bit

2009/12/21 André Luiz Bernardes <[email protected]>

> Adam, you mentioned that we can overcome the default att bit behavior with
> set-attached-bit feature. I suppose this could be used where default route
> between L1 and L1/L2 routers is wanted but they are on same areas.
>
> Any practical examples on that? I’m gonna google it…
> Andre
>
>
>
> On Mon, Dec 21, 2009 at 5:10 PM, Bryan Bartik <[email protected]>wrote:
>
>> Right, the key thing to remember is that it's not the level type (1 or 2)
>> that determines whether the ATT bit set, but whether there is a peer in a
>> different from an L1 peer on the router. That peer can be L1 (multi-area) or
>> L2. Example:
>>
>> R1-----area 49.0012-----R2-----area 49.0025-----R5
>>
>> R2 has two L1 neighbors (under two processes of course).
>> R2 would set the ATT bit in both L1 LSP, one for each process.
>> R1 and R5 would both have default routes towards R2.
>>
>>
>> On Mon, Dec 21, 2009 at 10:31 AM, Vitkovsky, Adam <
>> [email protected]> wrote:
>>
>>>  Yes please if there are multiple is-is processes configured -the router
>>> starts to advertise the att-bit in the l1 lsp
>>>
>>> But I believe this would only be true if the processes would have been
>>> configured with different area numbers
>>>
>>>
>>>
>>> Also I’d like to ask regarding the default route
>>>
>>> I’m not sure if I remember that correctly but I guess I didn’t see the
>>> default route in the link state database on the l1/2 router –and saw it only
>>> on the l1 router
>>>
>>> Almost like the l1 router creates the default route on its own based on
>>> the att bit received
>>>
>>>
>>>
>>> Adam Vitkovsky
>>>  ------------------------------
>>>
>>> *From:* Rick Mur [mailto:[email protected]]
>>> *Sent:* Monday, December 21, 2009 5:15 PM
>>> *To:* Bryan Bartik
>>> *Cc:* Vitkovsky, Adam; [email protected]
>>>
>>> *Subject:* Re: [OSL | CCIE_SP] CCIE_SP Digest, Vol 33, Issue 23
>>>
>>>
>>>
>>> Does this mean that the router is aware of multiple IS-IS processes and
>>> therefore advertises the default with ATT-bit? (only way to have multiple
>>> level 1 areas on 1 router)
>>>
>>>
>>>
>>> Learning something new every day :-) Thanks for correcting me Bryan.
>>>
>>>
>>>
>>>
>>>
>>> --
>>> Regards,
>>>
>>> Rick Mur
>>> CCIE2 #21946 (R&S / Service Provider)
>>> Sr. Support Engineer – IPexpert, Inc.
>>> URL: http://www.IPexpert.com
>>>
>>>
>>>
>>> On 21 dec 2009, at 17:02, Bryan Bartik wrote:
>>>
>>>
>>>
>>>  Adam,
>>>
>>> That is right. In fact, if you have a router connected to multiple areas
>>> via level 1, ATT bit is set as well by the router that is connected to
>>> multiple areas.
>>>
>>> On Mon, Dec 21, 2009 at 8:20 AM, Vitkovsky, Adam <
>>> [email protected]> wrote:
>>>
>>> Yes please I meant l2 routers in this case :)
>>> Though with the multi area feature the l1/2 router can connect to l1
>>> routers in different areas
>>> I believe the att bit is set automatically in l1 lsp only when the L1/2
>>> router interconnects different areas (via connections to l1 or l2 routers)
>>>
>>>
>>> Adam Vitkovsky
>>>
>>>
>>> -----Original Message-----
>>> From: Rick Mur [mailto:[email protected]]
>>>
>>> Sent: Monday, December 21, 2009 3:34 PM
>>> To: Vitkovsky, Adam
>>> Cc: [email protected]
>>>
>>> Subject: Re: [OSL | CCIE_SP] CCIE_SP Digest, Vol 33, Issue 23
>>>
>>> But an L1L2 router cannot connect to an L1 router in another area, since
>>> it shares the same L1 domain :-)
>>> You might be right though when the L2 part of the L1L2 router is
>>> connected to an L2 router in another area, but I can't recall that from
>>> things I tested.
>>>
>>> --
>>> Regards,
>>>
>>> Rick Mur
>>> CCIE2 #21946 (R&S / Service Provider)
>>> Sr. Support Engineer - IPexpert, Inc.
>>> URL: http://www.IPexpert.com
>>>
>>> On 21 dec 2009, at 13:09, Vitkovsky, Adam wrote:
>>>
>>> > I believe l1/l2 router will advertise a default route only when it
>>> connects two different areas as defined by the net id (ie 49.xxxx.)by
>>> default
>>> > -are the r1 and r5 routers connected to different area than the r2 is a
>>> part of?
>>> > You should be able to overcome this behavior by using the att bit
>>> control feature
>>> > -where you can set the att bit based on the match in the route-map
>>> > In that route-map you'll match for the net id (clns address) of the
>>> upstream router/s
>>> >
>>> >
>>> > Adam Vitkovsky
>>> >
>>> > -----Original Message-----
>>> > From: [email protected] [mailto:
>>> [email protected]] On Behalf Of
>>> [email protected]
>>> > Sent: Sunday, December 20, 2009 4:46 PM
>>> > To: [email protected]
>>> > Subject: CCIE_SP Digest, Vol 33, Issue 23
>>> >
>>> > Send CCIE_SP mailing list submissions to
>>> >       [email protected]
>>> >
>>> > To subscribe or unsubscribe via the World Wide Web, visit
>>> >       http://onlinestudylist.com/mailman/listinfo/ccie_sp
>>> > or, via email, send a message with subject or body 'help' to
>>> >       [email protected]
>>> >
>>> > You can reach the person managing the list at
>>> >       [email protected]
>>> >
>>> > When replying, please edit your Subject line so it is more specific
>>> > than "Re: Contents of CCIE_SP digest..."
>>> >
>>> >
>>> > Today's Topics:
>>> >
>>> >   1. Re: Volume 2 section 1 Multiprotocol Challange lab       ISIS task
>>> >      2.2 (Thomas J. Loran)
>>> >
>>> >
>>> > ----------------------------------------------------------------------
>>> >
>>> > Message: 1
>>> > Date: Sun, 20 Dec 2009 15:45:48 +0000
>>> > From: "Thomas J. Loran" <[email protected]>
>>> > Subject: Re: [OSL | CCIE_SP] Volume 2 section 1 Multiprotocol
>>> >       Challange lab   ISIS task 2.2
>>> > To: Rick Mur <[email protected]>
>>> > Cc: Marko, "[email protected]" <[email protected]>
>>> > Message-ID: <[email protected]>
>>> > Content-Type: text/plain; charset="us-ascii"
>>> >
>>> > Yes with the is-type level-1 command
>>> >
>>> > See the whole isis database and notice no L2 database.  I did a R1, R2,
>>> R5 reload and it did not fix the issue
>>> >
>>> > R2#sh isis database detail
>>> >
>>> > IS-IS Level-1 Link State Database:
>>> > LSPID                 LSP Seq Num  LSP Checksum  LSP Holdtime
>>>  ATT/P/OL
>>> > R1.00-00              0x00000015   0x737B        918
>>> 0/0/0
>>> >  Auth:         Length: 9
>>> >  Area Address: 49.1155
>>> >  NLPID:        0x81 0xCC
>>> >  Hostname: R1
>>> >  IP Address:   125.125.125.1
>>> >  Metric: 2          IP 172.16.21.0/24
>>> >  Metric: 2          IP 125.125.125.1/32
>>> >  Metric: 2          IS-Extended R2.00
>>> >  Metric: 0          ES R1
>>> > R2.00-00            * 0x0000000F   0x457A        1022
>>>  0/0/0
>>> >  Auth:         Length: 9
>>> >  Area Address: 49.1155
>>> >  NLPID:        0xCC
>>> >  Hostname: R2
>>> >  IP Address:   125.125.125.2
>>> >  Metric: 10         IP 172.16.52.0/24
>>> >  Metric: 10         IP 172.16.21.0/24
>>> >  Metric: 0          IP 125.125.125.2/32
>>> >  Metric: 10         IS-Extended R1.00
>>> >  Metric: 10         IS-Extended R5.03
>>> > R5.00-00              0x00000015   0xBBF9        1019
>>>  0/0/0
>>> >  Auth:         Length: 9
>>> >  Area Address: 49.1155
>>> >  NLPID:        0x81 0xCC
>>> >  Hostname: R5
>>> >  IP Address:   125.125.125.5
>>> >  Metric: 2          IP 172.16.52.0/24
>>> >  Metric: 2          IP 125.125.125.5/32
>>> >  Metric: 2          IS-Extended R5.03
>>> >  Metric: 0          ES R5
>>> > R5.03-00              0x00000001   0x3976        1019
>>>  0/0/0
>>> >  Auth:         Length: 9
>>> >  Metric: 0          IS-Extended R5.00
>>> >  Metric: 0          IS-Extended R2.00
>>> >
>>> > From: Rick Mur [mailto:[email protected]]
>>> > Sent: Sunday, December 20, 2009 8:42 AM
>>> > To: Thomas J. Loran
>>> > Cc: [email protected]; Marko Milivojevic
>>> > Subject: Re: Volume 2 section 1 Multiprotocol Challange lab ISIS task
>>> 2.2
>>> >
>>> > Is R2 a full L1 router? L1/L2 routers do not receive the default, only
>>> pure L1 ones.
>>> >
>>> > Otherwise try it with a different version, I know the proctorlabs 7200s
>>> have a bit old release and there are some strange bugs in it. Maybe this is
>>> one of them. For the workbook lab it doesn't matter, technological you are
>>> correct :-)
>>> >
>>> > Try it with Dynamips on the latest 12.2S release.
>>> >
>>> > --
>>> > Regards,
>>> >
>>> > Rick Mur
>>> > CCIE2 #21946 (R&S / Service Provider)
>>> > Sr. Support Engineer - IPexpert, Inc.
>>> > URL: http://www.IPexpert.com
>>> >
>>> > On 20 dec 2009, at 16:23, Thomas J. Loran wrote:
>>> >
>>> >
>>> > I should have mentioned that this was before route-leaking
>>> >
>>> >
>>> > From: Rick Mur [mailto:[email protected]]
>>> > Sent: Sunday, December 20, 2009 8:20 AM
>>> > To: Thomas J. Loran
>>> > Cc: [email protected]<mailto:[email protected]>;
>>> Marko Milivojevic
>>> > Subject: Re: Volume 2 section 1 Multiprotocol Challange lab ISIS task
>>> 2.2
>>> >
>>> > Once you enable route leaking, in some cases the default will not be
>>> advertised anymore. I believe this has to do with the S train specifically.
>>> Try and remove the route-leaking and see if you then receive a default on
>>> R2.
>>> >
>>> >
>>> > --
>>> > Regards,
>>> >
>>> > Rick Mur
>>> > CCIE2 #21946 (R&S / Service Provider)
>>> > Sr. Support Engineer - IPexpert, Inc.
>>> > URL: http://www.IPexpert.com
>>> >
>>> > On 19 dec 2009, at 22:43, Thomas J. Loran wrote:
>>> >
>>> >
>>> >
>>> > In the Vol II  section 1 multiprotocol Challenge  A  task 2.2  we have
>>> to leak additional routes  into R2  so everything stays connected if one of
>>> the link goes down.  I understand that.   I know that I need /32 for routes
>>> later on.
>>> >
>>> > What I do not understand is why R1 and R5, the L1/L2 routers  do not
>>> leak a default route into R2.    I thought the L1/L2 routers were supposed
>>> to set the ATT   bit so R2 has a way out of the network
>>> >
>>> > R2#
>>> > IS-IS Level-1 Link State Database:
>>> > LSPID                 LSP Seq Num  LSP Checksum  LSP Holdtime
>>>  ATT/P/OL
>>> > R1.00-00              0x00000031   0x6E4B        1058
>>>  0/0/0
>>> > R2.00-00            * 0x00000032   0x4661        977
>>> 0/0/0
>>> > R5.00-00              0x00000035   0xAECD        980
>>> 0/0/0
>>> > R5.03-00              0x00000008   0x2B7D        960
>>> 0/0/0
>>> >
>>> >
>>> > R1#sh isi database
>>> >
>>> > IS-IS Level-1 Link State Database:
>>> > LSPID                 LSP Seq Num  LSP Checksum  LSP Holdtime
>>>  ATT/P/OL
>>> > R1.00-00            * 0x00000031   0x6E4B        891
>>> 0/0/0
>>> > R2.00-00              0x00000032   0x4661        806
>>> 0/0/0
>>> > R5.00-00              0x00000035   0xAECD        809
>>> 0/0/0
>>> > R5.03-00              0x00000008   0x2B7D        790
>>> 0/0/0
>>> > IS-IS Level-2 Link State Database:
>>> > LSPID                 LSP Seq Num  LSP Checksum  LSP Holdtime
>>>  ATT/P/OL
>>> > R1.00-00            * 0x0000000B   0x0EBF        886
>>> 0/0/0
>>> > R5.00-00              0x0000000B   0x1228        518
>>> 0/0/0
>>> > R1#
>>> >
>>> > R5#sh is database
>>> >
>>> > IS-IS Level-1 Link State Database:
>>> > LSPID                 LSP Seq Num  LSP Checksum  LSP Holdtime
>>>  ATT/P/OL
>>> > R1.00-00              0x00000031   0x6E4B        844
>>> 0/0/0
>>> > R2.00-00              0x00000032   0x4661        763
>>> 0/0/0
>>> > R5.00-00            * 0x00000035   0xAECD        769
>>> 0/0/0
>>> > R5.03-00            * 0x00000008   0x2B7D        750
>>> 0/0/0
>>> > IS-IS Level-2 Link State Database:
>>> > LSPID                 LSP Seq Num  LSP Checksum  LSP Holdtime
>>>  ATT/P/OL
>>> > R1.00-00              0x0000000B   0x0EBF        840
>>> 0/0/0
>>> > R5.00-00            * 0x0000000B   0x1228        476
>>> 0/0/0
>>> >
>>> >
>>> >
>>> > I do not see any ATT bit being set and L2 is definitely an L1 only
>>> router and R! and R5 are definitely L1/L2 routers.  In Doyles' Volume 1
>>> Routing book page 580 he discusses this and suggests turning on CLNS routing
>>> on both the interface and the routers (default)  I did that and am still not
>>> getting the ATT bit set.
>>> >
>>> > What am I doing wrong?  AM I wrong to look for the ATT bit / default
>>> route in this scenario?
>>> >
>>> > All three routers are 7200 running version 12.2(18)S2
>>> >
>>> > thanks
>>> >
>>> > Thomas Loran
>>> > CCIE  #7113
>>> >
>>> >
>>> >
>>> > -------------- next part --------------
>>> > An HTML attachment was scrubbed...
>>> > URL:
>>> http://onlinestudylist.com/pipermail/ccie_sp/attachments/20091220/46d8af68/attachment.htm
>>> >
>>> > End of CCIE_SP Digest, Vol 33, Issue 23
>>> > ***************************************
>>> > _______________________________________________
>>> > For more information regarding industry leading CCIE Lab training,
>>> please visit www.ipexpert.com
>>>
>>> _______________________________________________
>>> For more information regarding industry leading CCIE Lab training, please
>>> visit www.ipexpert.com
>>>
>>>
>>>
>>>
>>> --
>>> Bryan Bartik
>>> CCIE #23707 (R&S, SP), CCNP
>>> Sr. Support Engineer - IPexpert, Inc.
>>> URL: http://www.IPexpert.com
>>>
>>>
>>>
>>
>>
>>
>> --
>> Bryan Bartik
>> CCIE #23707 (R&S, SP), CCNP
>> Sr. Support Engineer - IPexpert, Inc.
>> URL: http://www.IPexpert.com
>>
>> _______________________________________________
>> For more information regarding industry leading CCIE Lab training, please
>> visit www.ipexpert.com
>>
>>
>


-- 
Bryan Bartik
CCIE #23707 (R&S, SP), CCNP
Sr. Support Engineer - IPexpert, Inc.
URL: http://www.IPexpert.com
_______________________________________________
For more information regarding industry leading CCIE Lab training, please visit 
www.ipexpert.com

Reply via email to