I typically do the same thing when writing my rules as well.  I try and
write the rules so that a single rule would only apply to a single point
section rather than using the same rule across many different point
sections.  If you do have to make a tweak to a given rule, it's better to
only have that tweak affect one thing instead of 3+ things.

The called station ID format when coming from a WLC will be radio MAC
address + ":" + SSID name.  ie. 00-11-22-33-44-55:MySSID

So if you want to use a compound condition to match on the SSID, choose the
"ends with" matching option.

But to reiterate a previous point, we only get the SSID when the
authentication comes from a WLC, and not when it comes from an AP
(autonomous or HREAP).

Regards,



Jeff Rensink : Sr Instructor : iPexpert <http://www.ipexpert.com/>

CCIE # 24834 :: Wireless / R&S

:: World-Class Cisco Certification Training

Direct: +1.810.326.1444

:: Free Videos <http://www.youtube.com/ipexpertinc>

:: Free Training / Product Offerings <http://www.facebook.com/ipexpert>

:: CCIE Blog <http://blog.ipexpert.com/>
:: Twitter <http://www.twitter.com/ipexpert>


On Fri, Jan 3, 2014 at 11:39 AM, Kristján Ólafur Eðvarðsson <
[email protected]> wrote:

> Just a few cents on the ACS.
>
> I always try to be as specific when matching what I wanted. For example.
> Match WLC1 , match peap , match SSID and even user.
> Cause when complicating something in the future I would be more flexible.
> For example an extra requirement that I didnt forsee.
> We get those a lot in the real lab :)
>
> I always configure a default deny for the default network access. (not
> default permit) So I know that my configuration is really failing.
> When matching WLC SSID the compound condition with "ends-with" (or
> starts-with, I dont remember) was the trick since the WLC
> Sends the SSID and the radio MAC involved in one string to the ACS.
>
> Kristjan
>
> Today's Topics:
>
>    1. Re: WB1 Lab 3.2 (Jay Killion (jakillio))
>    2. Re: WB1 Lab 3.2 (Jeff Rensink)
>    3. Ang.:  WB1 Lab 3.2 (Andreas di Zazzo)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Fri, 3 Jan 2014 15:40:28 +0000
> From: "Jay Killion (jakillio)" <[email protected]>
> To: Justin Kurynny <[email protected]>
> Cc: "[email protected]"
>         <[email protected]>
> Subject: Re: [OSL | CCIE_Wireless] WB1 Lab 3.2
> Message-ID: <ceec355b.12742%[email protected]>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Thanks for the reply Justin.  Yes, I actually did exactly what you
> suggested and noticed that even when I created the end-station filter, I
> would never match on it and would just hit the default.  Going through the
> logs, I don't see the SSID (in this case, Test4-01) mentioned anywhere.
> I'm guessing that autonomous doesn't send the SSID in the request?
>
> I think I found the correct workaround, however.  Use Anyconnect NAM to
> create the network and lock-down the authentication method.
>
> Jay Killion, CCIE #17873 R/S
>
>
>
>
> On 1/3/14 12:11 AM, "Justin Kurynny" <[email protected]> wrote:
>
> >Jay,
> >
> >You're on the right track overall, but for that first step you may want
> >to take a close look at the ACS logs to see what attributes and
> >attribute values are included in the radius auth request packet coming
> from the AP.
> >
> >In a larger context, I found that a highly valuable exercise was to
> >compare the radius auth requests from the following three devices.
> >Their attributes differ depending on source and knowing those
> >differences is key when setting up access policies in ACS:
> >
> >* WLC
> >* Autonomous AP
> >* FlexConnect AP (standalone mode)
> >
> >hth,
> >Justin
> >
> >Disclaimer: I'm not familiar with the specific exercise you're working
> >on--just trying to help in a general sense on your outlined first step.
> >
> >typd on tny kybrd.
> >
> >> On Jan 2, 2014, at 19:42, "Jay Killion (jakillio)"
> >><[email protected]>
> >>wrote:
> >>
> >> Question on WB1 lab 3.2.  The requirements have you creating multiple
> >>SSID's (autonomous AP), each using different EAP methods in ACS ? such
> >>as PEAP for one and TLS for another.  The solution book doesn't show
> >>how this is done in ACS so I wanted to find out the correct method for
> this.
> >> I'm thinking the correct steps are as follows, but would appreciate
> >>any feedback.
> >>
> >> 1 ? Create an end-station filter to match on SSID
> >> 2 ? Create a new access service that only allows that specific EAP
> >>method (TLS, for example)
> >> 3 ? Create a new service selection rule that matches the end-station
> >>filter (from step 1) and returns the service created in step 2, thus
> >>only permitting that EAP method
> >>
> >> Is that correct?
> >>
> >> Thanks
> >>
> >> Jay Killion, CCIE #17873 R/S
> >> _______________________________________________
> >> Free CCIE R&S, Collaboration, Data Center, Wireless & Security Videos ::
> >>
> >> iPexpert on YouTube: www.youtube.com/ipexpertinc
>
>
>
> ------------------------------
>
> Message: 2
> Date: Fri, 3 Jan 2014 09:59:32 -0600
> From: Jeff Rensink <[email protected]>
> To: "Jay Killion (jakillio)" <[email protected]>
> Cc: "[email protected]"
>         <[email protected]>
> Subject: Re: [OSL | CCIE_Wireless] WB1 Lab 3.2
> Message-ID:
>         <CAG4_piWfFBr6j0E1CXnF20XAwZPFq=DzKrzG=
> [email protected]>
> Content-Type: text/plain; charset="windows-1252"
>
> The exercise was mainly geared at configuring the SSIDs.  There really
> isn't a specific ACS configuration that was supposed to go along with it.
>  In fact, there's no sane ACS config that would only allow the
> authentications as described on the APs.
>
> With autonomous APs, we don't get the SSID name in the called station ID
> field like we do when the authentications come from a WLC.  The same
> actually goes for H-REAP APs doing local authentication as well.  It's just
> the MAC address of the AP radio at that point.  So matching on SSIDs isn't
> really an option.  You may be able to do it with MBSSID mode and matching
> on the MAC address of the radio (different MAC per SSID at that point I
> believe), but that would be out of the realm of reality in my opinion.  The
> wireless lab doesn't really get into ridiculous scenarios like that.
>
> But let's say we were on WLCs, and we could match on SSIDs... While you
> could split this out into separate access services, I'd recommend just
> keeping it simple and doing it all under one.  So if SSID1 only allowed
> clients to use PEAP, just write 2 rules with the logic below...
>
> if SSID= SSID1 and EAP type = PEAP then Permit if SSID= SSID1 then Deny
>
> This would probably do the trick just fine.  Where you would absolutely
> have to split out into separate access services would be if they made
> reference that certain EAP authentications shouldn't even be allowed to be
> completed.  Then you'd need to push them over to an access service that
> only allowed the specified EAP type.  Here's why...
>
> In my first example, we'll pretend the access service allows all EAP types.
>  Say a client tries to authenticate with EAP-FAST.  The client would
> probably do a success ful authentication with EAP-FAST (correct
> username/password), but when it comes to the authorization phase, it would
> be denied based on the rules.  So EAP-FAST actually happened successfully,
> but the client wasn't allowed on the network.
>
> If you kicked the client over to a different access service that only
> allowed PEAP, then the client would try to use EAP-FAST but the server
> wouldn't go along with it.  So EAP-FAST would never have happened since the
> client and server wouldn't have agreed on an EAP type to use.
>
> So in the end, we always want to read very carefully and do exactly what
> is asked.  But when possible, try to keep things simple.  It's usually less
> to configure, less to verify, and less to troubleshoot.
>
>
> On Thu, Jan 2, 2014 at 9:35 PM, Jay Killion (jakillio)
> <[email protected]>wrote:
>
> >  Question on WB1 lab 3.2.  The requirements have you creating multiple
> > SSID's (autonomous AP), each using different EAP methods in ACS ? such
> > as PEAP for one and TLS for another.  The solution book doesn't show
> > how this is done in ACS so I wanted to find out the correct method for
> > this.  I'm thinking the correct steps are as follows, but would
> > appreciate any feedback.
> >
> >  1 ? Create an end-station filter to match on SSID
> > 2 ? Create a new access service that only allows that specific EAP
> > method (TLS, for example)
> > 3 ? Create a new service selection rule that matches the end-station
> > filter (from step 1) and returns the service created in step 2, thus
> > only permitting that EAP method
> >
> >  Is that correct?
> >
> >  Thanks
> >
> >  Jay Killion, CCIE #17873 R/S
> >
> > _______________________________________________
> > Free CCIE R&S, Collaboration, Data Center, Wireless & Security Videos ::
> >
> > iPexpert on YouTube: www.youtube.com/ipexpertinc
> >
> -------------- next part --------------
> An HTML attachment was scrubbed...
> URL:
> </archives/ccie_wireless/attachments/20140103/d1b0d92e/attachment-0001.html>
>
> ------------------------------
>
> Message: 3
> From: "Andreas di Zazzo" <[email protected]>
> To: "Jay Killion (jakillio)" <[email protected]>,      "Justin Kurynny"
>         <[email protected]>
> Cc: [email protected]
> Subject: [OSL | CCIE_Wireless] Ang.:  WB1 Lab 3.2
> Message-ID: <[email protected]>
> Content-Type: text/plain;       charset="utf-8"
>
>
> For the autonomous AP the called station ID contains the MAC address
> associated with the WLAN the client connected to. Since each WLAN contains
> a unique mac-address you can filter it this way, however one rule-condition
> per AP.
>
> Skickat fr?n min HTC
>
> ----- Reply message -----
> Fr?n: "Jay Killion (jakillio)" <[email protected]>
> Till: "Justin Kurynny" <[email protected]>
> Kopia: "[email protected]" <
> [email protected]>
> Rubrik: [OSL | CCIE_Wireless] WB1 Lab 3.2
> Datum: fre, jan 3, 2014 16:58
>
>
> Thanks for the reply Justin.  Yes, I actually did exactly what you
> suggested and noticed that even when I created the end-station filter, I
> would never match on it and would just hit the default.  Going through the
> logs, I don't see the SSID (in this case, Test4-01) mentioned anywhere.
> I'm guessing that autonomous doesn't send the SSID in the request?
>
> I think I found the correct workaround, however.  Use Anyconnect NAM to
> create the network and lock-down the authentication method.
>
> Jay Killion, CCIE #17873 R/S
>
>
>
>
> On 1/3/14 12:11 AM, "Justin Kurynny" <[email protected]> wrote:
>
> >Jay,
> >
> >You're on the right track overall, but for that first step you may want
> >to take a close look at the ACS logs to see what attributes and
> >attribute values are included in the radius auth request packet coming
> from the AP.
> >
> >In a larger context, I found that a highly valuable exercise was to
> >compare the radius auth requests from the following three devices.
> >Their attributes differ depending on source and knowing those
> >differences is key when setting up access policies in ACS:
> >
> >* WLC
> >* Autonomous AP
> >* FlexConnect AP (standalone mode)
> >
> >hth,
> >Justin
> >
> >Disclaimer: I'm not familiar with the specific exercise you're working
> >on--just trying to help in a general sense on your outlined first step.
> >
> >typd on tny kybrd.
> >
> >> On Jan 2, 2014, at 19:42, "Jay Killion (jakillio)"
> >><[email protected]>
> >>wrote:
> >>
> >> Question on WB1 lab 3.2.  The requirements have you creating multiple
> >>SSID's (autonomous AP), each using different EAP methods in ACS ? such
> >>as PEAP for one and TLS for another.  The solution book doesn't show
> >>how this is done in ACS so I wanted to find out the correct method for
> this.
> >> I'm thinking the correct steps are as follows, but would appreciate
> >>any feedback.
> >>
> >> 1 ? Create an end-station filter to match on SSID
> >> 2 ? Create a new access service that only allows that specific EAP
> >>method (TLS, for example)
> >> 3 ? Create a new service selection rule that matches the end-station
> >>filter (from step 1) and returns the service created in step 2, thus
> >>only permitting that EAP method
> >>
> >> Is that correct?
> >>
> >> Thanks
> >>
> >> Jay Killion, CCIE #17873 R/S
> >> _______________________________________________
> >> Free CCIE R&S, Collaboration, Data Center, Wireless & Security Videos ::
> >>
> >> iPexpert on YouTube: www.youtube.com/ipexpertinc
>
> _______________________________________________
> Free CCIE R&S, Collaboration, Data Center, Wireless & Security Videos ::
>
> iPexpert on YouTube: www.youtube.com/ipexpertinc
>
>
>
> ------------------------------
>
> _______________________________________________
> Free CCIE R&S, Collaboration, Data Center, Wireless & Security Videos ::
>
> iPexpert on YouTube: www.youtube.com/ipexpertinc
>
> End of CCIE_Wireless Digest, Vol 57, Issue 6
> ********************************************
> _______________________________________________
> Free CCIE R&S, Collaboration, Data Center, Wireless & Security Videos ::
>
> iPexpert on YouTube: www.youtube.com/ipexpertinc
>
_______________________________________________
Free CCIE R&S, Collaboration, Data Center, Wireless & Security Videos ::

iPexpert on YouTube: www.youtube.com/ipexpertinc

Reply via email to