[cisco-voip] Call Routing (Loop)
Wes Sisk
wsisk at cisco.com
Wed Jul 1 13:44:35 EDT 2009
Agreed. this is proper design. Any incoming device should not use a CSS
that contains a partition which contains a router pattern that points
back to the same gateway. This will prevent call routing loops.
Graphically:
Incoming call via Gateway1 -> CSS -> Partition(s) -> Route Pattern ->
Route List -> Route Group -> Outgoing call via Gateway1
This is the most basic example. There are other examples involving call
forwarding and translations patterns. For those CM has built in
mechanisms to kill call routing loops. Only challenge here is the loop
has to get pretty bad before it can be detected and stopped.
It is possible to look for call routing loops using CDR query. It is a 2
step process:
1. identify times with high rate of calls per second. the
datetimeorigination field in cdr is in epoch time which conveniently has
granularity to the second so:
select top 100 count(*) as cps, orignodeid, datetimeorigination,
DATEADD(ss, datetimeorigination, 'Dec 31, 1969 19:00:00') AS dts_EST
from calldetailrecord
where datetimeorigination>1173272400
group by orignodeid, datetimeorigination
order by cps desc
You might choose to replace 1173272400 with a more recent epoch time.
Recall the CDRs are written using GMT timestamps.
Any cps value over ~15 should raise suspicion.
2. Once you have a specific datetimeorigination and orignodeid then:
select
origdevicename,destdevicename,callingpartynumber,finalcalledpartynumber
from calldetailrecord
where
datetimeorigination='<datetimeorigination>'
and orignodeid = <nodeid>
substitute <datetimeorigination> for the specific value of interest you
found in the first query. Substitute <nodeid> for the orignodeid from
the same row of the first query.
If the devicenames or numbers repeat 15 or more times in a second then a
very good chance of call routing loop. This also gives you the numbers
and devices that triggered the problem. You can investigate the
specifics of the situation to identify a more desirable behavior.
Regards,
Wes
On Tuesday, June 30, 2009 9:23:03 PM, lelio at uoguelph.ca wrote:
> Daniel has it spot on. The incoming calling search space of any
> trunk/gateway/gatekeeper/etc should not include any partition that
> contains any route pattern that would end up routing a call back to
> itself.
>
>
>
> Lelio Fulgenzi, Senior Analyst
> Computing & Communications
> University of Guelph
> 519-824-4120 x56354
>
> ...sent from my iPod - please pardon my fat fingers ;)
>
> [XKJ2000]
>
> On 2009-06-30, at 7:46 PM, Daniel <dan.voip at danofive.id.au
> <mailto:dan.voip at danofive.id.au>> wrote:
>
>>
>> We had a very similar issue to this, ended up being that CSS was
>> configured wrong on our trunk to our gatekeeper.
>>
>> A call would come in from our external gateway or an internal call,
>> the externsion could not be found so the call would match the catch
>> all route pattern and go out the trunk to the gatekeeper. The
>> gatekeeper would have the correct zone prefixes and send the call
>> back to the same zone it came from. The inbound CSS of the gatekeeper
>> trunk had the wrong CSS configred which allowed calls to go back out
>> to the tieline which created a loop as call manager set the call back
>> to the gatekeeper and the gatekeeper sent the calls back to the same
>> call manager.
>>
>> I would advise to make sure that your gateways and gatekeepers are
>> only allowed to call internal extensions only, or at least the
>> gatekeeper so that itself is only allowed to call internal extensions.
>>
>> The call will still hit the gatekeeper once but when the call comes
>> back in from the gatekeeper to the call manager the call is rejected,
>> the extension can not be found and there is no partition in the css
>> that allows access to the route pattern for the trunk.
>>
>>
>>
>>
>>
>> On Wed, Jul 1, 2009 at 7:59 AM, Scott Voll <svoll.voip at gmail.com
>> <mailto:svoll.voip at gmail.com>> wrote:
>>
>> I don't think I follow all of scenario 2 but as for #1 we had the
>> same thing with our old PBX and our CM. if the extension wasn't
>> on either system you would loop until all channels were in use or
>> someone disconnected. I think the only fix would be to have a RP
>> with all the extensions not in use forwarded to an AA. But every
>> time you move an extension over you will need to delete the RP
>> extension. If you were doing a trunk and not a channel based
>> circuit to the other PBX you have the ability to kill the system
>> because it will keep looping until the system can't take it any
>> more.
>>
>> Scott
>>
>> On Tue, Jun 30, 2009 at 12:47 PM, <steve.siltman at assurant.com
>> <mailto:steve.siltman at assurant.com>> wrote:
>>
>>
>> Here is a couple of scenarios that I need help with.
>>
>> Scenario 1:
>> Cisco IP Phones and Avaya phones mixed at the same location.
>> MGCP router is the go-between. If an extension is dialed
>> and doesn't live on one system, it forwards the call to the
>> other system. This works great. What if the extension
>> doesn't live on either system? I was taking a trace from
>> Call Manager, on a different issue, and noticed a problem
>> that looks like a call is doing the above. Without adding
>> specific route patterns and continuing to use a large 1XXXX
>> pattern to send calls across to Avaya when they don't exist
>> on Call Manager, can I limit this route loop or stop it from
>> happening somehow?
>>
>> Scenario 2:
>> This one has me perplexed because I'm not sure why I didn't
>> notice this long ago or how it continues to loop. The call
>> comes into a remote H.323 gateway and the DNIS is translated
>> into a DN that lives on Call Manager. The dial-peer looks up
>> the DN on Call Manager but it doesn't exist. *sigh* We have
>> a route pattern configured to point all those extensions
>> towards the remote H.323 gateway. I believe the first office
>> was setup this way and its been copied for each additional
>> remote office install. We now have 20 offices that have a
>> route pattern pointing the Internal DN range back out to the
>> remote H.323 gateway where the phones live physically. I
>> believe the resolution to this is to remove these internal DN
>> range route patterns. Call Manager already knows them and
>> doesn't need this route pattern. Correct? I still don't
>> understand how this could be looping but it must be looping
>> within the Call Managers. I turned on ISDN Q931 and VOIP
>> DIALPEER debugs on the router and saw the call come in and
>> hit the dial-peer to Call Manager. Thats all I saw on the
>> router but yet the Call Manager had 250 trace files, each 1
>> meg in size and rolled after 9 minutes. Digit analysis shows
>> the called number and the extension, that doesn't exist, over
>> and over and over in just under 1 second intervals. I'm
>> pretty sure removing this internal DN range route pattern
>> will resolve this but I'd like to know how its looping.
>>
>> Any suggestions or you've seen this before would be
>> appreciated. Thanks!
>>
>> Steve Siltman
>> Assurant Corporate Technology
>> steve.siltman at assurant.com <mailto:steve.siltman at assurant.com>
>> ------------------------------------------------------------------------
>> This e-mail message and all attachments transmitted with it
>> may contain legally privileged and/or confidential
>> information intended solely for the use of the addressee(s).
>> If the reader of this message is not the intended recipient,
>> you are hereby notified that any reading, dissemination,
>> distribution, copying, forwarding or other use of this
>> message or its attachments is strictly prohibited. If you
>> have received this message in error, please notify the sender
>> immediately and delete this message and all copies and
>> backups thereof.
>>
>> Thank you.
>>
>> ------------------------------------------------------------------------
>>
>> _______________________________________________
>> cisco-voip mailing list
>> cisco-voip at puck.nether.net <mailto:cisco-voip at puck.nether.net>
>> https://puck.nether.net/mailman/listinfo/cisco-voip
>>
>>
>>
>> _______________________________________________
>> cisco-voip mailing list
>> cisco-voip at puck.nether.net <mailto:cisco-voip at puck.nether.net>
>> https://puck.nether.net/mailman/listinfo/cisco-voip
>>
>>
>> _______________________________________________
>> cisco-voip mailing list
>> cisco-voip at puck.nether.net <mailto:cisco-voip at puck.nether.net>
>> https://puck.nether.net/mailman/listinfo/cisco-voip
> ------------------------------------------------------------------------
>
> _______________________________________________
> cisco-voip mailing list
> cisco-voip at puck.nether.net
> https://puck.nether.net/mailman/listinfo/cisco-voip
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://puck.nether.net/pipermail/cisco-voip/attachments/20090701/1b20b081/attachment.html>
More information about the cisco-voip
mailing list