Monthly Archives: September 2011

It’s not me it’s you…

One of the niftiest features Call Manager offers is Single Number Reach (SNR). Configuring SNR takes a wee bit of work, but once it’s setup you’ll never love your life more. Why? Because SNR ends the tradition of giving out a work number *and* a cell phone number just so customers, end users, and sales piranas can abuse you all hours of the day. With SNR, you can send your work calls to your cell phone on your terms, which pretty much rocks my socks.

Recently I was troubleshooting this feature of awesomeness trying to determine why a user’s mobile phone wasn’t ringing when the work number was called. Few obvious things to look for when troubleshooting this:

  • Make sure there’s a Remote Destination Profile AND a Remote Destination configured – I know they sound like the same thing, but they are not and you’ll need both
  • Make sure the Calling Search Space on the Remote Destination Profile permits a call to the cell phone- meaning if it’s a long distance call to the cell make sure your chosen CSS allows long distance
  • Make sure that the user is associated with his/her Remote Destination Profile
  • Last but not least, tweak the timer settings- Answer Too Soon, Answer Too Late, Delay Before Ringing, and Just Hang Up the Damn Call Now are timers that are all extremely helpful when customizing this feature for Mr or Mrs Picky Pants. (Yes, I made up that last one, but it sounds like an option that should be there)

So, what if you’ve got it all configured correctly and the user’s mobile phone still doesn’t ring? Time for a sanity check – if you’re me, you put your own cell phone number in the Remote Destination and give it a whirl.  Now this may sound less like sanity and more like madness, but it means you’ve injected a known working quantity into the equation.  If there’s one thing I know my cell phone does, it’s ring. All. the. time.

But back to the story. Given the new parameter, SNR works like a charm.  So why did SNR like my phone so much better than the user’s? Mine does have a pretty cool case, I’ll give you that. However, doubting that this was the magic factor, I did some research and dug up some interesting information. I found that her model of phone had a TON of complaints posted online.  The complaints: that model – an EVO on the Sprint network- apparently has a nasty habit of not ringing when people call.  Fortunately, there were lots of suggestions on how to resolve this particular issue and my customer’s phone now rings all the time. Especially since I forwarded my work number there.

The moral of the story – voice is a murky, complex world in which we engineers often find ourselves trying to manipulate devices that lay just outside of our influence or control. Whether it’s an ancient fax machine located half way across the country with ECM turned on, a cell phone running a buggy version of code and a bad PRL list, or perhaps even another vendor’s video endpoint that refuses to make nice with the expensive equipment on your end of the conversation- we’ve got to hone our skills to narrow down the issue and face the fact that sometimes the fix will be out of our hands.

In case you’ve got an EVO phone with a ringing disorder, check out this forum:

And if you need to configure SNR from scratch, I’d recommend starting here:

And an official Cisco doc, complete with judgmental dude looking down at you while you read, can be found here:


Tags: , , ,

Voice Girl Goes to Storage Day

Who has two thumbs and got to attend the last Tech Field Day?  This girl!

In case you don’t know what Tech Field Day is, go here and check it out:   In case you don’t care what Tech Field Day is, I suggest you stop reading or make sure you have copious amounts of alcohol handy.  Actually, that last suggestion could improve the reading of any of my posts, so feel free to get started, you have my blessing.

Now, I’m sure we’ve all had that friend who goes on a vacation and brings back 10,000 pictures and insists on narrating them all in great, painstaking detail.  Fear not – I want to smack that guy as much as you do – so I will just be hitting the highlights of this expedition in this post.

So, without further ado, awesome thing number 1: hanging out with server admins.  I know, I know, for network and/or voice guys this hardly sounds like something that would make the list of awesome- unless that list were titled Ways In Which My Day Could Awesomely Suck – but it’s true and let me tell you why.

With roles in IT becoming less and less siloed, it’s clear us folks guarding the layer 2 and 3 keys to the castle are going to have to make nice with those folks rocking the upper layer data center knowledge.  As distasteful as that may initially sound to both parties involved, we all earn huge rewards.

Think about it- do you really want that server guy vMotioning all those production boxes across your precious WAN without any clue as to the implications?  I’m certain that server guy with the ponytail doesn’t want us well-intentioned network junkies screwing with SAN infrastructure when he/she thinks we don’t even know what random IO is. Of course do we do know what it is, but not the point…

Contrary to popular sysadmin belief, we network folks are capable of reading and do in fact know what a manual looks like.  Contrary to network admin belief, server guys do know what they are doing and don’t just break crap on purpose.  Given shrinking IT budgets, device consolidation, and technology overlap, our tiny sandbox has only gotten tinier and now it looks like we’re going to have to share the dump truck and not just the buckets.  (the dump truck was always my favorite)

So awesome thing number two:  presentations! Companies solving problems I was vaguely aware existed in ways I only wish I had imaged because retirement would be nice about now.  The quality of presentations was generally high and the technical level generally deep.  Perfect combination.

Let me offer a few brief take-aways from what I saw, you can catch the presentations here

  • Nutanix: Putting your VMs and storage on the same devices, have them utilize the same resources.  It has a kind of eggs in one basket feel – but the basket is really nice.  Interesting implications on the necessity for SAN administrator.
  • Nasuni: If you ever want tips on how to deliver a presentation, watch this one. The send-your-files-in-the-cloud-and-see-them-at-your-other-sites product was wicked cool. Matt Simmons had the product up and running during the time of the demo. Sweet.
  • Symantec Storage Foundation 6.0: Least favorite presentation style. So. many. power. point. slides. Clearly this product has some significant improvements over the previous version but the demo certainly wasn’t showing off this products nice curves, so to speak.
  • Data Direct Web Object Scaler: large-scale cloud storage wow-ness.  Keeping track of your massive amounts of cloud data using custom filing system to store and replicate data. Demo was super neat, product super fast.
  • Pure Storage- all SSD storage, forget tiering.  They wrote their own software to talk/write to SSD drives in a way that makes SSD drives very happy. In fact, drives never fail for Pure Storage, or so was said- a concept our little group of skeptics had some trouble with. Pure Storage held to their guns though and a promise was made to tweet the first drive failure.
  • Arista EOS:  Command line goodness. In the demo, the guy added the XMPP package to the Linux-based software running the switch, then chatted with the switch. Totally cool. Who doesn’t want to ask a switch how it’s day is going?
  • SolidFire- All SSD storage, optimized for providers who want to limit compute and/or storage on a per customer basis. If you are a cloud provider of storage, being able to establish very specific SLAs for customers I’m sure is extremely appealing.
  • Arekia- backup goodness.  Presentation went into detail on their particular brand of deduplication which provides quite a lot of benefit when backing up large amounts of data.

Last but not least, awesome thing number three: Stephen Foskett and Matt Simmons are freaking fantastic!  As the organizers, they coordinated every intricate detail and then made it look easy to the rest of us.  A very special thanks to those guys for making all of this happen, wishing them happy times in therapy as they attempt to recover.

For links to all things Tech Field Day 8:


Posted by on 2011/09/19 in Tech Field Day, Uncategorized



Just want to be heard…

One of the things call centers supervisors really like to do is listen in on agent calls.  I’m sure it’s not *just* because they are nosy-type people, [insert business justification here], so part of my job is to make sure their eavesdropping is configured and working properly in Cisco Contact Center Express (UCCX).

Now there are about 11 ways to Sunday monitoring and recording can be jacked up by various elements, not the least of these being the voice engineer at the configuration helm. So when, during a deployment, it was found that calls were not able to be monitored or recorded, I skipped right past the look of surprise and moved straight into the what-is-it-this-time expression.

First, the symptoms.  Agents were getting calls and their supervisors were recording these calls. This means a whole bunch of agent/supervisor/phone setup tasks were completed correctly. Plus one point to the competent voice engineer with the mad skills. The recorded files were then being played back, however, and the tracks contained no audio.  Minus one point to the slightly less competent voice engineer who may, in fact, just be mad.

This not being my first rodeo, I initially suspect a codec issue, quickly confirmed by using the question mark button on the phones when the calls are made.  The display on the phone shows me the codec the calls are using is g.722 which, while a lovely codec, is not actually supported by UCCX.  It having been a long day, I decide to take a hatchet to g.722 and disable it in the Call Manager system wide parameters – ensuring no more g.722 EVER. Or at least not in this cluster.

Fully expecting the new rounds of tests to be successful, I get to use my surprised look after all when, once again, the recorded tracks lack audio. Grrr.

Firing up trusty Wireshark shows something very interesting – there is no RTP traffic from the PC to the UCCX server. For those who don’t eat, sleep, breath voice, RTP is the transport for the audio portion of the call.  All the setup/control messages will generally use SCCP, SIP, or H323, but the packetized voice uses RTP over UDP. The fact that it is completely absent from my capture file is more than a bit disconcerting.

After a nice talk with my buddies at TAC, they inform me that this is commonly seen with the particular brand of antivirus being run on the client workstations. After uninstalling the antivirus product and running the capture file again, RTP packets make an appearance and victory is declared in my favor.  100 points to the cheeky voice engineer from Dallas.

In case you were wondering what this RTP traffic looks like in Wireshark, you are looking for something like this:

RTP Traffic

As an added bonus for making it to the end of this post, here are a few other things you should check on the phone device configuration page in Call Manager when having issues with recording and monitoring, they are pulled from this document: Cisco CAD Troubleshooting Guide CAD 8.0 for Cisco Unified Contact Center Express Release 8.0 Cisco Unified Communications Manager Edition revised April 2011

  • PC Port—Enabled. If the PC Port is not enabled, the agent PC that is connected to the port will not have network access. No voice streams will be seen by the desktop monitor module
  • PC Voice VLAN Access—Enabled. If the PC Voice VLAN Access is not enabled, no voice streams will be seen by the desktop if the desktop is not a member of the same VLAN as the phone.
  • Span to PC Port—Enabled. If the Span to PC Port is not enabled, the voice streams seen by the phone will not be seen by the desktop monitor module.


Posted by on 2011/09/09 in UCCX


Tags: , , ,