More on ShoreTel V switches!

When installing a Shoregear switch, one of the parameters that you enable is to indicate which server is to manage this switch!    Normally this is no big deal, but with the V switch, you need to pay attention to this parameter as it will impact your Automated Attendant and Hunt Groups.

Let us assume we have a V switch installed in Dallas, Texas which is in the CST zone.    If this switch is managed by a server in the PST zone, then we need to adjust the schedules.    When using a V switch in another time zone, which would normally not be managed by a local DVM,  the site will need separate schedules applied to hunt groups and auto attendants.   The hunt group schedule will need time expressed in the time zone of the HQ server (example: site is in CST but HQ is in PST, 8-5 CST would then be entered as 6-3 PST on the schedule for hunt groups).   The auto attendant schedule will use the time zone of the application server (the local V switch) so the time can be entered in the local time zone.

The first time you set this up it will undoubtedly bite you!  So be aware of this configuration issue and save yourself a service call from the impacted User Group!

ShoreTel Contact Center Overflow Concepts and Direct Calls to Agent!

I would like to kill call center challenges with one blog!   In many call center environments, it is possible that an agent has a Direct in Dial (e.g. DID) number that a client might call and bypass the entire call center process!   For a call center manager, this is very frustrating!  You create a Contact Center to organize the flow of calls to your Technical Assistance Center, and clients by pass the process by calling your Agents/technicians directly!

Clearly, you can eliminate this by not giving DID numbers to Agents, but we end up doing this to facilitate an orderly problem resolution strategy.  You might give a client a “homework” assignment and ask them to call you back.   They might object to being put at the back of the queue again, so you give them a DID number that gets them directly back to you.  The challenge is, that this call would not be accounted for in your contact center reporting!   So how then do you provide this feature and also create a mechanism for enabling your contact center to capture all calls for Agent Performance reporting?  One answer is to establish a “service” that has only one Agent in it!   Then build out a DNIS/DID route point to IRN relationship that brings that DID number into the call center and routes it to the Agent.

Actually, it is not a bad strategy!   The Contact Center can now account for all calls, even the ones that reach your Agent through a DID number and you can apply normal Contact Center routing tools like “overflow”, which brings me to my second point!   Is it possible to overflow calls to more than one other group?  Based on more than one overflow time in queue?

The Contact Center provides a wide variety of methods for handling callers in queue awaiting service.   One of the more interesting concepts is the ability to “overflow” a caller from one group to another group based on the amount of time they have been waiting in queue.   This contact center parameter is set within the “service” and can be found in within the tab labeled “Overflow”.

In figure One below, you can clearly see that we have a number of services defined, including a service named TAC1.   Let us assume that this is a technical service group and that TAC1 is comprised of Agents/technicians that include Agent Gandalf, Kipling, Regan and Jack.      You can also see that a service has been enabled by the name “Direct Kipling”.    This service was created to enable callers to Kipling’s DID number to be brought into the Contact Center, and routed to Kipling even though they did not enter through the TAC1 service.   Hopefully, we can now count the phone calls Agent Kipling is handling that otherwise would not be reportable!

ecc_figure_1

In Figure One we can also see that we have set an “Overflow” counter that,  should anyone be in the Kipling Queue for 10 seconds they will overflow to the TAC1 queue.  What is interesting is that when you overflow in the ShoreTel Contact Center,  you don’t leave the queue you are in, you basically add another queue containing agents that have a cumulative effect on the call.    Should that “overflow” not result in an available agent and the caller continues to wait for service, you can actually set a second timer that would “overflow” (e.g. expand the number of agents ) to yet another queue.

In Figure two, you can see the original call in the Queue for Kipling.  After the pre-configured overflow interval is met, the call is distributed to the “overflow” queue but is also available to the original queue.  You can see this in Figure Three, by noting that the call is now in queue in both the Kipling and TAC1 queue.   In this way, if an agent becomes available, in either the original queue or the “overflow” queue, the call will be answered.   Had we set up a second interval timer in the Service, we could expand the number of Agents to include the additional group specified by that timer.  This is one of the more interesting, if not misunderstood capabilities of the ShoreTel “overflow” concept!

ecc-figure-21ecc_figure-31

Putty Telnet into ShoreGear SG50V/SG90V

We had a lot of requests for examples of telneting into the SG Voice switches, so we though we would do a quick video on that process. The V switch is a very different animal from the rest of the SG family. Basically, a small Linux server lives inside the switch and contains a Flash memory card to hold Automated Attendant and Voice Mail for users assigned to that “server”. There are some characteristics about the V switch that you should know and we have tried to summarize them below. One of the most important characteristics, is that a V switch MUST have a time source or bad things will happen!

The entire telnet process into this switch is different. You can find your way down to the subdirectory that contains the famous IPBXCTL security shell and run it using the normal command sequence. You will see the “telnet enabled” message, but you will pull you hair out before you can telnet into the V switch. To get inside the switch, you will need to use an SSH client like Putty. You will also need to log in as either the Admin or the Root users. The Admin gets you to a safe menu configuration option and you should have no problem configuring the switch. Remember, you have to have that NTS in your configuration! Log in as the root user however, gets you a more powerful set of trouble shooting command and enable you to back up the Flash card, and to some other debugs that you can not do from the admin user.

The video clip below walks you through the process. Here are some of the highlights we have discovered working with the V switch,

· Limited media stream capability.

· 1U Half Width Switch

· Voice Mail stored as 8-bit WAV files from G.711/G.729

· Switch can negotiate ADPCM but can Not proxy G.729

· Switch runs under Linux (not VxWorks)

· Voicemail Switch uses Qmail not SMTP and does not support SMDI

· SG90V supports 90 voice mail boxes

· 1GB Compact Flash stores about 1500 minutes or 15 minutes per user

· Full CF Card = “voice mailbox full” message

· Holds all Automated Attendant Messages?

· Store prompts for up to four languages

· At boot, requires connectivity to HQ server for configuration

· Once operational, does not need server connectivity for VM/AA

· V switch MUST have NTS SMTP and FTP resource

· Console port = Linux Shell STCLI is default

ShoreTel SIP Trunk Configuration Parameters for Level 3

Generally, we do not fill out forms given to us by Carriers.   Most Carrier forms use nomenclature that is “Marketing Speak” and not technically driven.   It seems we have 100 different ways to refer to a TDM with PRI signaling and only one of them means anything to an engineer.   There are a variety of new delivery methods being offered to clients for PRI service and it is important to understand exactly what type of circuit you are actually being given.    A PRI may in fact be a SIP trunk from a Softswitch, through a CISCO IAD with a TDM interface for your SGT1.   Not necessarily a bad thing (though you should expect problems with Fax service) but it is a hybrid circuit and not a traditional TDM T1 to a circuit switched central office.   We are always eager to assist the client by speaking with the carrier and having a dialog as to the exact nature of the circuit, but we steer clear of filling out the forms they expect their new customers to complete!

Recently, while planning the implementation of a ShoreTel IPBX with 230 SIP trunks, I  had the experience of pulling together answers to L3 request for configuration data.  With the help of  Rod Davis, Juan Rubio and Steve Weinstock we were able to respond.   To my surprise, this time there was not a single marketing expression in the document!   The questions were specifically engineering oriented and designed to understand the capabilities of the IPBX that the SIP trunk service was being ordered to support.    I though it would be interesting to share that information with our readers.  The IPBX is the ShoreTel and the assumption made in these answers is that the ShoreTel is being front-ended by an Ingate Separator.   Here are the SIP configuration questions and answers:

level31

How to set up Backup for ShoreTel VG50 and VG90 V siwtches!

Both models of the ShoreTel VG90 and VG50 are availabe in V switch configurations.   The V switch configuration enables you to provide local voice mail and autoamted attendant functionality to a smalll site economically.  Compared to the cost of adding a distributed voice mail server, the SG50V and SG90V are economically viable alaternatives that should be considered as you plan for resilency and site continuity during a WAN outage.

Due to the CF card capacity, daily backup of  voice mail and automated-attendnat is advised.  When automatic backup is enabled in the ShoreWare Director, it begins immediately after the server completes its daily house-keeping operations.  Automatic Backup Stores Voicemail, Auot Attendant Data and switch  log fiels to an FTP server.  After completion of the daily file-system cleanup taks, the switch begings automatic backup.  A time stamp is appended to the name of the files copied to the target server.

Automatic backup provides a source for the most recent days voice mail and the other data in the event of a system failue.  It is not intended to be an archive of voice messages or a sourc fo retrieving deleted voice mail.  The following silent film clip walks you through the process of setting up an FTP site and properly configuring the ShoreWare Director for V switch backup!

The missing ShoreTel Beep!

In both ShoreTel Release 8 and 9 there is a very interesting behavior that will ultimately bring a client call to your support center.   The behavior is most obvious in a multi-site deployment in which there is only one HQ server providing voice mail services.    A call to a user at a remote site is RNA (i.e. ring no answer) transferred to the voice mail system.   The call entered the system on a PRI at the remote site and is then transferred across an MPLS WAN to the HQ server.   The caller hears the greeting: “I am not here, at the beep please leave a message”.   The caller dutifully waits for the “beep”, but it never comes.  What happens to it?

The fact of the matter is the “beep” was in fact played, but the caller did not hear it.   This has to do with the use of the G729 Codec on Inter site calls in this scenario.   The G729 is distorted to the point that the caller can not hear the beep.   The distortion appears to only be related to the implementation of the G.729 codec on the virtual soft switch.  As a troubleshooting step, I used a different phone system and an IP phone that was set to only use G.729.  When we called the HQ virtual soft switch over a local PRI trunk where ULAW was being used from the PRI switch to the HQ server, we heard the voicemail beep without distortion (through the G.729 codec as implemented on a different PBX).  It is only distorted when the virtual soft switch is trying to do G.729.

Outlined below are two WAV files of the normal beep (ULAW) and the distorted beep (Softswitch G.729).  We have also attached a visual representation of the audio files.  The distortion is audibly and visibly clear.  Our thinking is the “beep” is to short and to high a frequency, but we are not developers!  Currently, the only known fix for this is to change the Codec to G711.  That fixes the issue, but it may have a negative impact on your WAN plan!  I am sure there is a fix in the works back at the ShoreTel Mother ship, but we have not seen it yet.

waveforms

ShoreTel “Call Screening”

Call Screening is a popular telephone feature though some people consider it both annoying and insulting.    In ShoreTel 9 you have this option using “find me follow me” to ask the caller something equivalent to “who may I say is calling”?   The caller then speaks their name or wisecrack, they are asked to “please hold while I locate your party” and then you are called on your specified” find me follow me” location.   When your phone rings and you hear the name of the individual that is calling.  You must explicitly accept the call by pressing the digit 1, or the call will be taken back and the caller will hear:  “ I am sorry I could not reach your party at the tone please leave a message”.

Actually, you can use this feature in the office if you are prepared to use two extensions on your ShoreTel phone (read two user licenses).   Set up the first extension (e.g. 114) with a call handling mode that has the following options checked in “find me follow me”.  First, check the box “Enable record callers name for Find me” and also “Enable Find me for incoming calls before playing Greeting”.   Set your First Number Find Me location to your second extension (e.g. 115).  When a call is received by your first extension, it will automatically prompt the caller with ‘the person you are calling requests your name”.  The System will then put them on hold and call your second extension, play the name and give you the option to accept or reject the call.  It is actually quite clever and it works!

9.1 Find Me Option for "Call Screening"
9.1 Find Me Option for “Call Screening”

Route Points, IRN’s and Media Streams!

Lately the subject of ShoreTel Route Points and IRN’s has been getting a lot of “mind share” from those working with them.   Two questions:  How many Route Points can a ShoreTel server support?  How many IRN’s can an ECC support?  How many DNIS numbers can a ShoreTel support.   Seems like a simple set of questions, but the answers are something like “herding cats”.    Route Points are used in a ShoreTel system to accomplish call routing and TAPI connectivity between the IPBX and the Enterprise Contact Center, for example.   Let us assume, for purpose of illustration, that you have a large call center and you want to track the effectiveness of your advertising campaigns.   You determine to do this by tracking DNIS numbers.  Each DNIS represents a different advertising campaign.  Recently, I came across a client who was creating a route point – IRN pair for each DNIS number.   Nothing particularly wrong with this, but using the DNIS map would have required only one Route Point if the DNIs was pointing to the same Group of Agents (e.g. Service) in the Contact Center.

The Contact Center, currently can only map one DNIS number per IRN.  It is my understanding that ShoreTel may be moving toward 1000-1500 DNIS numbers in future ECC releases.  So if you want to run report about DNIS you might have to look in two databases: (a) the ShoreTel MySQL CDR database on the IPBX; and (b) the C2G database on the ECC.  You will need to use the GUID if you want to tie the two databases together to get the DNIS and Agent Events and Wrap codes.   If you bring the DNIS into the IRN,  to enable DNIS reporting in the ECC, you can see the list of IRN’s could get very long and there is a  limit.

I suspect that DNIS has a limitation as well.  My thinking is that DNIS numbers would live in the various SG switches.  So the ultimate answer to how many DNIS numbers you can have may be memory constraint and  configuration specific (e.g. different for each configuration).    If you have any input on this, I would welcome it!   At the end of the day what we know is that you can create Route Points all day long in ShoreTel (though I have not attempted to find the real limit).  If the Route Points live on the ShoreTel server the real limitation is the 254 simultaneous media streams that a Microsoft Server can support.   If you point 12 PRI’s at one server and you have the potential of having 276 simultaneous phone calls, bad things are going to happen.   You need to spread your media stream over multiple servers.   In the contact center, you generally shape media streams (e.g. IVR ports) to 150 per server.

The “devil is in the details”.

What is all the Hub about?

So you can’t get any help from the IT department getting a private printer connected to your office PC.  So what do you do?  You run down to your local favorite “we sell cheap computer stuff” store and buy a “hub” or “usb multiplier” or something that looks like a small “Ethernet Switch”.  You take it back to the office, and having send the quick start guide, you unplug your computer from the wall jack, plug it into your shiny new thing and run a cable from it to the wall jack.  Then you plug in your new printer!  Maybe you go really geek and add in your own scanner and wireless access port!

Enter VoIP phone deployment.  Anyone that has done a VoIP deployment knows or has learned the hard way, that “hubs” will kill a VoIP phone deployment!   Multiple devices in a hub will immediately change your ethernet port to half-duplex.   If you put the VoIP phone behind the hub, you are either using a local power brick or you will not have access to the POE on the home ethernet switch.   Running around a 400 desktop installation trying to reconfigure the wiring so that the wall jack feeds the phone and the hub goes into the bottom of the phone is one option. That is always a time sink and usually outside the statement of work (SOW) of any knowledgeable systems integrator.

Hubs need to go the way of “buggy wips”.   They mark a network as unmanageable and general indicate that there is no professional network administration being conducted at that location!   Personally, hubs should have the same warning label the government wants you to put on cigarette packages!   Hubs and toy switches may be ok for you home network, but in this day and age they have no place in an enterprise deployment and when it comes to VoIP deployments, they just don’t make sense and they don’t work!

Preview ShoreTel ECC 5.1 Agent Tool Bar!

ShoreTel has now release the Enterprise Contact Center 5.1 for Beta users.  We are just now playin with it, but it has quite a few hot new features and even a new “look and feel” as it relates to icons.    For those of you who have installed the ECC, one of the more time consuming aspects of this deployment is the need to visit each desk to install the Agent Toolbar software.   Well, your prayers have been answered! An MSI is now available!  This will make upgrading, even a maintenance release, just that much more fun!   The 5.1 version has some other exciting capabilites that we will explore in detail in later blogs and film clips.  One of these is the ability to generate Real Time Adherance capability, a workforce management function that many call centers are demanding.  There are some new tabs and configuration options and as I said, there is a new look and feel.    Again, we will take a closer look at these features as we get more Beta experience but for now, lets just take a look at the Agent Tool Bar!