T minus 20 minutes
It’s the big day. You’re in the lobby waiting for the proctor to relieve you of the agony of waiting. Yes. The waiting is almost over. It’s judgment day. What happens in the next 8 or so hours could have an irreversible effect on your life- for the better I hasten to add. Your heart is racing and you catch a glance at the others who signed up for the same ordeal. They are talking, they are laughing, they make reference to the other CCIE certs they already possess, they make reference to painful experiences on previous attempts at the CCIE Voice lab exam. They seem relaxed and unfazed. But for you it’s different. You need this. You want your life back. This is hopefully the one and only time you will EVER sit the Voice lab.
Handle the pressure
The Proctor is satisfied everybody is at the correct desk and the housekeeping routine is over. Time starts now. It’s crunch time. You’ve shed plenty of blood, sweat and tears, been through countless practice labs over too many weekends to remember, read way too many documents and attended the bootcamp. But nothing you have done has simulated the pressure cooker you find yourself in. You know the ability to handle the pressure and think rationally is going to be critical if you are to have a successful day. If you can’t handle the pressure then you don’t have a chance. This is where second/third timers have the advantage over you.
How to take your nemesis down
You open up the spiral bound folder that represents your nemesis for the day. There are just so many pages and it feels more like a novel than an 8 hour lab. Good time management is vital! Every minute is going to count. But what does it mean? What does good time management really mean? It means that you try and read and interpret every word that is contained in the lab as quickly as possible without re-reading tasks over and over again and then translating that into configuration with a high degree of efficiency. Configure with a high degree of efficiency does not mean navigating to each part the UCM GUI one time and attempting to “group-configure” absolutely everything in one go. The time taken to read the entire lab with microscopic lenses, learn and appreciate all the requirements and then configure all entities in order to meet requirements IS GREATER THAN reading and configuring the lab section by section.
Don’t fall into the trap of spending too much time at the beginning of the lab analyzing requirements. You only have a handful of phones on your desk and if you need to return to the “device” pages for the phones 3 or 4 times then that’s ok. If Site A Phone 1 ends up being an ICD Agent phone, with a BLF Speed Dial with a new softkey template and the ability to be used for Extension Mobility then these are all things you don’t need to know upfront. You can return to all these settings later on. Right now you don’t even know the IP Address of the servers. You don’t even know if SiteA is a UCM or CME site. Why are you concerned with advanced features and applications given these limitations?
The theme of your strategy is simple. Walk before you can run. Take a modular approach. Hopefully you will be more accurate too since some requirements can only be truly appreciated by doing and not simply by reading.
“Speed Read” the lab
Spend 10 minutes glancing over the lab capturing a few high-level requirements. Which sites are UCM and which sites are CME? Which gateways are H323 and which are MGCP? Read the QoS section because you should be doing this earlier rather than later. Do you have MLP or FRF.12? Can I run AutoQos or not? Is your CUE module going to register with UCM or CME? You need to get this in a steady state earlier rather than later.
That’s enough reading. No point going into detailed information in Voicemail or Call Routing or Contact Center. You will only have to read it again. And you are going to forget what was contained in those sections. Remember what good time management means.
Notepad is your friend
Spend 15-20 minutes capturing all the information you are going to need repeatedly throughout the day in notepad. This information is dispersed over several pages and it’s going to be awkward visiting page 4 (which is the reverse side of page 3 which means turning over the leaf of paper) when you are on page 27. Have a section in notepad for each site or even a separate notepad file for each site. Do you need to know that SiteA Phone 1 is an ICD agent phone, has Extension Mobility enabled or has a BLF Speed dial on multiple occasions? No! You just need that information one time- at the time of configuration. You are not responding to an RFP and you don’t have thousands of phones to “pre-configure”. Walk before you can run.
Type into the notepad the IP Address of your servers. Take a note of usernames/password if they are unfamiliar to you. You will need the IP Addresses of your UCM servers over and over again. “session target ipv4:”, “mgcp call-agent”, “sccp ccm”, “option 150″, in CCX/CUPS/CUC/CUE, etc, etc. Mess it up one time and it will be costly. And for heavens sake whatever you do, type it in correctly in notepad because you are going to copy and paste this critical bit of information a hundred times.
Take a note of all the gateway ip addresses that you are going to be using- loopback interfaces and LAN interfaces. You will use this when defining gatekeepers, gateways, SRST references, SIP Trunks, etc, etc. Again- copy/paste them into the configuration.
Knowing and loving the full E164 numbers used at every site is imperative. External Number Mask. Voice Translation Profile. Ephone description. Route Pattern. Translation Pattern. Calling / Called Transformation Pattern. You could need it for all the above. So get it down in notepad and don’t make a mistake. Copy / Paste as and when you need it.
Leave the off-piste skiing for another day
30 minutes has elapsed and your neighbor is configuring voicemail and is “taking a moment to personalize his greeting”. Ignore him. Focus. Don’t get distracted. Stick to your strategy, don’t veer off-piste.
In saying that- you had better start configuring something soon. Work through the infrastructure tasks and don’t get stuck! Don’t spend 30 minutes figuring out NTP. It’s a slow protocol, check synchronization in an hour. If your phones are not picking up an IP Address from the DHCP server and you are sure everything is looking good, don’t spend 45 minutes trying to fix the DHCP server. Change the DHCP server- if the DHCP server on UCM doesn’t work then what about when the router is the DHCP server? Statically assign the Layer 3 settings and TFTP server on the phones as a last resort and re-visit DHCP at a later time. You only need 80. Changing things around is going to help you isolate problems. Find out what works and what doesn’t. This will give you a clue as to where the problem lies. Don’t reboot and hope for the best. The UCM is not a windows box these days (ouch!). Try not to get stuck as the 8 hours is going to fly by!
Incorporate the QoS section when completing the Infrastructure tasks. When you configure the interface with “switchport voice vlan” then you can add “mls qos trust cos” or “auto qos…” as an example. If you have MLP LFI over your FR links it is important to get your Virtual Template attached to your PVC and FRTS enabled sooner rather than later.
Base Configuration for the UCM Phones
The next step is to work your way down the System menu in the UCM admin GUI and configure as much as you can without having to delve too much deeper into the lab. System > Server / UCM / UCM Group / DateTime Group / Region / Location / Device Pool / Enterprise and Service Parameters are all areas of UCM that you should be able to configure without much requirements capture.
Class or restriction is something that is assigned to every entity within UCM and so it is important to address the issue of Calling Search Spaces and Partitions as soon as possible. If you find that this information is not immediately obvious then the creation of placeholder CSS’s per device could be helpful- “CSS-SiteA-Phone1 could be created as an example. At this stage the CSS might be empty (or just contain the partition in which the internal DN’s reside). Creating a placeholder CSS for AAR and CFUR is definitely going to be helpful in case you are required to configure one or both of these High Availability features. Generally speaking you should isolate everything you do for AAR and CFUR with respect to call routing- this will help avoid any unintentional overlap with “normal” routing of calls.
At this stage you are ready to get your devices registered and assigned with the appropriate “base settings”. In UCM Serviceability ensure necessary services are activated (if not done already) and auto-register the phones (assuming that all phones are using the same protocol). Navigate to the device pages under UCM Administration and configure each auto-registered device with the base settings. At minimum try and assign the following “base” settings:
- Description (e.g. SiteA Phone1 5001)
- Device Pool (e.g. SiteA)
- Calling Search Space (e.g. css-sA-p1)
- AAR CSS (e.g. css-aar-cfur)
- Subscribe CSS (e.g. css-internal)
- DN / Partition (e.g. 5001 / pt-internal)
- Call Forward Noanswer/Busy checkboxes for devices with voicemail
- AAR Group
- Display Name
- External Number Mask
- Line Text Label
Yes you will most likely have to return to the devices configuration for more advanced features at a later stage- but that’s ok. There are only a handful of phones in the lab and not every phone will have every feature! So don’t fret about that.
At least an hour will most likely have elapsed- and you will need to progress onto the next section asap but if what has been outlined above resembles anything like your first hour I would say you are in good shape and you could, given a good dose of lady luck, be well on your way!
Good luck with your studies and thanks for reading!
Vik Malhi – CCIE #13890
Managing Partner / Instructor – IPexpert, Inc.