Google Wave has been in private-beta status since Google I/O 2009, and one year later it’s being opened to the public.

Google Wave is a realtime multi-user combination of both Instant Messaging, Email and Document collaboration service. Basically it means that as you write, the other people who are in the Wave with you (Think of it like a chat conversation) see what is being typed as it happens.

So, how do we get started?

If you already have Google Apps for your Domain, you can easily enable this feature for all of your accounts.

To enable this, simply login to your Google Domain dashboard:

https://www.google.com/a/cpanel/yourdomain.com/Dashboard

Obviously adjust yourdomain.com to your actual domain.

It will probably prompt you for your password.

When you’re logged in, you should see a list of of all your enabled services, such as Email, Calendar etc

You want to click on “Add more services” which is just above the enabled services.

You’ll be taken to a new screen where you can select the service you wish to Add to your domain, in this case you want to select Wave.

Anyways, follow your nose a little, but one useful thing that I found was the ability to change the url you access your domains Waves from, much like you can with mail. For example, instead of using:

https://wave.google.com/a/yourdomain.com/

You can have:

https://wave.yourdomain.com/

Much better!

To do this, all you need is to add a CNAME record from wave.yourdomain.com to ghs.google.com

Then, back in your Google Apps Dashboard, if you click on Google Wave, it’ll take you to the subpage to edit the Wave settings for the domain. The first option is: Web Address

Click on Change URL, then enter in your custom URL just as you’ve set it up with your DNS provider.

All done!

Stay tuned as I look to migrate blog hosts to my own provider, so I can start playing around a lot more with Wave and other fun things.

As always, if this has helped you, please leave me a quick message just to say Hi.

Cheers

Chill.

http://webmproject.blogspot.com/2010/05/introducing-webm-open-web-media-project.html

All I can say is well done Google, they’ve truly done the right thing and could have single-handedly helped to preserve the future of video on the web! They’ve also hit the ground running with a very impressive list of people backing the format, such as:

nVidia
Intel
ARM
Qualcom
Adobe Flash platform
Opera
Mozilla / Firefox
CORE Codec
AMD
Logitech and more!

Why is this all so important?

Well, without an “open” standard for the web, essentially if people continued down the path of using a proprietary codec such as h.264, come 2014 the world could literally be held to ransom by the MPEG-LA. They’ve stated that (in laymans terms) it’s OK to use h.264 video for the web and will review this again by 2016 (http://en.wikipedia.org/wiki/H.264#Patent_licensing). What that means is they could decide to start charging for the use of this codec in a few ways, but it’d potentially destroy video on the web with a mad dash for people to find alternatives in order to avoid paying royalties.

Not only that, but with the emerging HTML5 standard (In laymans terms, it’s a way of defining how browsers should display web code) there is the need to define the codec used with the <video> tag, and that means getting a lot of people to agree on it where previously they wouldn’t. Ogg Theora wasn’t good enough quality (amongst other things) but was open-source. h.264 was, but wasn’t open-source.

By preempting all of this, Google has not only given us the chance to use what is essentially a better codec in MANY ways, but they’ve saved the future of video for the web.

Not only this, but by having such an array of people back the format when it’s launched, the WebM project is pretty much a guaranteed success right away! You don’t have to be a rocket scientist to see that when Google puts video in HD on YouTube (It’s currently re-encoding all video that is 720p or 1080p to WebM format), that people are going to start using the format right away.

Same for Skype, they’ve seen that the ex-VP8 codec is pretty much king of the hill in terms of Video Conferencing, and hopefully other 3rd party video conferencing applications and hardware will begin to support it very shortly.

Then there’s support from hardware vendors too, hopefully this means that there will be hardware acceleration for the format that will continue to improve performance, especially with respect to playback on handheld and portable devices like Cellphones or iPods.

Not to mention rather than shun Adobe, they’ve pulled them onboard and the Flash platform will support VP8 also. They’ve got all their bases covered!!

Now, the only thing left for them to do is get Apple onboard with their iPods, iPads and MacOSX. That’s a huge slice of the market that you don’t want to leave out in the cold, but technically speaking there is very little from WebM being supported simply through the push of a software update. Heck, what greater way for Apple to say “Upgrade to the latest iPhone OS 4.0 and you get support for the latest video format for 720p video from YouTube etc.”, as I know that’d be a killer feature and reason to upgrade.

Same goes for Microsoft, they’ve gotta get them onboard. While Microsoft with Internet Explorer haven’t yet (from what I can tell) ruled out the idea of supporting this format, they had announced that they would back h.264. In an ideal world, Microsoft would support WebM in Internet Explorer 9 (IE9 is in heavy development right now), and potentially even push out an optional software update for the likes of Windows Media Player that provides it with support. It would be genius on their behalf as it would mean less people over time *need* to install 3rd party media players and codecs, and can continue to use Windows Media Player.

In summary, I’m absolutely stoaked that Google have released VP8 as the WebM project. It was just 7 hours ago when I was trying the homepage, http://www.webmproject.org only to find nothing on it.

Time will truly tell how this tale unfolds, if VP8 / WebM flies or flops.

All I can say is this: Google, you have my support, well done for thinking of the future of the world, and well done for launching with such an impressive list of companies backing you too!

This post summarizes why NZ broad band doesn’t “suck”, why FTTH isn’t going to solve what your problems are, and what you can do about it in the mean time.

It’ll also explain to you how “The Internet” works, here in New Zealand, as well as giving you some figures to compare us against the rest of the world.

What’s on offer globally? NZ vs the World:

Lets start by looking at Japan.

They have 90% of their homes capable of getting Fibre (1), yet only 30% actually *have* it. Why?

Likely because it’s more cost-effective to get something like ADSL broadband instead, and the benefits that Fibre provides over ADSL2+ for most people is very slim.

What about Korea? They’re apparently well known for their high-speed connections, yet speedtest.net (2) shows an average speed of 22.5m/bit (NZ it down at 4.9m/bit currently). Sure, there’s a lot of things that can affect their speeds, but it’s a decent indication of a “real world” scenario.

Lets compare us to England. In England, the Govt has just pledged to get the whole country on to 2m/bit connections by the end of 2012 (3), whereas estimates are only 1/3 the country will get their “next generation” broadband, of 25mbps.

Now, lets look at New Zealand.

How things were, how they are, and how they’re going to be:

Telecom launched “Jetstream Broadband” back a decade ago now, in the year 2000. They were on the cutting edge of technology, considering ADSL had only been approved as a standard less than 24 months earlier!

Since then, there have been 5 major upgrades in the likes of the Telephone exchanges (The central place where all the phones meet in your suburb, internet is pumped out from there), each one just as costly for Telecom to install as the previous. Can you imagine how that’s got to look trying to justify that to a board of directors?
“Hey guys, we know we’ve only *just* finished the last technology upgrade in each of the exchanges across the country, but there’s something slightly better available now and we want it. Yes, it’s gonna costs millions and millions of dollars but we want to give NZ the best available!”

I don’t know how they pulled that one off time and time again, but they did!

So fast-forward 10 years down the track and the Govt is mandating that Telecom install roadside cabinets across the country. There’s 3600 of them in fact, and they’re over half-way done right now.  The Cabinets have the same technology as the Exchange, they’re effectively a “mini-exchange” if you will, so that houses within a 2KM radius get the absolute *best* in speeds. This is because, the way that ADSL broadband works, the further you are from your Exchange (or roadside cabinet), the slower your attainable speed.

This is a good visual explanation: http://www.chorus.co.nz/f472,15074/15074_Fast_broadband.jpg

Telecom are promising that over 80% of NZ will have 10m/bit+ by the end of 2011.

Englands Govt is promising England just 2m/bit by 2012…

But it gets better. In areas that have been upgraded, NZ’ers are seeing an average speed of 16m/bit! (1)

But it gets even better:

Where the real problem is now:

Right now, many people get half-decent speeds between their home and their Exchange, but that’s not where the problem lies.

The real problem is that because there are so, so many homes that have broadband, all going back to the one Exchange, that the pipe from “The internet” to the Exchange just simply can’t provide everybody with enough to reach their maximum attainable speed.

What does this mean in laymans terms?

Imagine the road to your home is a nice wide road, and you’ve got a pretty decent car too that goes pretty fast. Now imagine that around peak time (Rush hour?), everybody is on the road to your home. What happens? Traffic slows down, and so instead of doing 50Kmph you only end up doing 10Kmph. Make sense?

Picture the roadside-cabinet rollout like this:

Now there’s not just one main road in your area, but potentially dozens and dozens, but the speed limit has been increased from 50KMph to 200KMph!

That’s what the roadside cabinets are doing. Now, if only we could have the same sort of improvements with rush-hour traffic in Auckland!

As Telecom Chorus roll out the new roadside cabinets, with the new ADSL2+ & VDSL2 technology, they’re also preparing for the possibility of running Fibre from these roadside cabinets, and as such the back-end infrastructure is also significantly upgraded from the cabinet to “The Internet” so that there’s always far more than enough room for you to speed along at 200KMph, and then-some!

But won’t Fibre be even better?

Well that’s a really good question. In terms of max raw speeds, yes.

VDSL2 (Which is also being rolled out at the moment) has a max of 50m/bit. Fibre has a max of 100m/bit. However that’s not the issue.

Right now, most people across NZ have a 3GB cap. This is because of two things:

1) The base-cost associated with providing your home with an internet service of some form
2) The high per-GB cost of international data

Data is estimated to cost USD$0.20 per-GB for international data at wholesale costs (What Telecom gets it for). It gets resold in NZ for around NZD$2 per-GB.

Why is this relevant? Well, regardless of if you’ve got Dial-Up, ADSL, Cable or Fibre, it costs your ISP for you to use data internationally.

Ask yourself this: Is it even remotely logical that we would get a super high speed plan, capable of downloading masses of data in a short time, for the same price as you currently get just 3GB to 10GB?

The long and short: No.

With having a higher speed fibre plan, you’re either going to chew through your measly 3GB data cap in seconds, or it’s going to cost you significantly more to have a bigger monthly data allowance. What’s the point in having a line that can burn through your data allowance in under a minute? Imagine if you get infected with a virus and it starts sending out to the internet, your bill at the end of the month would be astronomically higher because of the faster speeds.

Sure, prices may have adjusted then, data costs may be less, but why would the ISPs suddenly *give* away their margins?

I certainly wouldn’t roll out a multi-billion dollar fibre network, and then make a loss. Heck, if it’s the latest and greatest, then I’m going to charge for it and make it a “premium” product, just like everything else in this world is when they’re brand new.

Even when it’s not “brand new”, because it’s still “top of the line”, it’s never going to be as cheap as the likes of ADSL2+ or VDSL2, because there’s less involved in cabling to your house. They just have to lay new cable to each of the Cabinets as they upgrade them (3600 of them), not to millions of homes and businesses.

It’s also worth noting that new technology has shown speeds of 500m/bit achievable through using just a standard copper telephone line (4), and given Telecom’s history, if this technology is decent, do you think it’ll take long to get to NZ?

But I can’t survive with ADSL broadband!

Well actually you most probably can.

In fact, almost all the issues that you’ll find you have with ADSL, you’ll have with fibre.

If you’re going to be watching High-Def video on YouTube, gaming, VoIP calls and browsing the web all at once, you’re potentially going to run into issues on both Fibre and on ADSL2+.

Why?

Well when you load up YouTube, it just goes full-throttle downloading the video as quickly as it can. That’s just the nature of the beast. If you’ve got ADSL2+ or Fibre, you’re still going to run into issues where it uses up all of your connection and leaves no room for gaming, browsing or VoIP calls.

What you need is your ISP or your Router to manage the flow of things. Because in NZ we currently don’t have any ISP’s who will do this (Although I’ve discussed with multiple Big-Cheeses at Telecom the prospect of this), you have to do it yourself.

Basically what you do with your Router is what’s called “QoS”, which stands for “Quality of Service”. In a nutshell, it’s going to give some types of traffic, such as Gaming, or Voice calls priority over other things, such as file-sharing or video-streaming.

Some routers do this better than others, ask around your techno-savvy friends, on PressF1.co.nz for advice, or even here in the Comments

So, what are you saying with all this?

I’m suggesting that Fibre to the Home in New Zealand is *not* going to be the long-awaited savior people are hoping, but rather education about the different types of broadband available, and helping people get on to ADSL2+ (And cabinetized) connections will solve pretty much all of New Zealands issues that can be solved at an infrastructure level. Not only this, but it will all happen within the next 18 months, rather than 10 years.

Naturally there are exceptions to this rule, some people want to do things like real-time High-Def video broadcasting which you simply must have Fibre for. They’re the exception, not the norm. For the vast majority of NZ, Fibre’s not going to be any cheaper nor provide any tangible benefits in terms of usable speed or data caps.

Do yourself a favor, and get ADSL2+!

References:

1) http://www.nbr.co.nz/opinion/chris-keall/average-nz-internet-speed-just-297mbits-akamai

2) http://speedtest.net/global.php#0

3) http://news.bbc.co.uk/2/hi/technology/8618507.stm

4) http://www.gizmag.com/ericsson-vdsl2-500mbits/11296/

That’s great if you have ABE, but not so good if you’re one of the millions of installs out there that uses vanilla Asterisk.

This guide presumes you already have ADA downloaded and installed. If not, you can get the latest (At the time of writing) version 1.1 from here:

http://dl1.digium.com/ADA1.1/ADAInstall.exe

Now you’re going to ignore any instructions you’re previously read, and we’ll start from scratch. Don’t worry, you’ll be up and running in a matter of moments!

Once ADA is downloaded, installed and running, you should be greeted with a login window.

What you now need is the “manager” user to be setup.

The “manager” user is different from your regular extension, in that it allows you to control the actions of Asterisk. With this in mind, you need to be careful when setting up an Ext, not to have a weak password, and make sure you use the permit / deny lines, even for remote extensions. You can get a password generated for you here: http://www.thebitmill.com/tools/password.html

From the shell, we’re going to edit /etc/asterisk/manager_custom.conf like this:

nano -w /etc/asterisk/manager_custom.conf

You’re now going to create a new entry with the following details, in my case I’m Ext 199:

[199]
secret=mkHHa031kHMsGRu
deny=0.0.0.0/0.0.0.0
permit=192.168.0.0/255.255.255.0
read = system,call,log,verbose,command,agent,user
write = system,call,log,verbose,command,agent,user

Now you need to create a manager user for every Ext that will be using ADA. This is because ADA uses the username part as the Ext for it to ring to connect the call to.
Without it, it would be the same as you dialing the number, and the system just sitting there, holding it.

Ctrl + X to quit nano, and you can safely get your ADA desktop client to login with the username: Ext@asteriskIP

In my case it would be something like: 199@192.168.0.250

Use the password you specified above in the manager user, and not your regular SIP Secret.

If you find you’re unable to connect, confirm that you’ve got the right password, and adjust the “permit=” line as applicable. I would highly recommend leaving it in there and just adjusting it for your IP range. Same goes for external or remote clients, make sure that you have their IP Address.

Now that you’ve done that, you need to hook ADA into your dial-plan.

Again, the standard documentation is no good here, so, we’re going to edit another file:

nano -w /etc/asterisk/extensions_custom.conf

At the top of this file (Well under the first tow lines of comments) we’re going to add in the following:

[ada]
include => from-internal

Then hit Ctrl + X and follow the prompts to save and quit.

Again the standard documentation is no good for us there, as the “ADA” context is actually “ada” in lower case, and we’ll use the “from-internal” instead of “default”.

Now you need to refresh the dial-plan (Just this once) so run this:

asterisk -rx 'dialplan reload'

You’re now all good to go! On your ADA desktop client, you can create a test call! It’s easiest to test to another Ext in my opinion, so enter in the Ext # of another extension on the system and hit “enter”.

What should happen:

Your phone will ring immediately

You pick it up and will hear ringing, the destination phone should now also be ringing

Easy!

If this has helped you in any way, please leave a comment and say Hi.

Cheers

Chill.

VP8 is potentially a game-changer in the realm of both streaming, High-Definition, and mobile video devices. It also has the potential to help in the killing of Adobe Flash, and solve the HTML5 Video format war.

Currently there are two formats that are used in HTML5 Video by differing browsers: Ogg Theora and h.264

Both have support from different browsers, both have different camps fighting for their success, and both have their pros and cons.

Ogg Theora is open-source, and is supported by Mozilla Firefox and Opera.

h.264 is a proprietary codec with paid licenses, it’s supported by Safari and the upcoming Internet Explorer 9 (1).

Google Chrome currently supports both h.264 and Ogg Theora video.

A little history

Ogg Theora was open-sourced a while back now. Previously it was known as VP3, initially made by a company called On2 Technologies. On2 have recently been purchased by Google, and their latest and greatest codec VP8 is a real contender for h.264 video, but currently is under a restrictive license. An example of a “standard” piece of encoding software costs around USD$39.

The arguement

There are reasons why you want to use Ogg Theora and there are reasons why you don’t. I’ll list them all in bullet points, just for ease of reading, and because I’m lazy:

• Ogg Theora is open-source, therefor free and can be easily incorporated into a browser
• Because it is open-source, people fear it may suffer lawsuits due to software patent infringement (2)
• h.264 is arguably better quality, though those from the Theora camp say that they can match or better the quality of h.264 with fine tuning, personally I’ve found that h.264 is better.
• High Definition video requires quite a bit of processing power to decode (Well and even to encode), so it can suck battery life out of things like Cellphones or iPods etc like crazy. h.264 is particularly CPU intensive even when standard or mobile definitions.
• VP8 is supposed to use around 50% the CPU requirements to decode the same video, when compared with h.264 (3)
• VP8 uses less bandwidth to achieve the same quality (4)
• VP8 is able to adjust its quality in realtime, based on things such as CPU use and bandwidth available, and as such would be superb for video conferencing (5)
• h.264 video is currently free for use in browsers until the year 2016, but after that who knows, the patent owners could begin charging. If h.264 is *the* codec that’s used on the web, why wouldn’t they, and make millions? (6)
• Apple is also heavily invested in h.264 with their iPhone / iPad / iPod Touch platforms, they’re specifically against flash

Why this all matters? Why not flash?

Well, in a world that’s heading towards Touchscreens amongst other things, there’s less and less reasons why people would want to use Flash.

Known for being horribly insecure and one of the main reasons why an OS would crash, it’s renowned for being CPU and bandwidth intensive (7).

HTML5 is a way to replace Flash for almost everything, and it’s going to die a slow death (My own prediction there), but it’s going to happen.

What matters is finding a suitable replacement to the current Flash-Video, prior to the finalizing of the HTML5 specification.

The options

If we go with Ogg Theora, it’s free and open-source, but may come under fire from lawsuits. It’s debated that the quality isn’t as good as h.264 either

If we go with h.264, it’s licensed and popular already, but the owners could begin to hold the world ransom from 2016, and then the world would either have to pay up, or overnight switch to something else. That’s what the team over at Mozilla is worried about.

We could go with something like VP8, if open-sourced. It’s got Google behind it, so immediately if they start using it on YouTube, then millions of PC’s would suddenly support the codec in order to play back the video.

If it’s open-source, it’ll keep the Opera & Mozilla camps happy. If it’s not Flash, it’ll keep Apple (Safari) happy, and they could potentially use it in their iPhones etc with an update. Google own it, so naturally they’d support it in Chrome, which just leaves Microsoft. They really want their own Silverlight, but they’re the only ones, and nobody else seems to like that idea, so they’ll basically just have to suck it up and go along with it even if they don’t like it. Besides, it’d be in their best interests in the long run, compared with h.264 anyway.

Summary

This is why we need VP8 to be open-sourced, and for Google to push it via YouTube.

Flash will die

VP8 will use less bandwidth and CPU compared with h.264, so it’s great for mobile devices like iPhones / iPods / Laptops on battery, and embedded devices like Media Centers. Being open-source will mean people can work on hardware acceleration for encoding / decoding for their particular device (if applicable), and it’ll mean that the world isn’t about to be held to ransom over the code (Worst-case is Google does though ..).

References:

1) http://blogs.msdn.com/ie/archive/2010/04/29/html5-video.aspx

2) http://yro.slashdot.org/story/10/04/30/237238/Steve-Jobs-Hints-At-Theora-Lawsuit

3) http://www.on2.com/index.php?605

4) http://www.on2.com/index.php?564

5) http://www.on2.com/index.php?606

6) http://newteevee.com/2010/02/04/good-news-for-html5-h-264-streaming-will-remain-free/

7) http://www.apple.com/hotnews/thoughts-on-flash/

If you’ve disconnected from a Domain for example and are now part of a workgroup but want to retain your existing “setup”, or perhaps you’ve got yourself something like an EeePC 900 and your C: Drive is running out of space, this can be a great way to free up additional space, and it’s a better all-round solution that purely moving the “My Documents” folder.

The whole process should take only 2-3 minutes!

Firstly you need to have another user account on the machine. Ideally you can just use the Administrator login, but it’s just as easy if you have another Admin user on your PC.

So, what you need to have done is logged in as your “local” user (If moving from a domain). You may need to create a new User, we’re going to use “Bob” as the example in this. Login as Bob, then log out again.

Now, log in as a *different* administrative user, such as Administrator

Move C:\Documents and Settings\Bob, and put it where you want it to be (If applicable).

Click Start –> Run –> regedit

Go to HKEY_LOCAL_MACHINE > SOFTWARE > Microsoft > Windows NT > CurrentVersion > ProfileList

You should see several entries in there. You’ll need to look around at them all, and you want to check the “ProfileImagePath” key. It’ll be something like:

%SystemDrive%\Documents and Settings\Bob

You can now adjust it to something like:

%SystemDrive%\Documents and Settings\Bob.DOMAIN

Or:

D:\BobFiles

All done!

Reboot and login as Bob and you should be OK!

This basic How-To shows you how to set it up to work with your Elastix system:

First off, you’ll want the following config which is a bit of a mash-up from some other documentation on making it work with an Aastra 800 (Include the header):

#----------------------------------------------------------------#
#                                                                #
# Template for SN4554/2BIS/UI to use with Elastix                #
# Use with firmware R5.1 or higher                               #
#                                                                #
# This template uses dhcp to retrieve an IP address. Comments in #
# the file indicate what to change (Start with '#')              #
#                                                                #
# See the SmartWare Software Configuration guide for command     #
# details (http://www.patton.com/manuals/SCG-r52.pdf)            #
# Modified by Josiah Spackman                                    #
# http://chillingsilence.wordpress.com                           #
#----------------------------------------------------------------#

cli version 3.20
webserver port 80 language en

system

  ic voice 0
    low-bitrate-codec g711alaw64k

system
  clock-source 1 bri 0 0
  clock-source 2 bri 0 1

profile ppp default

profile tone-set default

profile voip default
  codec 1 g711alaw64k rx-length 20 tx-length 20
  codec 2 g711ulaw64k rx-length 20 tx-length 20
  fax transmission 1 relay t38-udp

profile pstn default

profile sip default

profile aaa default
  method 1 local
  method 2 none

context ip router

  interface IF_IP_WAN
# replace 'dhcp' with your fix IP if needed, e.g. "ipaddress 172.16.1.20 255.255.0.0"
    ipaddress dhcp
    tcp adjust-mss rx mtu
    tcp adjust-mss tx mtu

context ip router
# uncomment the following line if you need to add routing table entries
  # route 0.0.0.0 0.0.0.0 172.16.1.1

context cs switch
  national-prefix 0
  international-prefix 00

  routing-table called-e164 RT_ISDN_TO_SIP
# 'T' in the following lines specifies "digit collection". The time for collection can be adjusted replacing 'T' wih 'T2' for 2 seconds.
    route T dest-interface IF_SIP

  routing-table calling-e164 RT_SIP_TO_ISDN
    route default dest-service SV_HUNT_PSTN MP_Unknown-Subscriber

# This mapping table sets the ISDN type of number for calls towards ISDN to 'subscriber'
  mapping-table calling-e164 to calling-type-of-number MP_Unknown-Subscriber
    map default to subscriber

  interface isdn IF_ISDN_0
    route call dest-table RT_ISDN_TO_SIP

  interface isdn IF_ISDN_1
    route call dest-table RT_ISDN_TO_SIP

  interface sip IF_SIP
    bind context sip-gateway GW_SIP
    route call dest-table RT_SIP_TO_ISDN
# This is the IP of your Asterisk. Replace with domain name if DNS server is available
    remote 192.168.0.250

  service hunt-group SV_HUNT_PSTN
    drop-cause normal-unspecified
    drop-cause no-circuit-channel-available
    drop-cause network-out-of-order
    drop-cause temporary-failure
    drop-cause switching-equipment-congestion
    drop-cause access-info-discarded
    drop-cause circuit-channel-not-available
    drop-cause resources-unavailable
    route call 1 dest-interface IF_ISDN_0
    route call 2 dest-interface IF_ISDN_1

context cs switch
  no shutdown

# The parameters realm, username and password have to match your Asterisk configuration
authentication-service AUTH_AASTRA800
  realm 1 smartnode-gw
  username patton password 6953789

# The parameters domain, username, identity (=username) below have to match your Aastra800 configuration
location-service LS_AASTRA800
  domain 1 smartnode-gw

  identity-group default

    authentication inbound
      authenticate 1 authentication-service AUTH_AASTRA800 patton 6953789

  identity 6953789

    authentication inbound
      authenticate 1 authentication-service AUTH_AASTRA800 patton 6953789

    registration inbound

context sip-gateway GW_SIP

  interface WAN
    bind interface IF_IP_WAN context router port 5060

context sip-gateway GW_SIP
  bind location-service LS_AASTRA800
  no shutdown

port ethernet 0 0
  medium auto
  encapsulation ip
  bind interface IF_IP_WAN router
  no shutdown

port bri 0 0
  clock auto
  encapsulation q921

  q921
    protocol pp
    uni-side auto
    encapsulation q931

    q931
      protocol dss1
      uni-side user
      bchan-number-order ascending
      encapsulation cc-isdn
      bind interface IF_ISDN_0 switch

port bri 0 0
  no shutdown

port bri 0 1
  clock auto
  encapsulation q921

  q921
    protocol pp
    uni-side auto
    encapsulation q931

    q931
      protocol dss1
      uni-side user
      bchan-number-order ascending
      encapsulation cc-isdn
      bind interface IF_ISDN_1 switch

port bri 0 1
  no shutdown

Copy this all into a notepad window and save it as “SN4554.cfg”

What you’re going to want to modify is the references to “192.168.0.250″ and point it to your Elastix / Asterisk system.

All the ISDN ports are set to “Point-to-Point”. You can adjust the “pp” settings to “pmp” if you require “Point-to-Multipoint”, though to be honest ISDN isn’t my strong point so I’m not entirely sure how relevant that is.

This will leave the WAN port using DHCP, hopefully you’ve got a semi-intelligent DHCP server that will give out the same IP each time.

Open the WebGUI of your SN4554 up, login with the default Username “administrator” and a blank password.

Click on “Import / Export” on the left-hand side, then hit the Import Configuration tab.

Select the file, import it, then reload the device.

Now, in asterisk you want to add a new SIP trunk. Name it “ISDN” and put the following in the SIP PEER Details:

username=patton
type=friend
secret=6953789
qualify=1200
insecure=very
host=192.168.0.141
dtmfmode=RFC2833
disallow=all
context=from-pstn
canreinvite=no
allow=alaw&g729

There’s no register string or USER details, though it may be worth specifying the max channels as “4″.

You’ll need to adjust the “host” setting to the WAN IP Address of your Patton box, as we’re semi-insecure with such a basic password.

You *could* adjust the password and username in the config etc as applicable, but specifying the host should be enough, especially considering (in my instance) the whole system is LAN-accessible only.

When you’re finished, you should be able to login to your Elastix system via SSH and run:

asterisk -rx ‘sip show peers’

And see:

ISDN/patton                192.168.0.141        N      5060     OK (19 ms)

If you do, congratulations, you’re ready to go!

Now you just need to setup inbound and outbound routes as applicable.

If this was useful to you, please leave a comment and say hi.

I’d also like to thank Byron from SnapperNet in New Zealand, he’s been such a great help, and also provided me with the initial configuration samples.

This How-To has also been re-posted here: http://blogs.elastix.org/en/2010/05/using-the-patton-sn4554-for-isdn-with-elastix/

Everybody is starting to talk about FreeSWITCH as the next big PBX software (Amongst other things), and FreePBXv3 is shaping up to be a damn fine GUI.

The more I use it, the more I start to like FreeSWITCH as a progression from Asterisk. One will likely never completely kill the other, I can see a world where they both live happily, but for me I struggle knowing that Asterisk doesn’t 100% accurately do all CDR stuff, and it bugs me recommending that to a client.

I’ve blogged before on FreePBXv3, and the dev guys are still as friendly as ever (Granted they seem to be a truckload busier lately, but who can blame them).

What follows is a guide on booting a LiveCD (Debian!) and running FreeSWITCH & FreePBXv3 from this live system. From there, it’s up to you if you want to simply reboot and forget it ever happened, or potentially install it to a HDD. The choice is yours, I’m just going to give you a quick How-To on getting FreeSWITCH & FreePBXv3 up and running nicely, and in a flash! In fact, on a decent machine (C2D 2Ghz+) and a decent internet connection, you could be up n running from scratch, in less than 20 minutes

You’ll need a Blank CD (Or VirtualBox for example), and potentially some form of external thumbdrive that you can format. In my case I used an 8GB SDHC card.

Start by grabbing yourself the LiveCD: http://cdimage.debian.org/cdimage/release/current-live/

I used: i386/iso-cd/debian-live-504-i386-standard.iso

Once you’ve downloaded it, burn it off or boot it up in VirtualBox

Great! Now, become root:

sudo bash

Change the root password:rm

passwd

Enter something nice, coz we’re presuming you’re going to copy / paste via SSH for the rest of this.

Now run:

apt-get update
apt-get install openssh-server

Once that’s done, login remotely with ssh and carry on as-per usual. This is kind of optional but saves you typing package names manually etc.

OK now before we run out of HDD space, we’re going to mount the USB drive.

Use cfdisk to create a linux partition, then format it with (something like):

mkfs.ext3 /dev/sdb1
mkdir /mnt/usb
mount /dev/sdb1 /mnt/usb

You may need this if you’re working on a machine with a low amount of RAM and you find you run out of “disk space” on the ramdisk:

mkdir /mnt/usb/src
mkdir /mnt/usb/freeswitch
mkdir /mnt/usb/archives
rm -rf /usr/src
ln -s /usr/src /mnt/usb/src
ln -s /usr/local/freeswitch /mnt/usb/freeswitch
ln -s /var/cache/apt/archives /mnt/usb/archives

Now we want you to install all the following:

apt-get install php5 php5-xcache php5-xmlrpc php5-mysql php5-gd php5-cli \
build-essential git git-core autoconf libgdbm-dev libdb-dev subversion php5-curl \
automake libtool libncurses5 libncurses5-dev php-db mysql-server-5.0 mysql-client-5.0

They’re all going to come in handy at some point for either FreePBXv3 or FreeSWITCH.

Make note of your MySQL root password, you’ll need it for later on!

Now we begin with the FreeSWITCH compile:

cd /usr/src
git clone git://git.freeswitch.org/freeswitch.git

This will pull the latest version from subversion, which for most applications is a bit of a No-no, but FreeSWITCH seems to be relatively stable on a day-to-day basis and it’s actually recommended from what I can tell.

cd freeswitch.git
./bootstrap.sh

This is the beginning of what could be a very long compile if you don’t have semi-recent hardware…

Provided there’s no errors, proceed on to:

./configure

Now, I’m doing it this way rather than the proposed “quick n nasty” because for some reason it kept being unable to find the right libraries when I *know* they’re on the system. YMMV.

Next we make with:

make

Then we make install:

make install

IF you get errors right away like I did, then try running:

make current

I’m unsure why, but it worked for me after make install errors (Thanks to bruce from #freeswitch).

Next we install sounds:

make sounds-install moh-install

Tada, all done (For FreeSWITCH at least)!

Note, you can also do the High Definition sounds with:

make uhd-sounds-install uhd-moh-install hd-sounds-install hd-moh-install

Start the service with:

/usr/local/freeswitch/bin/freeswitch

You should be able to point a SIP device at your debian box’ IP address, with Ext 1000 & Secret of 1234, then dial 9999 or 5000 to test, as-per the instructions here: http://wiki.freeswitch.org/wiki/Quick_and_Dirty_Install

You can quit it at any time by typing “shutdown”.

FreeSWITCH down. Now on to FreePBXv3!

cd /var/www
svn co http://www.freepbx.org/v3/svn/trunk/ freepbx-v3/

Now we’re going to sort out the permissions:

chgrp -R www-data /usr/local/freeswitch/conf/*
chmod -R g+w /usr/local/freeswitch/conf/*
chmod 777 freepbx-v3/freepbx/logs/
chmod 777 freepbx-v3/freepbx/cache/
chgrp -R www-data freepbx-v3/freepbx/config/*
chmod -R g+w freepbx-v3/freepbx/config/*
mkdir freepbx-v3/upload
chgrp -R www-data freepbx-v3/upload
chmod -R g+w freepbx-v3/upload

Debian uses www-data for the http group, instead of apache.

We’re not going to bother with mod_rewrite, I’m lazy and it’s not crucial.

Now, go to http://debian.ip/freepbx-v3/

Follow through the first part of the install, accepting the terms, the pre-flight checklist should be all good except for mod_rewrite, and now we’re onto the actual configuration:

Leave the host as 127.0.0.1, but set the username as “root” and the password as the one you defined earlier. Pick your timezone as applicable and hit Next.

Create a Master Admin account and hit Next

The defaults are fine on the next page, leave it as the “FreeSWITCH Driver” and hit Next, twice (To tell it it’s safe to overwrite the config files).

The default selections are fine for “packages” to be installed, click next, twice (to ignore the errors).

We’re not ready to install! Hit the button to begin the process

It’ll then say Installation Complete, click the link and begin using FreePBXv3!

Don’t forget, you may need to restart freeswitch if you killed it earlier, with:

/usr/local/freeswitch/bin/freeswitch

I’ll post a brief intro to FreePBXv3 shortly. In the mean time, enjoy, and please post feedback!!

NOTE: You will no longer be able to connect immediately to FreeSWITCH with your SIP device as you did earlier. This is because FreePBXv3 has overwritten the config files with it’s own, however we’ve accomplished the goal of this post, to get FreeSWITCH & FreePBXv3 up and running on a LiveCD in no time at all! From here, you’ll need to configure everything with FreePBXv3

Contents:

1. Installing the base system
2. Updating the config
3. Fixing up HTTPS
4. Updating dial-plan
5. Closing words

1) Installing the base system

First you’re going to need to ssh in to your Elastix PBX System.
Then, run the following:

cd /var/www/html
wget http://www.kefa.it/hotel.tar.gz
tar zxvf hotel.tar.gz
rm hotel.tar.gz
cd hotel
mysql -uroot -peLaStIx.2oo7 < Hotel.sql

Very similar to the default steps taken on the official documentation, except we’re going to use the Elastix SQL password rather than the script they’ve got there. This is also easier than the official doco, just one step instead of two.

2) Updating the config

We’re now going to modify the configuration file so that it works with Elastix due to the different Username / Passwords used between distributions.
Edit config.inc.php with your favorite editor. In my case I use nano:
nano -w config.inc.php

Now you’re going to modify the following three paragraphs so that they match below:

//Hotel database
$dbhost = "localhost";
$dbpass = "eLaStIx.2oo7";
$dbuser = "root";
$dbname = "Hotel";

//Asterisk database
$dbhost2 = "localhost";
$dbpass2 = "eLaStIx.2oo7";
$dbuser2 = "root";
$dbname2 = "asterisk";

//Asteriskcdrdb database
$dbhost3 = "localhost";
$dbpass3 = "eLaStIx.2oo7";
$dbuser3 = "root";
$dbname3 = "asteriskcdrdb";

Save and exit and you’re done with the config.

3) Fixing up HTTPS

When you visit the Elastix WebGUI, there’s a condition that rewrites the URL to include HTTPS. However, when this system tries to check the state of an Extension, it must do it via HTTP and not HTTPS, so we add an exception.

Edit /etc/httpd/conf.d/elastix.conf with your favorite editor and add these lines at the very bottom:

RewriteEngine Off

What this will do is add the /hotel directory as an exception so you don’t have to use HTTPS to get to it. We leave HTTPS on for the rest of the PBX, just because there’s no real reason for us to turn it off.
Now, restart apache with:

/etc/init.d/httpd restart

You should now be able to access your WebGUI without HTTPS at: http://pbxip/hotel/
If not, then go back and try again, as this part is crucial to the redirection to Reception when a phone is not Checked In

4) Updating dial-plan

Now we’re going to add in to the Dial-Plan and tell the PBX what calls can and cannot be made when the phone is not Checked-In.

Edit /etc/asterisk/extensions_custom.conf in your favorite editor and add the following underneath the [from-internal-custom] header:

;lets guests call other rooms without being checked in
exten => _XXX,1,Dial(SIP/${EXTEN})
exten => _XXX,n,Hangup()
;lets guests call freephonenumbers without being checked in
exten => _1800NXXXXXX,1,Goto(outbound-allroutes,${EXTEN},1)
exten => _1800NXXXXXX,n,Hangup()
exten => _X.,1,Set(result=${CURL(http://127.0.0.1/hotel/checkuser.php?Ext=${CALLERID(num)})})
exten => _X.,n,NoOp(Results: ${result})
exten => _X.,n,GotoIf($["${result}" = "OK"]?OK)
exten => _X.,n,GotoIf($["${result}" = "KO"]?KO)
exten => _X.,n(OK),Goto(outbound-allroutes,${EXTEN},1)
exten => _X.,n,Hangup()
;call front desk if not checked in
exten => _X.,n(KO),Dial(SIP/200) ; used for just one extension
;exten => _X.,n(KO),Goto(ext-group,0,1) ; used for ring groups
exten => _X.,n,Hangup()

If you want to allow additional free calling, just duplicate the lines with “1800″ in them. Mines quite different for New Zealand, but I’ve left it as american to give you a good idea of what you can do there.
You’re also going to want to adjust the third-to-last line where it says “SIP/200″, replacing 200 with the Extension of Reception.
Or, you could just comment that line out, and uncomment the second-to-last line to use Ring Group 0 as your “Reception” extension. The choice is yours.
Save the file and run:

asterisk -rx "dialplan reload"

5) Closing words

From here, you’re on your own. It’s dead easy to log in and get started. You’ll first need to import all the rooms which it takes from the Ext Caller ID name, so it pays to give your Ext’s decent names like “Room 201″ for Ext 201, just for example.
If you run into troubles, post here and myself or the Elastix team would be more than happy to assist.
If you get this successfully installed, just leave us a quick comment to say Hi and tell us you’ve done so, it’s always nice to know our time is appreciated.
Cheers

Chill.

Imagine a client of yours happens to move their PBX, perhaps they are shifting premises (Without notifying you — Their prerogative I suppose). Now, with the change of ISP’s they’ve changed router and network settings, new public IP Address etc, so until you fix the sip_nat.conf settings calls are dead quiet. What do you do?

Well, you could ring them up and get them to port-forward their router and let you in. Or, you can do everything yourself via your shiny remote-control VPN!

Not only does it not matter where your customer plugs in the box, provided DHCP gives it an IP Address and they don’t have some silly proxy in the way, it will tunnel out NAT, past their routers firewalls, and let you in! It’s a service-operators dream! Used in conjunction with the SSH Tunnelling blog article, you’ll find this gives you total control of their local endpoints as well, just as if you were right there sitting on their LAN. You can even access your clients ADSL Router WebGUI and make the required port-forward changes yourself, provided they give you the admin password of course.

Table of Contents

1. Expected outcomes of this How-To
2. Example final VPN layout
3. Preperation & IP Addressing
4. About the structure
5. Server setup
6. Client setup
7. Testing
8. Last words

1) Expected outcomes of this How-To

At the end of this How-To you should:

• Be proficient in generating Keys for clients
• Understand basic security, what files are private and must be kept secret
• Understand the basics of how OpenVPN is setup and works
• Know how the VPN is routed, what boxes are able to see each other
• Have setup a VPS with OpenVPN
• Have clients that initiate the connection to the server, tunneling out past NAT etc
• Hopefully have a fully working VPN that allows total remote management of your PBX systems

2) Example final VPN layout

You can see that even though each PBX box (And the Admin Laptop) have different IP ranges, they’re all connected to the VPN Server. The PBX’s can “see” each other as well as the Admin Laptop, and vice versa. What cannot be seen is the other devices on remote LAN’s, such as Phones, Computers, and other devices. That requires SSH tunneling. Click the picture for a larger version:

In my case, I’ve used a VPS (Virtual Private Server) to host the VPN, it’s just a vanilla box running Debian. I do this because it’s then hosted in a Data Center, it’s always-on, always accessible, and ultra reliable!

3) Preperation & IP Addressing

Keep in mind that if you’re using 192.168.1.1 on your client LAN and 192.168.1.1 for your VPN “LAN”, then you’re going to run in to routing issues and it’s not going to work.

If you use 10.0.0.0 with a subnet of 255.255.255.0 then you’re also going to potentially run into issues. However if you use a subnet of 10.0.0.0 with a subnet mask of 255.255.255.0 then you’re less likely to hit issues.

With this in mind we recommend you pick something in the middle of the 10.X.X.X range (Such as 10.99.88.0/24). We’re also going to start off with a /24 which will give you approx 62 usable clients. OpenVPN uses 4 IP Addresses per-client, I’m not entirely sure what for, but I believe one is the real IP, one for broadcast traffic etc.

Starting off with a /24 just because we can, because we’ve got no immediate need to have more than 60 devices, and because it’s easier to try and preach to the masses about what the IP Addresses are that will be used.

4) About the structure

We’re going to be setting up the VPN in a One-to-Many structure, with one server and many clients.

Every client gets one Key / Certificate file. Each client (key) can only be used once at a time, trying to have two machines use the same will result in issues. You can have multiple clients with different keys connected concurrently however (That’s the main purpose of this).

Also, the way it’s setup is so that each client (The Elastix PBX) can see every other client. Same goes for any “Administration” devices, such as a Windows Laptop used to administer the PBX systems, it’ll be able to see all the devices and they will all be able to see it. If this is not desirable, I would suggest that you use something like iptables on the main VPN Server to prevent this.

Another thing to keep in mind is that this can be used for direct Elastix -> Elastix links, for secure tunneling and communications. All you need to do is replace the VPS VPN Server with an Elastix machine. It really makes little difference.

One thing to keep in mind is that the other devices on the LAN of the clients cannot see anything else on the VPN, in fact they’re generally completely unaware of the fact that the Elastix PBX is even connected to a VPN. If you want to remotely administer phones via this VPN, you’ll need to read up on the SSH Tunneling How-To, and use your imagination a little bit. It’s incredibly easy.

5) Server setup

We’re going to begin by installing the relevant RPM’s from the Dag repo’s: http://dag.wieers.com/rpm/

Depending on your distribution, you may simply need to run something like: apt-get install openvpn

Note: I’m now assuming that you have OpenVPN installed. There’s no difference between client / server, so it’s just the one “openvpn” package. I’m also assuming you’re running a Linux server. You *can* use Windows, I’ve done it twice before with great success, but this documentation is written with a Linux server in mind. You can bridge the gaps by reading here: http://www.openvpn.net/index.php/open-source/documentation/howto.html#quick

Now, we’re going to edit /etc/openvpn/server.conf using your favorite editor (I’m a nano fan) and put in the following:

port                    1194
proto                   tcp
dev                     tun8
tls-server
cipher                  BF-CBC
ca                      /etc/openvpn/server/ca.crt
cert                    /etc/openvpn/server/server.crt
key                     /etc/openvpn/server/server.key
dh                      /etc/openvpn/server/dh1024.pem
client-to-client
mssfix                  1200
comp-lzo                yes
keepalive               7 20
server                  10.99.88.0 255.255.255.0
ifconfig-pool-persist   /etc/openvpn/server/vpn.hosts
route-up                "route add -net 10.99.88.0/24 tun8"
push                    "route 10.99.88.0 255.255.255.0"

Now I’ll explain:

We use the default port of 1194, TCP (Not UDP), with the BF-CBC encryption (because it’s slightly lighter on the CPU than the likes of AES, though feel free to debate that with yourself).

We’re telling it that everything is gong to be in the /etc/openvpn/server directory, and a few other little “fixes” that are useful to have.

Now, we run:

cp -Rv /usr/share/doc/openvpn/examples/easy-rsa/ /etc/openvpn/

This copies the files needed to generate the keys into an easy location in /etc. In my case, there’s a 1.0 and a 2.0 directory. I’ll presume you’re in the 2.0 directory whenever I refer to it.

So, you should now be in /etc/openvpn/easy-rsa/2.0

Now you’re going to edit the file “vars”. Go right down to the very end and edit the last 5 lines.

Run:

source vars
./clean-all
./build-ca

This is going to clean everything up and build your Certificate Authority (VPN Server).

Where it asks for values, just press Enter to accept the defaults that you’ve entered in to the “vars” file.

Where it asks for the Common Name (e.g. server Hostname) you can press Enter and it will call it “CA”.

Now we’re going to generate the Certificate & Keyfile for the server:

./build-key-server server

It’ll ask if you want a password, for the scope of this document we’re just going to press Enter and skip the password.

When you’ve pressed Enter and gone through it all, you’ll see a confirmation right before the files are created:

Check that the request matches the signature
Signature ok
The Subject's Distinguished Name is as follows
countryName           : PRINTABLE:'NZ'
stateOrProvinceName   : PRINTABLE:'AK'
localityName          : PRINTABLE:'Auckland'
organizationName      : PRINTABLE:'Anonymous VoIP Company'
commonName            : PRINTABLE:'server'
emailAddress          : IA5STRING:'Fake@anonymousvoipcompany.co.nz'
Certificate is to be certified until Mar 14 03:34:48 2020 GMT (3650 days)
Sign the certificate? [y/n]:y
1 out of 1 certificate requests certified, commit? [y/n]y
Write out database with 1 new entries
Data Base Updated

Note: I’ve put in fake values here just so that I don’t get spammed, but I recommend you use real details.

Now we’re going to generate Diffie Hellman parameters:

./build-dh

Now we put the files into the right location with:

mkdir /etc/openvpn/server
cp keys/dh1024.pem /etc/openvpn/server
cp keys/ca.crt /etc/openvpn/server
cp keys/server.crt keys/server.key /etc/openvpn/server

Everything on the server should now be all done, so you can restart the service:

/etc/init.d/openvpn restart
echo 1 > /proc/sys/net/ipv4/ip_forward

The last line is optional to enable forwarding.

Congratulations! Now we can move on to the client setup!

6) Client setup

We’re still working on the server at the moment, but we’re going to presume you’ve come back to re-generate another key.

To start with on the server run:

cd /etc/openvpn/easy-rsa/2.0/
source vars

This sets up the server to use the variables you’ve entered when you first setup the server.

Now we’re going to generate the key with:

./build-key clientcompany1

I call it “clientcompany1″ as I support multiple clients and this links them. You could also call it something such as “southernoffice” or anything you like really.

Now you want to copy the following files off the server via SCP or similar:

/etc/openvpn/easy-rsa/2.0/keys/clientcompany1.crt
/etc/openvpn/easy-rsa/2.0/keys/clientcompany1.key
/etc/openvpn/easy-rsa/2.0/keys/ca.crt

You also need the ca.crt file from

There are additional files (.csr — Certificate Signing Request) that you can just leave there.

If you’re wondering about what files are private and what are not, see here: http://www.openvpn.net/index.php/open-source/documentation/howto.html

Press Ctrl + F and search the page for “Key Files” to skip straight to the relevant table.

When the files are copied off, you’ll want to install OpenVPN on the client. It doesn’t matter if it’s a windows desktop or another Elastix server.

Assuming you’re using an Elastix PBX as a client, install OpenVPN from the Dag repos, or if you’re using Windows then get the OpenVPN-GUI packages from: http://openvpn.se/

Now you’re going to SCP the clientcompany1.crt & .key file in to /etc/openvpn

port            1194
proto           tcp-client
dev             tun8
remote          vpn.servername.co.nz
tls-client
ca              ca.crt
cert            clientcompany1.crt
key             clientcompany1.key
ns-cert-type    server
comp-lzo        yes
pull

This is going to tell the client to connect to vpn.servername.co.nz which you’ll need to update with the appropriate IP Address or domain name of your VPN Server.

That’s it, you’re now all good to restart the service and connect!

7) Testing

You should now have your client connected. What you want to do is ping from the server to the client and the client to the server. Then if you connect multiple clients, try and ping between them. Provided there’s no firewall in the way, and that you’ve run this on the main VPN Server:

echo 1 > /proc/sys/net/ipv4/ip_forward

You should be able to ping from client to client.

Once you have established that pings are working, you should then be able to follow through with other traffic, such as IAX2 trunks, HTTPS access to the WebGUI, SIP traffic, anything you can shake a stick at.

If you cannot ping, make sure that you’re successfully connected and that both your clients and server have an IP Address. Many times you’ll find that during the connection process it will log its reason for failing, and that’s generally what prevents things from working is the actual connection.

Also make sure you’re using the VPN-internal IP Address to ping, not the external IP.

8 ) Last words

IP Addresses for clients start at .6 (Such as 10.99.88.6), and increase by 4 every time. So the second machine to connect will be 10.99.88.10

If you run into troubles, feel free to post in the blog comments.

If you have a success story, please also post, it’s great to know the time taken to write this has had good results.

Look forward to hearing from you

Cheers

Chill.