This morning it looks like a majority of ISP’s using BT’s Wholesale Broadband Connect went offline due to a failure in one single exchange - link here to my ISP status page.    Exact scope of outage is still uncertain.   What is known is this affects ISP’s who interlink to WBC in Docklands, which will be a lot of ISP’s as most link in Docklands due to the cheap comms in the Redbus/Telehouse buildings there.     Wholesale Broadband Connect is used to link ISP’s to the 21CN and to exchanges providing the BT ADSL2 products.   Ie it connects ISP’s to the exchanges that have been upgraded to 21CN.    BT have you not heard of Single points of failure, and how you should design them out of your network?  The ISP I use, Andrews and Arnold are particularly affected with a total loss of service on customers who have been migrated to the 21CN this week.   It thankfully isn’t affecting all their customers as all have not been moved yet as it depends on your local exchange - I personally am unaffected thankfully.     BT will apparently be releasing a press release on this outage today to explain the reasons, which I do await with anticipation.

This problem really makes ISPs connecting to BT from 2 exchanges from 2 POPs look quite silly - as what’s the point in the ISP having a resilient link to BT when BT’s network itself is not resilient?

Further to this BT has been annoying a lot of smaller ISP’s, especially mine by changing their plans  with the WBMC product - which to explain is a product that allows ISP’s to remove the legacy old style BT ATM links into the ADSL network, and instead use one link to cover users on ADSL2, and those on exchanges which have not been connected to the 21CN yet.     By the looks of the new rollout announced, a lot of exchanges will be 21CN before the WBMC can even be used, which questions the need for the product (Q2 09).    For small ISP’s this is a big deal as the BT central links used to provide service to users present one of their biggest cost bases - price per MBit into the BT cloud is way higher than cost of transit to to the internet as a whole…

My ISP are one of I believe 2 in the UK who are currently offering native IPv6 provision.   Its useful for those of us who want to test real world Ipv6 from home prior to general adoption on the Internet.  All  ISP’s will have to use Ipv6 in the future due to address space depletion in the Ipv4 world..

Over the past few months a few of the users of A&A noticed that when using native IpV6 over L2TP to A&A small Ipv6 packets do not get through in some circumstances due to what looks to be a firmware issue in some Cisco equipment used in the BT 21CN and general ADSL platform - with BT’s equipment silent dropping the packets L2TP as they transit from the ISP back to the customer.     L2TP is used to pass a PPP session from a customers ADSL line directly to an ISP’s termination equipment, from which it goes onwards to the Internet.    BT should not be interfering with L2TP traffic.   BT have claimed to my ISP  this will not happen with the 21CN, but it shold not be happening even on the old service in my opinion at least.

A&A raised a ticket with BT to get this fixed a few months ago but have got no timeframe to fix this problem as it stands currently.

Users of AAISP doing research found the bug looks to be a IOS firmware issue within the BT network affecting certain pieces of Cisco kit used within their network.   Full (very technical) details can be found on this forum thread.    A workaround for now is to tunnel IPv6 in IPv4, but this does have drawbacks, namely a smaller Maximum transmission unit (MTU) - meaning you get less data per packet due to the encapsulation overheads.

To directly quote one the owner of A&A (Adrian Kennard) this morning in a online chat with myself:

“AAISP feel that it is incorrect for BT to be interfering with any PPP packets and have been trying to get BT to address the issue for months. We believe the root cause is an old (2 year) cisco IOS bug for which there is a fix”

“We are still trying to get a clear response from BT as to whether they even consider it a fault or not”

“We appreciate that if they agree it is a fault it may take time to roll out IOS upgrades, but would like a clear statement from BT as to what they propose doing”

With the advent of 7Digital now providing DRM free music in 320kbit/sec mp3’s (See this article for full details) from all 4 major labels, I’m confused about the iTunes offering in AAC.

You see, I like iTunes plus music, ie the no-usage limits, no DRM music in iTunes, and prefer this to buying a CD….however its not always available.  In fact only one major label is signed up.

Nowadays, my choices when buying music are… use play.com, amazon.co.uk or similar for physical media.    Use iTunes if I want it now… but have to accept the DRM on the majority of tunes.

Admittedly for me iTunes DRM doesn’t get in the way much, but it does stop me playing music on a non Apple device, so I only spend £10-15 a month there normally, for the few single tracks I can’t get in Album format from the aforementioned retailers.    DRM is important to me as at some point I’ll will be getting a Squeezebox Duet, or Sonos device - so I am becoming all too aware of how my “singles” library is proprietary to Apple only playback.   Currently I resort to waiting a day or two and getting most music albums via the post and CD method, upon which its ripped to 320Kbit mp3 VBR format, and then added to my library.

7Digital changes this completely.     Now I don’t have to wait for albums, I’ve checked their prices and they are very reasonable, in fact cheaper than the shops in many cases.   I can see myself using 7Digital instead of physical media now for almost all purchases so I congratulate them for getting the labels to agree to this change.

Apple, take note - I’m voting with my feet and will move my single purchases to 7Digital also.     I will come back to the iTunes fold, if and when you make all tunes Itunes plus.

Over the weekend (on Friday night) I upgraded firmware to 2.1…

1.  Battery life is indeed seemlingly slightly improved by 20-30% when on standby.   Still not enough with my usage levels for a 2 day runtime though.
2.  The Genius playlist selection on device is very very nice.
3.  iPhone appeared to be more responsive when web-surfing.
4. I know its only visual, but the GPRS (dot), and 3g symbol are now more in keeping with the rest of screen (rather than being bold).

Overall, I’m very happy with the update, as the Genius playlists in particular show why Apple is king of mp3 players.     Over the weekend - I just selected one track in genius, and it played me music I liked for the entire 2 hour car journey - I  would rarely choose these tracks in a playlist of my own making.   It included one track I didn’t even recognise but really liked - causing me to look at phone to check (thankfully I wasn’t driving!).

Genius is better than shuffle because shuffle does tend to put a hard rock track into a mix where you want just “chilled music”.   Genius just works, and I’m very impressed with the results.

How are you finding Genius?    Is 2.1 working as well for you as for me?

Nokia, still my second favourite handset manufacturer have decided to release a mobile that comes with “free” music downloads for 12 months after purchase.   This guardian article goes over the fine print better than I can.

Now my points here are varied so bear with me…

1. This has to be limited - theres GOT to be a fair usage snarfu - despite the unlimited rumblings in the article linked above.   If these isn’t Nokia are foolish… they must be being charged per item by the music companies - as the download continues to work past 12 months.    So if someone wanted to download all 2.1 million tracks… could they?   If they did, I imagine it would cost Nokia a considerable amount more than the phone, as even at 10c a track (v.low compare to what Apple pays on itunes I believe), that would cost Nokia $210,000 if my maths is right!
2. More to the point why wouldn’t you download every single track? -especially as you’ve only got 12 months of unlimited service.   If you don’t you lose access to the one’s you didn’t download!
3. Phones not available via operators, only sim-free.   Key point… who pays for a phone nowadays if they are on a contract?    I can see the point in pay and go… except again its not an operator pay and go deal, you can only really get as a sim-free.
4. Free downloads only for 12 months.     I can see why the operator’s didn’t like this - They prefer 18 month contracts for one, also it competes with their own music service, which are pay per track generally.
5. 2.1 million tracks isn’t a huge variety.     iTunes has over 5 million at least.    Even the Three music store had a similar variety (which I found very limiting whilst I was a customer)

The guardian is spot on, I can imagine this also risks Nokia’s usually good relationship with the carriers as it competes directly with their own profit making services.

Nokia think that if you buy a handset every 12 months its good for their profit, but due to reasons above I don’t think it will be that good for them, especially if they get a few customers taking advantage of their generous offer

Personally I won’t be buying one, but I’ll be interested to read about the 1st person who does try to download all the items in the catalogue.   Being unlimited, someone really should try this !

What are your thoughts on this deal?   Will you be buying a pay and go access phone to get 2.1 million music tracks free of charge?

Admittedly you can only use the tracks on PC and phone and it probably uses Microsofts awful DRM.    But at the prices quoted its  still way cheaper than buying all that music on itunes

I started a new blog last week based on my exploits in EVE online…   really because this blog isn’t really the place for it - despite my recent mini-review, the politics and trading exploits in EVE are NOT the place on an IT blog, and I’m sure the editors here would promptly tell me off if I was to post all EVE content.

Anyhow I wanted to see how many readers I was getting, where the viewers were from etc.   Reason, purely my interest, as I want to see who’s reading my content, as a blog with no readers is surely not a blog but a journal!
I installed both Google Analytics and Feedburner on my blog.    GA to analyse the breakdown of visiting users, and Feedburner to actually track people subscribing to the RSS feed.   So I posted some content, and  publicised in a few locations, and I finally am seeing statistics.

Of the 2 services, currently I prefer Feedburner.   Google Analytics has a 24 hour delay on its statistics it seems, so thus isn’t as useful up front.   However Google Analytics does have extra detail on referrers to name just one item, which I’m sure in time I’ll find interesting - however it doesn’t give the immediacy Feedburner does.

Feedburner for example shows me who is reading the RSS and when, from where, and is instant updating- for example today I know I had 2 visitors from Denmark, though the majority of my readers are in the USA.

So… my question to you guys out there is:    Is there a feedburner like real time analysis platform for web hits - thats invisible on the page like GA is?

I totally agree with the Conservative policy mooted in this Register article.

Being a member of an IT security team you realise that user-education and actions are what invariably lead to data loss… and a problem with users is their apathy and reluctance to change.

If you tell a user to do data transfer in this “secure” manner - you’re safe, If you use your old process you risk going to jail. I think this one change would focus their minds quite well..

Users in large companies sometimes do try and hide behind the “process” shield, instead of challenging a potentially risky insecure data request from a client/partner in many cases…

For example I still see users internally who are unaware that email is by nature an insecure medium - of course unless a secure pgp or s/mime link is setup in advance of the email being sent…

Thankfully we now have technology in place to spot and stop many such instances from occurring now (in email at least), with the email’s in question being redirected to compliance instead of the end-recipient so they can be educated as to proper data transfer methods. Of course the technology isn’t perfect, user education is the main thing here, and legislation and personal responsibilty for loss is the good thing.

So, hats off to the conservatives - a step in the right direction.

I’ve posted about this before in comments to this post, but I think it deserves a seperate update post:

1 month and a few weeks in, I’m finding the iPhone 3G battery life to still be sufficient for what I use my phone for. There is, however, noticible difference in battery left at end of day if I don’t micromanage (ie turn off/on) the relevant chipsets on the phone. My major gripe is there is no Office/Work/etc profiles to control what is turned on… Doing it manually can be a pain. Right now I have the below setup:

At home: Need Bluetooth, Wifi on, and 3g off (non 3g area)

At office/travelling on train: Need Wifi off, BT off, 3g on

In car: BT on, 3g off, wifi off

Providing I change settings as above, in my usage - yesterday I got (starting coming off charge at 6:30am):

1. 1.5-2 hours of mp3 playback (during half of this was also surfing web on train (mix of 2g/3g)
2. 1 hour of websurfing and email (I have 2 email accounts, one push, one checking hourly) via 3g
3. One 2 hour phone call (2g)

At end of day (midnight), I still had ~ 40% of the battery free and  I realised this isn’t that bad.  Without micromanaging the battery the day before, I had 5% (an estimate) - with an almost identical usage pattern.

In comparison my old Nokia N73 (also 3g) could only handle 2 hours of intensive web surfing/video use - a few more hours if only being used for mp3 playback… and it had a FAR smaller screen. The N73 also ran out of battery with a just being on standby and a 3 hour call in fact…

I think peoples expectations with the iPhone are that it’ll perform the same as their old phone and this is the problem. On their old phone they didn’t spend a minimum 2 hours of the day websurfing, emailing etc (and it has a smaller screen!).   They were, in many cases just using it for calls. My experience on Three with the N73 made me know that a 3g phone with very intensive use does need 2 charges a day. This surprisingly isn’t the case with iPhone for me - at least right now.

My only problem is I may have to get a Morphie and a second plug-in charger to keep me going in the future if I continue increasing my phone usage every day.

iPhone 3G users - have you come to same conclusion as me - in that the battery life issue is mainly due to increased usage of a converged device - not the phone having awful battery life?   Does yours last the day without the micromanagement I do?

Well, after many years with zero anti-spam technologies (and manual deletion of ~ 200 items a day) I decided it was time to move my mail host and implement anti-spam technologies.

Now I already have a home SAMBA server, running Debian, which also acts as a mini desktop. I decided to use this as my mail volume isn’t huge… I get ~20 valid emails a day, ~200-500 spams depending on the day of week really.

SpamAssassin looked to be the premier anti-spam solution out there for Linux, and I selected a Debian EXIM integration. Took a while to learn exim, but I’m now mostly impressed with the configuration. I’ve used dovecot as a IMAP server. All these are the standard Debian stable packages……

Basic procedure for me was I installed the packages - then I followed this guide and got a basic system up and running… and moved a “test” domain name to point inbound SMTP at the box so I could then fully test all the options and tune the anti-spam.

Tricks the above guide missed:

Using CPAN (perl -m CPAN -e shell) to install Net::DNS. Without this vital step Spamassassin missed out on ALL DNS tests, which are quite good for scoring.

Bayesian filtering.

• Set this up to use a system wide database, in a folder you control with world read/write access. The default isn’t right.
• You may wish to increase the default size of the bayes database. I increased mine 10 times.
• It seems to require 200 spams and 200 non-spams to be learnt before its operational - at first I did not realise this. I fed Bayes a folder of 2000 spams, and let it read my (already filtered of spam) archive of personal mails as non-spam (3400 items). This trained the spam filter quite well.. I used a variation of this script
• If you run sa-learn with -D for debug it does tend to show faults in your SA config.
• Increasing score of BAYES_99 for me at least results in better results.
• I’ve set up learn as spam folders in my mailfile, which is learnt and deleted every 6 hours (ie mails making it through SA I drag to this folder).

Setting SpamAssassin up is NOT easy, and requires a lot of tinkering to get runnign as you want (hence my playing with a test domain). Once complete however, its an brilliant system in my opinion at least.

Now its up and running, only 4 spams have hit my mailbox (though I’m still storing all spam - aim is to not store very high scoring spams in future, and only store “uncertain” results. Though right now, with ~5000 spams not hitting my mailbox I’m a happy bunny.

SpamAssassin is also available as a windows version I believe. For Exchange users with nothing else it may be worth a look.

Quick post:

I updated last night to the new 2.0.1 iphone software… and it fixes a few of the bugs in 2.0.0 - namely the browser/other apps crashing regularly and randomly (this isn’t a huge issue in browser as it restarts and usually goes back to page you are looking at) - but since installation, this morning, I had zero crashes in one hour of surfing on the train to work.

So overall, I’d say if you have an iPhone, you want to install 2.0.1 asap.

After day one of Cuil, I’m not super impressed

Half the UK day the site was down due to lack of capacity… guys if you want to beat Google you need to allow us to actually search… after all the primary reason people use Google is its search is quick, pretty accurate.. and seemingly always available.

Cuil, however, well, not so good on day one.   Several searches I ran came up with no results (on my name for example, which is very common - what are the chances of Google not knowing about a Dan Jones (at least one!))…. this may have been teething problems, but isn’t a good start.

In positives however, when I did try other searches, the results were slightly more accurate than Google.     The problem overall from my day one experience is the engine either comes back with good results… or nothing.

I’ll continue playing with Cuil today, and report my findings, but overall I’m having to stick with Google right now…

Do you like Cuil?