Google and IPv6 Day

June 8, 2011 is the official worldwide IPv6 day. This “special” day is intended to increase awareness and there by adoption of the IPv6 protocol.

In admirable support for IPv6 world day many major sites, including Google and Facebook just to name a couple, are enabling IPv6 on their main sites, for 24 hours. These aren’t IPv6 only but rather IPv4 and IPv6 in what is referred to as a dual stack configuration. This way, the sites will be reachable via both IPv4 and IPv6 protocol users. So, why they are only doing it for 24 hours seems a bit of a mystery. But, that’s not the point of this post.

Google, in anticipation of IPv6 day is running tests in advance. They are putting up a banner at the top of www.google.com announcing the big day and offering a link to test your connection. This “banner ad” worked on the Net Codger and he clicked the link just to see what it was all about. After all, the Net Codger already knew well that there was no IPv6 connectivity between his machine and the outside world. Neither the Codger’s firewall/router nor his ISP provide IPv6 support at this time.

So the Net Codger was quite surprised to see Google announce a successful IPv6 test between the Net Codger and Google’s test site. Opening up Wireshark, a network protocol analyzer, the Codger repeated the test to see if perhaps Google was implementing an IPv4 to IPv6 tunnel of some kind. But, analysis showed that it was all IPv4. There was no IPv6 involved.

So, what’s going on here? Did Google screw up the test? Is this some sort of trickery from Google? Whatever it is, it is all very strange.

Playstation Network Down – I’m Warning You

Playstation 3

On April 20, 2011 the Sony Playstation Network(PSN), utilized for Playstation online activities, went dark. Again! Those attempting to get onto the PSN were greeted with the intentionally misleading message that they had logged out of the PSN. Not that they had failed to login or that they had been kicked but, rather implying that they had chosen to log out. Repeatedly attempting to login and encountering this message finally leads to the realization that it is in fact, the PSN that is down.

The next day a Sony blog post said that the outage may last a day or two. At the end of the fourth day, the network is still down and there is no word from Sony as to when it will be back up. Unsurprisingly, conjecture is flooding the forums with claims of hacking, server outages, claims that it is related to Amazon Web Services(AWS) outage, a denial of service attack by the group Anonymous and so on. No one really knows what is going on and Sony is tight lipped about the matter.

This outage highlights, yet again, one of the issues the Net Codger has with today’s rather expensive online gaming systems. Today’s games, for the most part, don’t have any means of playing them offline. So if the game’s servers are down or are retired, you can’t play the game you paid for. Or worst of all, as is the case for the past few days, when the console network goes down, none of your games can be played. Want to play Call of Duty? Denied. Team Fortress 2? Not today. Just got the new SoCom 4? Too bad. Portal 2 came out Tuesday, the day before the outage began. Forget about the co-op features.

Of course, owners of other gaming consoles are using this as an opportunity to say that their system is superior or at least that Sony is inferior. Whether or not there is any truth to those assertions is irrelevant. These systems and games work in the same way and therefore share the same vulnerabilities.

You, the consumer, need to realize the risk and demand better from the console and game vendors. Suppose you had a 30 year old Atari 2600, you would still be able to use it and play the games with no reliance on Atari itself. But, in 30 years or probably much less, you will not be able to play your Sony Playstation and its games because Sony won’t care to maintain the Playstation Network or the game servers required for today’s game play. Some argue that this won’t happen but, Sony has already changed its mind and forcibly removed preexisting features from consoles in people’s living rooms. Shuttering the PSN won’t be given a second thought when the time comes and the same is true of Xbox Live.

Update: Sony got around to making a formal announcement on the sixth day of the PSN outage. Read their blog post here. While they avoid coming right out and saying it, it doesn’t take much reading between the lines to see that PSN was hacked in a major way and only a fool would doubt the leakage of credit card data.

Many might think that they are safe since it’s a free service and doesn’t require a credit card. But, Sony has been collecting credit card numbers all along, even for the free service. They claimed it was for age verification. Don’t even get me started.

Sony has become a disaster and all by their own actions. When the lawsuits flood in due to identity theft and credit card fraud from this breach of their service, it will simply be their just deserts.

Cavalry SSD Disappointing

15 months ago, I purchased an ultra small fanless Intel Atom based PC to test on a small PBX project. Silence and durability were the critical components of this project and I was going to use a CF card for the “disk” drive in this PC. The application required only read access for the OS and a small partition for logging and voicemail recordings. Low disk I/O access and utilization, to be sure. Not wanting to worry about write wear on the CF card or an additional drive just for logging, I decided to try out one of the “new fangled” Solid State Disk(SSD) drives. They were supposed to be super fast and, with no moving parts, much less prone to failure than a standard spinning disk drive. Silent operation and low heat were bonus features.

I spent a few dollars more than I should have on a Cavalry 32GB 2.5” SSD drive. At the time it seemed like a neat unit with built in SATA and USB 2.0 interfaces right on the drive. The unit has been in production since then and besides slight disappointment in the speed of the drive things have been working great and the project has worked out well.

This morning, voicemail stopped working. Checking the PC I found a disk error that I was unfamiliar with. I found lots of references to this error online where it was most often referred to as being transient and rebooting usually resolved the issue at least temporarily. But, when I rebooted, I got an error that was all too familiar; Operating System Not Found. Bah!

Further checking lead me to find that the drive wasn’t even being seen by the PC at the BiOS level so, I pulled the drive and put it into a known good PC to eliminate the possibility of a motherboard controller error. Nothing. No drive detected. It seems like it won’t even power on but, it’s impossible to say as there is no power indicator on the drive and being SSD there is no platter spin up noise as there is on normal drives. Regardless of whether it is the drive controller or a power issue inside the drive, it’s failed.

Drive failures are an unfortunate fact of life. I had really hoped that I would be less likely to see drive failure with an SSD but, it happens and I’m willing to accept it without too much frustration. So, I contact Cavalry regarding at least getting a warranty replacement. Contact it seems is only possible via their web interface as there doesn’t seem to be a phone number for Cavalry support anywhere. That’s when the penny drops. This drive is 15 months old and is therefore 3 months outside the one year warranty. Sorry. Have a nice day.

I’ll get another drive for the PBX but, I won’t be buying anymore Cavalry drives!

Babysitting Progressbars

On a recent project I needed to do a migration from Microsoft Small Business Server 2003 to a Microsoft Small Business Server 2011 server. This newly released server suite has numerous installation requirements including a big CPU of 4 or more cores, lots of memory, lots more disk space, on site installation, internet connectivity(!) and at least two hours of just waiting for the initial installation to complete.

It was this waiting that got me all codgerly again. Sure, you can start the install and go to lunch or what have you and come back in two hours. But, the install can fail at any moment for any of a bazillion reasons. So, you start the install walk away and three minutes later, it fails. You come back after two hours to see that it failed due to some trivial detail and you must now start again, waiting another two hours and hoping that it doesn’t fail again!

Now in all fairness to Microsoft and this particular software, the above scenario is common with lots of software, especially operating system(OS) installs or upgrades. It got me thinking, why don’t the software installers have a notification mechanism?

It is already required that the system be connected to the internet, why not include a small utility in the installer that will send an SMS alert upon success or failure of the install. During the initial install wizard it could pop up a screen just before the initiation of the install asking for the cell phone number of the installing user. Then you click START and go to lunch or for a nooner or whatever. If the install fails prematurely, it sends you an SMS message stating as much so that you can return early and resolve the issue.

With such a simple feature added to the installers, you could initiate the lengthy server install process and move on to doing something more productive than babysitting a progress bar without worrying about premature failures going unnoticed. If all goes well it’s done when you return. If it fails prematurely you can address the matter immediately rather than wasting two hours only to start it all over again.

I thought these kids knew everything already. So, how come none of them have implemented this as a standard component in all OS installers?

IPv6 – Imminent Disaster or 2010 Bogeyman?

While there can be little argument about the certainty of IPv4 address exhaustion, there is continued speculation as to when it will really become a problem. While IPv4 exhaustion is surely more imminent than peak oil, some people argue that citations of exhaustion occurring 10 weeks from today is hyperbole generated by manufacturers wanting to sell more equipment.

Indeed, IPv4 addresses are being gobbled up at a historic rate but, let’s face facts, it is occurring in a very wasteful fashion. Properly managed the exhaustion could be delayed for a long time forward. Were the assignment of public addresses to cell phones and refrigerators reduced with increased reliance on the dreaded NAT, 10 weeks could turn in to several years.

The shrill pitched alarm about IPv4 exhaustion usually implies the implosion of the internet. ‘Total collapse a certainty!’ But, at the eventual point of total and complete IPv4 address exhaustion, the internet will continue to work just as perfectly as it always has. You simply won’t be able to add another publicly addressed device without taking an existing one off. While this will be very inconvenient, it is a much less scary proposition.

The question is; is IPv4 address exhaustion an imminent disaster or is this just a bogeyman like Y2K? I think the answer is both. The exhaustion is certain and difficulties are ahead. But, they are further out than 10 weeks and the impact won’t be felt by the very vast majority of internet users.

Has your broadband ISP said anything at all to you about it? Could you even get IPv6 service from your present provider, if you wanted it now? Probably not. And IPv4 exhaustion will be a big problem for them, long before it is any problem for you.

RoadRunner Returns To DNS Hijack Tactics

Back in early 2008 RoadRunner, the Time Warner/Bright House Networks ISP started a new policy of redirecting failed DNS queries to their “helpful” search and suggestions page. It was a hijacking tactic that was increasingly popular with ISPs at the time. It allowed them to monetize all those typos and non-existent domains that people were looking for. As if they weren’t making enough money off the service already.

Needless to say, most people were so enthralled with MySpace and YouTube that they didn’t know or care about it and just blindly accepted it. But, there were quite a few people savvy enough about the workings of the internet to realize what was happening and they didn’t care for it at all. Indeed, such a ruckus was raised that RoadRunner implemented an opt out page where you could go and opt out of the “fine service” that they were providing. This was enough to calm most folks and the world continued to turn.

Over the past week, I’ve been noticing a lag or delay on RoadRunner networks. Whenever I entered a new hostname into a browser or any other application, there was a new and very pronounce pause before the hostname was resolved. I wasn’t sure what was happening and frankly, my first thought was that spyware might be the cause, despite the very low likelihood. But then I got a couple of these pages in my browser.

After being annoyed by a couple of these redirects, I realized that the DNS hijacking service was enabled and that this was likely the slowdown or lag in hostname resolution that I had been noticing. So, I went to the opt out page and opted out. It was easy, the link was right there on the redirect page. And done.

Everything was fine again. That is, until the next day when the DNS hijacking was enabled again. Repeatedly, I opted out and every time, I was opted back in the next day. While it is possible that this unauthorized opting back in is a technical issue, I can’t help but think that this is RoadRunner’s new policy. A way to force users back into using the service and increasing their revenue.

Now, I do realize that there are ways of avoiding this issue. One can use a different DNS service or even setup their own DNS server. Indeed, it might come to this in my case. But, why must I pay for service to have it slowed by underhanded tactics intended to further increase the ISP’s revenue at the expense of my already under-delivered performance?

Net neutrality in the making folks.

Boxee Committing Suicide

Have you ever heard of Boxee? It’s a company that produces media center software. Through a joint effort with D-Link they are about to release a hardware bundle called Boxee Box.

Pre-orders for the, as yet to be released, Boxee Box are presently being accepted on Amazon for $229. Silly names aside, that’s a pretty gutsy or stupid move for Boxee. Especially since Apple has announced that their second generation Apple TV product, which offers much of (though not all of) the same functionality as Boxee Box, will go on sale very shortly for $99! Also, already shipping devices from Western Digital and Roku are also priced below Boxee Box.

But, don’t be too harsh with Boxee and D-Link about the bad pricing. They were probably too distracted by the atrocious design that they came up with for Boxee Box to be worried about naming and pricing.

Do these guys really believe that people want yet another set top device that does not fit in with the rest? One that is obtrusive and aesthetically displeasing in most living rooms? What else have they failed to consider? Does this thing have a fan that howls like a banshee too? Or does the Boxee logo light up in a brilliant and distracting LED green?

Sorry Boxee, you took a ‘me too’ product idea and applied abysmal execution. You and D-Link seem to be the only ones that don’t know where this will end.

Fish Cam

I’ve played around with closed circuit television(CCTV) and security solutions for a few years now. Nothing professional, just hobby and home security stuff. On a recent holiday trip to the tropics, I thought it would be entertaining to try some underwater CCTV. So, I bought a cheap Chinese underwater video camera from a company called Panvigor, via an eBay auction.

This small camera claims to be water tight and includes built in LED lights for night or low light use. It seems specifically designed for being towed behind a boat. Although I never tested it in this fashion, I suspect that it would waggle under tow and be nearly useless. Instead of towing it behind a boat, I opted to strap it onto a pole three feet below a dock for some quick and easy testing.

As you can see, the results were excellent. The above picture is a snapshot I took of the underwater image being displayed on a 46 inch LCD television inside the Codger’s holiday residence. The fish cam was very entertaining and was enjoyed by all who saw it. This particular image shows an inquisitive French grunt, that was probably about 3 inches long, mugging for the camera.

Considering the pleasure we got from this camera, it was well worth the price. But, all was not perfect with this camera. First, because of its 6 millimeter wide angle lens, objects need to be quite close in order to see them well and when they are close, they look huge. As is the case with my diminutive friend above.

The built in LED lights were a disappointment. There was no way to turn them off, even during the day and I suspect that the close proximity of the LEDs to the lens likely caused some washout of the picture color. Worse yet was that at night, it attracted a swarm of millions of near microscopic copepods directly in front of the lens.. These copepods showed up very well in the light as they swarmed the lens. At dusk, as they first started to arrive, the image would get “snowy” as the light reflected off of them. But as darkness set in and the copepod swarm reached its peak, the picture was a complete whiteout as they obscured the camera’s view completely. An external light, a few feet away, would have been much better for evening viewing.

Within a couple of days, algae was starting to grow on the camera’s lens. This caused a clouding of the image. The effect was nearly imperceptible due to the slowness of its onset. But, wiping down the lens with a wash cloth every three days would reveal a noticeably clear, sharp and colorful picture again.

Sadly, the time came for me to leave my seaside lair but, I left the camera behind for others to enjoy. I was informed that it lasted about three months before water finally penetrated its defenses and it stopped working. It was a great little toy and I’ll probably get another one for my next trip unless I find a better one at as reasonable a price.

openSuse 11.3 – Evolution 2.30

I missed an appointment yesterday. Hardly Earth shattering but, it was annoying. How could it have happened, I had it on my Evolution calendar and I should have gotten a notification. Double checking my calendar to be sure, I noticed that there should have been several reminders that I had not seen. I decided to test the notification system by creating a test appointment a few minutes into the future. Unsurprisingly, the time came and went without any notification.

This launched an exploratory mission into why Evolution alarm notifications weren’t working. After a bit of wasted time, I found that the evolution-alarm-notify service was not running. I started it manually at the command line and up popped several alarm notifications. I thought I had figured it out. Apparently, evolution-alarm-notify must have died for some reason. Just to make sure that was all, I decided to restart the system to make sure that evolution-alarm-notify would come back up properly and I wouldn’t have this problem again in the future. But, it didn’t come back up. It didn’t even try to come back up. So, now I had to figure out how it was supposed to be started and where the problem was.

As it turns out, Evolution has changed the way that evolution-alarm-notify is started several times over the course of a few versions. Sometimes it was started as a service and sometimes it was started as a sub process of the main application, which would seem to be the most logical. But, with Evolution 2.30 the service is supposed to be started by the FreeDesktop XDG autostart facility. This means that a evolution-alarm-notify.desktop file in the autostart directory is supposed to kickoff the evolution-alarm-notify service when users login to their desktops.

Checking the /etc/xdg/autostart/ directory on my openSuse 11.3 system I saw that there was indeed such a file. But, upon examination of that file, I saw that KDE was excluded from the desktop environments that were supposed to run the service. I felt that this omission was odd and the lack of online complaints about it lead me to suspect that it was not a common or known issue with the distribution.

To test that theory, I ran a couple of new installs in virtual machines. I changed the installation options a little bit each time to see if there was a particular combination that was triggering the issue. But, the problem occurs no matter what you do. In all cases of new openSuse 11.3 installs, the setup of autostart for evolution-alarm-notify excludes KDE. It’s obviously a small bug but, I’m surprised that the issue hasn’t been more widely discovered. I thought that a lot of people were using Evolution.

Once you know what the problem is, the fix is fairly straight forward. You need to edit the file and enable autostart under KDE as well. The following bash shell commands will do this for you:

sudo vi /etc/xdg/autostart/evolution-alarm-notify.desktop

Then edit the file so that the OnlyShowIn line looks like it does below:

[Desktop Entry]
X-SuSE-translate=true
Type=Application
Name=Evolution Alarm Notify
Comment=Calendar event notifications
Icon=appointment-soon
Exec=evolution-alarm-notify
Terminal=false
Type=Application
Categories=
OnlyShowIn=GNOME;XFCE;KDE;
X-GNOME-Bugzilla-Bugzilla=GNOME
X-GNOME-Bugzilla-Product=evolution
X-GNOME-Bugzilla-Component=calendar
X-GNOME-Bugzilla-Version=@VERSION@

Press the escape key. Then type
:wq
to save the file. Finally, restart the system and your Evolution alarms will work as expected.

openSuse 11.3 – Bit Me Again

I guess I’m a glutton for punishment. No matter how many times I’m disappointed by openSuse, I keep coming back to it for more punishment. My latest issues are with the latest version, openSuse 11.3.

Now, let’s be fair, overall this is a fine distribution. It is feature rich and it mostly just works. Some of my biggest issues with it can be attributed to taste, or the lack there of, and KDE 4. But, I’ll avoid those issues for now and concentrate on real issues.

Issue number one, immediately after installation, is the lack of any graphical means of configuring the video card and monitor settings. The openSuse folks decided that the Sax2 utility, formerly used to graphically configure the video card and monitor settings, wasn’t very good. Which was very true! So, they decided that it should be removed and perhaps it should have been. But, they couldn’t be bothered to provide any form of  replacement. This means that if Xorg doesn’t successfully detect and automatically configure all your video settings, which it most often does, then you must manually configure your settings in the obscure, convoluted and completely absent xorg.conf file.

That’s right, I said absent xorg.conf. The fact is that by default, xorg autdetect does not create any xorg.conf file so, you can’t even do a simple edit to at least get a VGA frame buffer session going, let alone a graphical configuration tool to do the job. With openSuse 11.3 it’s all or nothing. Either your system automatically sets up your video settings just right, or you are up the proverbial creek without a paddle.

Now, the die hard xorg users will be quick to jump in and start saying that you simply need to run xorg -configure to create a basic xorg.conf file that you can then copy into the xorg directory and edit as you need. Simple!(?) But, the fact is that xorg.conf is somewhere that no human should ever have to go and, until the removal of Sax2, they mostly never had to. Now, with the release of openSuse 11.3, we advance backwards from graphical video tools to arcane text file configs that will make even the seasoned veteran weary. For the new user that encounters video problems, abandon all hope ye who enter here.

Distributing openSuse without a graphical means of setting up the video card and monitor is a regression that is as shameful as the feature free regression release that was/is KDE 4. Oops, I said I wouldn’t do that.

For those that refuse to give up, here is a link to a openSuse knowledge base article on how to setup a video card. I wish you the best of luck and promise that, if you get the video working properly, the rest of the distribution works well and you’ll likely enjoy it.