Sun’s university challenge

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

Last week I blogged about Microsoft’s contest for the New Year and here’s Sun’s – a university challenge to encourage programming students to create a new application or port an existing application to the Solaris platform. The winner gets $5,000 and their university gets $100,000 to spend on Sun equipment.

There’s even a list of ideas for projects – and after just a few weeks of using the platform I agree with many of these (like a QuickTime player – Apple should have come up with one for Linux/Unix by now – after all, isn’t OS X based on Unix?).

Of course, some of the ideas are just plain laziness on the part of Sun – they should really be writing their own bug fixes and creating new device drivers, but then again, now that Solaris has become open source, maybe that’s the whole idea…

Windows Vista – how original?

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

I know that imitation is reckoned to be the most sincere form of flattery, but this morning I was listening to episode 37 of the This Week in Tech podcast and my ears pricked up when the guys referred to a video doing the rounds on the ‘net with Bill Gates’ Windows Vista CES keynote over the top of some Mac OS X demonstrations.

I’ve tracked it down (on MacLive.net, although that’s not the original source of the video), and it appears that it was so popular the creator has made two follow-ups:

They’re very funny!

I’m not a Mac user, but I have previously expressed doubts about the Windows Vista interface (codenamed Aero). I also commented last month that the Aero interface seems to be a mix of the Windows XP Luna interface with hints of Apple OS X and KDE. It’s a fine line to tread between plagiarism and a familiar user interface but personally I don’t like any of those big icons.

Right now, my favourite interface is Sun Java Desktop (based on Gnome and borrowing heavily from Windows in its aesthetics and operation, but much “prettier”). I guess it’s all a trade off between user familiarly and innovation, but then Microsoft has always been good at buying other people’s ideas and then promoting them.

Useful command for controlling Windows services

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

Back in 2004, I blogged about some new commands in recent Windows releases and yesterday, I came across another one – sc.

More details may be found in the Microsoft Windows XP Professional product documentation or by entering sc /? in a Windows XP or Windows Server 2003 cmd shell.

Updating the FTSE 100 symbol for Microsoft Money 2000

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

I use Microsoft Money 2000 to manage my home finances. It’s old (read out of support) and a bit confusing when it comes to investments, but it suits my needs and I don’t see any need to upgrade to the latest version right now.

One of Money’s many features is the ability to download stock quotes from MSN Money but mid-way through last year this stopped working properly, resulting in the following message when attempting to update online quotes:

Online Quotes and Exchange Rates:

Money was unable to retrieve a quote for FTSE (London). The symbol $US:FTSE is invalid. To correct an invalid or expired symbol, go to the Details for that investment from the Portfolio and correct or delete the symbol.

The problem is that versions of Money up to 2001 do not allow the investment symbols to be edited. A search of the microsoft.public.uk.money newsgroup turned up some useful information (along with lots of ranting about how Microsoft should fix this – they should, it’s easy enough but hey!) but with help from the advice in one particular thread, I got things working again.

  1. Firstly, I backed up my money data file (*.mny).
  2. Then, I downloaded a hex editor. The evaluation version of the one mentioned in the thread only handles 10MB files (mine was 13MB), so I used Cygnus Hex Editor (v2.50).
  3. Next, I used the Find Data function to search for the old symbol of 55 00 53 00 3A 00 46 00 54 00 53 00 45 (U.S.:.F.T.S.E.). In my data file, this appeared twice.
  4. I replaced both instances of the old symbol with 47 00 42 00 3A 00 55 00 4B 00 58 00 00 00 (G.B.:.U.K.X…) and saved the edited data file.
  5. After relaunching Money, I was able to successfully download FTSE data.

The less technical method of tracking the FTSE is to create an investment to watch, using the $UKX symbol. Whilst this works, it is quoted in pounds (not pence), so is inaccurate by a factor of 100, and doesn’t seem to allow comparisons (for instance, it is often useful to compare a stock against an index).

Now all I need to fix is that investment I made in March 2001, just before the markets took a tumble…

Stock market performance

Apple launches Intel Macs (and new iPod accessories)

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

So, Apple finally launched the new Intel-based iMacs yesterday. These use Intel Core Duo processors (formerly codenamed Yonah), which I think means they are still 32-bit, at a time when the rest of the PC world is finally starting to wake up to 64-bit technologies. Still, at least it’s a step in the right direction (and dual core should whip up performance nicely, although I do doubt Apple’s claims of a doubling in performance simply by going dual core, unless each core is also much faster than the old PowerPC processor).

Of more interest to me are new iPod accessories – like the iPod Radio Remote. Shame it needs an iPod Nano or Video and I’ve only got a Mini.

Telnet issues on Windows Server 2003

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

I’ve spent a good chunk of this morning trying to get the Telnet service working on Windows Server 2003. By default, this service is disabled and configuring the Telnet service to start automatically and then starting the service is straightforward enough, but when I tried to connect, after supplying username and password details, my clients received the following message:

Failure in initializing the telnet session. Shell process may not have been launched.

Telnet Server has closed the connection

Although the troubleshooting advice for Telnet at the Microsoft Windows Server 2003 TechCenter appears reasonably comprehensive, it didn’t detail this message at all; however Microsoft knowledge base article 309523 does – but only for 64-bit versions of Windows. My telnet server is also a domain controller, and as the advice involves demoting and promoting the server, I chose not to try it on a 32-bit version of Windows (a Google search revealed plenty of anecdotal evidence that it doesn’t work).

I had been trying to establish the connection from a Unix client, so thinking that was the issue, I tried telnet locahost from the server itself but received the same result and checking the application event log revealed various entries similar to the following:

Event Type: Error
Event Source: TlntSvr
Event Category: None
Event ID: 4049
Date: 11/01/2006
Time: 10:00:40
User: N/A
Computer:
servername
Description: Error in creating CMD process. System Error: A required privilege is not held by the client.

For more information, see Help and Support Center at http://go.microsoft.com/fwlink/events.asp.

Following the link in this event led me to check for the presence of %windir%\system32\login.cmd

One post on the microsoft.public.windows.server.general newsgroup looked hopeful, suggesting that the Secondary Logon service also needs to be started; however, when I checked, the Secondary Logon service was already running on my server. I restarted the Secondary Logon and Telnet services just in case (that made no difference) but the newsgroup post had got me thinking about the credentials used by the Telnet service (by default, this was NT AUTHORITY\LocalService). I changed the service to start using the Local System account, restarted the Telnet service and it started accepting connections. Just to be sure, I changed the credentials back to NT AUTHORITY\LocalService (which does not need a password) and restarted the service, breaking Telnet again. That confirmed that it was related to the service’s logon credentials and I went back to using the Local System account.

I’m not sure what the security implications of this hack are. Pretty severe I imagine, as membership of the TelnetClients security group doesn’t seem to make any difference to whether my users can logon to the server using Telnet or not, but at least I can get a console connection to my server from a Unix client now (my Windows clients can use RDP).

Unknown hostname for Solaris 10 DHCP client

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

When I installed Solaris 10 x86, my computer thought its hostname was unknown. Aside from being annoying, this seemed to be causing a few issues, so I set about trying to set it to a name of my choice.

Using the uname -S hostname command set the hostname for me but this information didn’t persist on reboot. A bit of googling turned up various references to editing /etc/init.d/network so that it read /etc/nodename and set the hostname accordingly (as well as a script to set the hostname), but my system didn’t have an /etc/nodename file.

I understood that /etc/nodename should contain my computer’s name, but didn’t know if any other settings were required (I later found Jeff Hunter’s TCP/IP quick configuration guide, which confirmed that the file just contains the computer’s name – in my case laptop3).

It turns out that these hacks are for Solaris 8/9 – Solaris 10 is quite happy to set the hostname based on the contents of /etc/nodename. Once I had created /etc/nodename and rebooted, /etc/hosts read:

#
# Internet host table
#
127.0.0.1 localhost
192.168.7.106 laptop3 # Added by DHCP

and the computer was no longer anonymous!

Understanding ‘net lingo

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

Some time ago, I linked to the parent’s primer to computer slang and the netiquette 101 for new netizens. Well, as I dive ever deeper into the world of the ‘net, I’m constantly coming up against acronyms that I don’t understand and variations on smileys/emoticons that I haven’t seen before. That’s where the Net Lingo dictionary of Internet terms has helped me a few times recently, providing acronyms, text messaging shorthand, the most extensive list of emoticons I’ve ever seen (as well as assicons and boobicons – I’m sure you can work out what they are!) and a substantial dictionary of Internet terms.

Resources for Solaris newbies

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

A few weeks back, I blogged that I was trying out Solaris 10 on a spare PC. I haven’t spent a lot of time with the product yet, but so far I’m impressed. I’m sure I’ll be writing more as I get to grips with Unix – an operating system family that I haven’t used in anger for many years.

I also began to gather a collection of Solaris hints and tips, ‘net resources and the like. This is my list so far (some of these are written for SPARC but most should apply to the x86 version too):

Some of these I found myself, others have been recommended by Unix sysadmins and architects, who by and large seem keen to help a Microsoftie learn about Unix (actually, I did a fair amount of Unix system programming at uni’, but that was a long time back). The other comments that I got back were that Linux is also “grown up” contender these days (and I’ve been suitably adminished for suggesting otherwise)!

Indeed, a couple of people have suggested that the best way to learn Solaris would be to start out with one of the main Linux distributions to get the hang of the start-up, configuration and shell. Ubuntu seems to be well recommended, as is Mandriva and SUSE. Apparently the Ubuntu route allows trade up to full Debian (“expert class” when you get there). Knoppix is an even easier way in – basically Debian on a live CD – impossible to mess up! Following that, Solaris is a case of working out what’s different. One sysadmin who had been through this journey commented that he now uses Gentoo at home “100% roll your own, compile everything from scratch”.

I did actually have a play with a couple of the Unix/Linux live CDs last week. One was eLux NG embedded Linux but far more impressive was the Sun Java Desktop live CD that Alex gave me – running Morphix 0.4 (based on Knoppix).

Getting back to Solaris, printed reference manuals (i.e. books) seem thin on the ground (although as can be seen from the list above, there is plenty on of material the ‘net). One which has been recommended to me is Solaris 10: The Complete Reference but I haven’t looked at it myself and the Amazon reviews for this are all very poor, commenting that this is really a Solaris 9 reference with very little new for Solaris 10 and suggesting that some other Solaris 10 books should be considered instead.

Mystery ADSL upgrade

This content is 20 years old. I don't routinely update old blog posts as they are only intended to represent a view at a particular point in time. Please be warned that the information here may be out of date.

Having written earlier this evening about getting started with ADSL, one of the services that I use is the ADSL guide speed test. When we got our broadband connection at home back in 2003 (primarily for my wife’s business, but I also frequently work from home), 512Kbps was the fastest available ADSL connection. We haven’t knowingly upgraded since but recent tests suggested that the connection is delivering 1800-1900Kbps anyway. Being suspicious of the ISP’s own speed tests I tried the ADSL guide test instead and over the last week or so I have consistently recorded results similar to the following:

Results from broadband speed test recorded on Friday, 6 January 2006, 19:27
Downstream
– Actual speed: 1803 Kbps (225.4 KB/sec)
– True speed estimate: 1947 Kbps (including overheads)
Upstream
– Actual speed: 239 Kbps (29.9 KB/sec)
– True speed estimate: 258 Kbps (including overheads)

It’s interesting to note that the actual speed figure is the amount of useful data that the connection can transmit/receive per second, whilst the true speed figure includes an approximation of data overheads (estimated at around 8%). Whichever figure is used, it looks like the line has been upgraded to around 2Mbps and, as we’re not paying any more than we did for 512Kbps (although prices have dropped considerably over the same period, so what we pay is about the market price for 2Mbps), I haven’t checked with the ISP in case it’s all a mistake!