This error that I came across today typifies what I hate about error messages that I see in various programs.
I was trying to change a drive letter in the system management console in Windows Server 2008 R2. I had upgraded my drive to a new drive and when I restarted my Windows Server VHD installation, it had reassigned all my drive letters. So I tried to set the Drive D drive letter to Drive L (back to the same drive letter it used to have). I had previously changed the Drive E to Drive G without an issue. But when I tried to make this change, the error message popped up saying "The parameter is incorrect".
What exactly is a normal user supposed to make of that error message? Surely we developers can do better than this. (I'm using the collective "we" here to include the Microsoft Windows Server developers).
What was the real problem? It was that Windows had decided, to allocate a page file onto that drive. I had no idea it had done so. But because it was using the file, it wouldn't let me change the drive letter for the drive that was holding the file. Changing the placement of the paging file followed by a reboot, then allowed me to change the drive letter back to what it used to be.
But surely, there's some point in the code where the real problem is detected and surely "we" could surface a better error message than "The parameter is incorrect".
This message reminded me of an error that was common in VB6 days where the system would say "insufficient memory". I've lost track of the number of users that I've seen trying to add memory to machines to fix that problem. What the user was supposed to have instead interpreted from the message was "You have moved or removed a DLL that I was depending upon". The person who wrote the error message has obviously decided that the only reason that we can't load a DLL that used to be there is that we must have run out of memory. Surely "we" developers can do better than this. The next time that one of us feels the need to expose a message that says "The parameter is incorrect" or "Unexpected error" or "Catastrophic error", etc. etc., please can we spend a few more cycles thinking about how to surface something more meaningful that would give the user some chance of understanding the issue?
Hope this helps someone. After upgrading to IE9 and SQL Server 2008 R2 CU7, I found that I couldn't access Report Manager anymore. I found that even though I was in the BUILTIN\Administrators group, that I had to specifically add myself to the Content Managers role in Reporting Services for the root folder of Report Manager. (I have no idea why as it used to work ok) NOTE: I had to do this with UAC turned off, otherwise, Site Settings, etc. were not visible. Remember to turn UAC back on if you did this temporarily.
But the bigger issue was that Report Builder 3.0 would not launch, even after I got the button back on the screen. The message was that I needed to install the .NET Framework 3.5, even though I already had .NET Framework 4.0 installed. Nothing I did seemed to fix the problem.
In the end, the issue is that Report Builder 3.0 will not launch from Report Manager in IE9 unless you enable Compatibility Mode (normally used for older web sites that don't render properly in IE9) for the Report Manager web site. Once I set Compatibility Mode, Report Builder 3.0 launched as expected.
Hope this helps save someone else a bunch of time.
I've published another SQL Down Under podcast. This time it's Mark Tabladillo discussing SQL Server data mining. It's available now at: http://www.sqldownunder.com.
At our company we teach quite a lot of classes and that means we do a lot of printing. So, we decided to move up to a really serious printer. Whenever we go to a high-end print shop, they all use Fuji Xerox Docucenter printers. So we thought we should get one.
These are truly amazing printers/copiers. The print quality is the best available. The print speed is awesome (and just based on your budget). The capabilties are exactly what we need. The printers are renowned for their reliability and the price of the printers reflects their high-end status.
However, given a company claims to have the best product available, charges a price that matches those claims and also says that it almost never breaks down, how long would you imagine they would warrant it for? 12 months? 24 months? 36 months? 5 years?
I was a little stunned to see that they cover only the cost of parts for 3 months. Yes, that's no misprint. It's 3 months.
I am assured this is "normal" for this segment of the industry. Am I alone in finding this more than a little odd? A small car manufacturer would warrant their similarly-priced product for 3 or 5 years today.
I understand that most people buy these with some sort of ongoing service agreement but is there any other industry where you sell items that cost as much as a car and only warrant it for 3 months? What message does that really send about the manufacturer's confidence in their own product?
I had a day of mixed emotions today.
I (and I assume most of the world) have been horrified by this "religious" massacre in Afghanistan: http://www.theblaze.com/stories/two-beheaded-florida-quran-burning-triggers-massacre-at-un-office-in-afghanistan/
Today, however, I also had the reverse experience and a seriously uplifting one. This afternoon I drove through Marysville. For those that don't know it, it's a town that was wiped off the face of the map by bushfires in Victoria a while back. Hundreds of people died. I was roughly in the area and so I made a point of popping into Marysville to see how they were coping. In general, I find that the people in those towns don't want hand-outs so much as business. So I try to visit them and spend some cash in their shops, etc.
But it was the people in the bakery cafe that impressed me today. I saw a tin-can for donations on the counter. I wandered up to look at it. I thought it was a good idea to contribute to helping the locals. What impressed me most was that the collections were for flood victims in Queensland and Victoria, not for Marysville at all. Given what a mess they are currently dealing with themselves, that's seriously impressive and uplifting to see.
There was a discussion on our internal mailing list today about how to get a list of reserved words for SQL Server. It strikes me that there should be a system view that returns this. It could also return details of the version of the product where the word was added and an indication of if the use of the word is deprecated.
If you agree, you know the drill. Vote once, vote often 🙂
This one is a bit annoying. When you apply SP1 for Visual Studio 2010 (VS2010), one of the side effects seems to be that you lose Intellisense in SQL Server Management Studio 2008 R2 (SSMS).
If Intellisense matters to you, you might want to wait for a cumulative update to fix it.
Details here: https://connect.microsoft.com/SQLServer/feedback/details/650569/ssms-2008-r2-is-losing-intellisense-after-installing-visual-studio-2010-sp1
Great to see Jacob and the guys updating their T-SQL Quiz. It's free and it's fun. Get involved:
Since the removal of the 8KB limit on serialization, the ability to define new data types using SQL CLR integration is now almost at a usable level, apart from one key omission: indexes.
We have no ability to create our own types of index to support our data types. As a good example of this, consider that when Microsoft introduced the geometry and geography (spatial) data types, they did so as system CLR data types but also needed to introduce a spatial index as a new type of index. Those of us that need to work with the product as it's supplied can't just create our own new types of index objects.
What would have been far preferable would have been for the ability to create user-defined indexes to have been added to the product and for spatial indexes to have been one instance of that.
Other database engines (such as Oracle) have this capability. This makes it impossible to migrate applications that use Oracle Data Cartridges to SQL Server in an effective way. It also just makes the creation of data types in SQL Server that much more limiting than it could be.
One alternative is to promote properties of CLR data types via persisted calculated columns and then index those but that's somewhat awkward and more importantly, doesn't really do the same thing.
If you'd like to see user-defined index types be considered in the future, you know what to do. Vote once, vote often 🙂
Just posting this blog tonight hoping it might save someone else a bunch of time. For call recording on Skype, I use a program called Pamela. Lately, when I'd first installed it, it would work fine. Later, however, it would come up and say:
"Another application (Pamela.exe) is attempting to access Skype, but we are unable to respond".
You just have to love these sorts of messages that don't give you the slightest clue about what the problem is.
While I saw the problem with Pamela, it can happen with any Skype add-in. The problem actually is related to UAC. If Skype is running as non-admin (startup app in windows) and you launch the other app as admin, you'll get this error message. Amazingly, running the add-in without admin privileges makes it work. The irony is that I was trying it as admin because I thought that would make sure it works.
Hope that helps someone else as I certainly wasted endless hours trying to work out what was wrong. And a big raspberry to the Skype guys for the "useful" error message.