Async Gotchas in C#

If you ever had “strange” problems with async/await, the Async in C# and F#: Asynchronous gotchas in C# might have an answer.

TypeScript IntelliSense

We are going to use TypeScript in our product, so imagine my disappointment when I realized that IntelliSense is not working for TypeScript files in Visual Studio 2012, despite it should.

I have tried everything I could – played with VS settings, uninstalled and reinstalled TypeScript, ReSharper, whole Visual Studio… Nothing. Finally, I had to reset my Windows and start from scratch. After two full days completely wasted trying to get IntelliSense working, I finally got to the magic point of having fresh Windows, Visual Studio, and TypeScript. Started it all up, and bang – no IntelliSense again! After pulling some hair out, and crying quietly in the corner, I started second round of uninstalling and installing TypeScript, including older version 0.8.2, Web Essentials, resetting VS settings, etc. Same result. Finally I checked my Visual Studio plugins, and, to my surprise, spotted MySQL Tools for Visual Studio extension, despite I did not want to install it in the first place when MySQL was installed earlier. Removed this plugin, and voila, TypeScript IntelliSense at its best!

What is strange, I have tried to install MySQL server and its Connector/.NET at home (although a newer version 5.6.10.1 package), and surely got the unwelcome MySQL Tools for Visual Studio extension with them. But! The TypeScript IntelliSense works nevertheless, despite a completely different picture was observed at work twice. Go figure.

So, if TypeScript’s IntelliSense is not working, I would recommend to remove all extensions, starting from those more exotic (e.g. MySQL, and those less used by wide population) – it might actually repair the thing.

Troubles with Xbox 360 and Live

Was installing today my second Xbox 360. Everything went smoothly in the beginning: Xbox calibrated, asked its questions, found WiFi, connected to Live, and downloaded an update from it.

The next step would be to download my profile, and this is where the sucking part started, as it turned out that the Xbox cannot connect to Live. Connection test was able to find network and internet, but not Live. I have tried everything – restarting, unplugging power, restarting router, reading forums, jumping on one leg, singing shaman songs, beating the drum, standing on head. Nothing. Then I noticed that the DNS configuration on the Xbox has only primary server set, and it is pointing to the router, which has no DNS server of its own. Changed DNS to manual and pointed it to Google’s 8.8.8.8 and 8.8.4.4 servers. Nothing. Changed from WiFi to Ethernet, rebooted a few more times, changed DNS back to automatic, changed back to WiFi, and bingo – Live test succeeded. No idea why though, as Xbox settings now were identical to the initial factory settings, and those were not working before. Go figure. And all these while my second Xbox, with absolutely identical configuration, was happily working in the other room, and had no problems with Live whatsoever, including passing tests!!!

Fine, Live sorted out, so now it was time to download the account. Meant to be easy. Well, it was not. After entering email and password, I was present with the list of my accounts, selected the needed one, the download process would start, go about 10%, and stop with error message (80151909 and 80151103, temporarily cannot download profile). Retried it a few times (each time you have to enter both email and password using shit-fuck-crappy-on-screen keyboard, fuuuuuuuuuuck!). Someone from Microsoft recommended to authenticate the Live account, but the account was not asking for it. Then there was recommendation to temporarily change the Live password. Changed it, rebooted Xbox, and managed to download the profile. Finally!

The last step was to get back my old password – I have it on my Windows Phone, and some other devices, so changing it everywhere would be rather painful. Well, the password change page said I cannot reuse the original password, and that I must choose a new password each time. As stubborn as I am, I started changing the temporary password by changing one letter a time, and then retrying to set my original password. Something like 5-7 retries later I was able to reuse my old password, and the fairy tale got a happy end.

Aftermath: working Xbox with Live, and two hours of my precious Saturday time miserably wasted. Thank you, dear Microsoft! Can somebody tell me, how a normal person can set up a new Xbox with the internet connection and Live? He/she cannot, you say? Well, that is what I think too. Context: I am a software developer, and have dealt with computers for 27 years now, since I was kid.

Visual Studio Behind Proxy: Expectation Failed

We have a transparent proxy at work, and it was causing troubles in Visual Studio 2012 when going to the Extensions and Updates. Contrary to your expectations, “The remote server returned an unexpected response: (417) Expectation failed” message was appearing. The fix is pretty easy, and involves editing devenv.exe.config file in the Visual Studio 2012 folder (normally C:\Program Files (x86)\Microsoft Visual Studio 11.0\Common7\IDE\). Somewhere towards the end of that file there is a

<system.net>
    <settings>
        <ipv6 enabled="true"/>
    </settings>
</system.net>

fragment, which has to be fixed to read

<system.net>
    <settings>
        <ipv6 enabled="true"/>
        <servicePointManager expect100Continue="false"/>
    </settings>
</system.net>

That’s it.

Problem with SugarSync’s AutoSync on Android

I have all auto sync categories (i.e. folders, photos, and videos) off in settings, but SugarSync Android app periodically wakes the tablet up for syncing, despite there is nothing to sync. The interval of these wake ups is the one specified in the auto sync settings (which are disabled, i.e. grayed out).

This results in unnecessary battery drain (my setting for the period was 1 hour, and it “ate” about 10% of battery overnight).

I have switched the syncing interval to 24 hours, so I will see in the evening if it alleviated the issue as a temporary workaround. The real fix must be done in the SugarSync app – to stop waking up when auto sync is off. Reported this to SugarSync, let’s see what will happen.

LocalDB and NHibernate

We’ve got a request to support SQL Server 2012 in our product at work. Well, the LocalDB (coming with Visual Studio 2012; a rough equivalent of SQL Server Express of the past) is practically SQL Server 2012 I thought, so it was reasonable to do initial testing on it. We use NHibernate 3 in our system, so I went to the .nhibernate.config file, and changed the connection.connection_string property (the Data Source part of it, to be precise) from

Data Source=localhost\SQLEXPRESS

to the logically sounding

Data Source=(localdb)\Projects

and… It did not work. A bit of googling pointed me to a solution: named pipe connection. To get its name, run

sqllocaldb info <InstanceName>

in the command line, which in my case translated into sqllocaldb info Projects. The output will contain the Instance pipe name line, and the value of it is the data source name you are after. So, in my case the correct Data Source read

Data Source=np:\\.\pipe\LOCALDB#5E0F8FF8\tsql\query

(as far as I understand, usually only the number after LOCALDB# will change).

Windows Phone 8 SDK + VirtualBox = Problem

Just installed the Windows Phone 8 SDK and got disappointed when running the sample app on the phone emulator failed with the following cryptic message:

The Windows Phone Emulator wasn’t able to create the virtual machine. Something happened while creating a switch: Xde couldn’t find an IPv4 address for the host machine.

Some people suggested that the problem might be related to the VirtualBox being installed on the machine (I had it). Uninstalling VirtualBox helped, although it is a bit annoying, as I need VirtualBox to access my work VPN.

Toolbox Scrollbar Disappeared in Visual Studio

My colleague just hit into a problem with his Visual Studio 2012 – the Toolbox (the one keeping UI controls) lost its scrollbar. The scrollbar was appearing if he was filtering controls by name, and disappearing again when filtering was off. After searching a bit on the net I found that people were seeing the same problem in Visual Studio 2010 as well. The workaround Barney Nicholls, that post’s author, proposes for Visual Studio 2010 involves deleting toolbox*.* files from the Visual Studio settings’ folder:

1. Close Visual Studio.
2. Delete toolbox settings using the commands below.
3. Start Visual Studio.

The commands (script) to delete the toolbox settings (those files are normally hidden, thus we need to remove the hidden file attribute before):

attrib -h %userprofile%\appdata\local\microsoft\visualstudio\10.0\toolbox*.*
del %userprofile%\appdata\local\microsoft\visualstudio\10.0\toolbox*.*


Needless to say, that this worked for Visual Studio 2012 as well – just substitute 10.0 with 11.0 in the commands above. And, of course, you can perform these operations in Explorer, FAR, or any other file manager as well.

PS Just occurred to me, that there is one more problem, similar in its nature, but with different symptoms – IntelliSense stops working after some actions, e.g. renaming some function. Usually, deleting the hidden .suo file with the same name as the solution .sol file and located in the same folder as the solution file resolves the problem. Of course, Visual Studio must be closed when performing this operation.

Microsoft Message Analyzer

Microsoft Message Analyzer (currently in Beta) has been released to the public. It is a smart successor to the Microsoft Network Monitor that hosts lots of cool new features, like parsing and validation of protocol messages and sequences, user-controlled on-the-fly grouping by message attributes, re-assembly of- and rendering of payloads, etc. It has also a separate blog at Technet.

Windows 8: First Experience

Yesterday, finally, I installed my new SSD drive, and this was a good opportunity to bring Windows 8 into the picture. So, just a few notes on what to expect:

1. Installation was really simple. Regional settings, and serial number was all info requested. It was really fast too, although it is difficult to say whether it was much faster than Windows 7 to SSD.
2. The thing is booting mega-fast. Both cold boot, and Windows 8′s strange-own-way-of-kinda-hibernating-without-calling-it-like-that are fast; to the extent that most of the cold-boot time is spent by BIOS. No idea how MS made it, but it is great!
3. The new Start screen and desktop are not that horrible as people claim they are – at least I cannot see any large problems with them. Should I say “so far?” :) Of course, they require certain re-learning, but that should not be too big of a problem. If people handle phone calls on Android phones (and gosh, Android’s UI of phone call related functions is really horrible; but that is a totally different story anyway), then they must be able to master the new Windows UI. :)
4. Installing Visual Studio 2012 took just a few minutes (again, is it because of SSD? who knows), and the thing starts in just a few seconds. Fantastic!

Of course, not to be called Microsoft-fanboy, or whatever we are called nowadays, I must mention some problems:

1. Power management sucked badly out of the box on my computer. It was not able to wake the computer up from sleep (nothing was helping, not even THE power button on the PC itself); then, after some tweaking, I was able to wake it up, but the screen was black; then I was not able to put it to sleep; then… and so on and so on. Took me good few hours, but now it seems to be quite OK, although I am still not completely satisfied. What helped me, was turning hibernation off (powercfg -h off in admin mode, to those interested), and then tweaking power options (e.g. not allowing my mouse to wake up).
2. Metro media players suck, IMHO. They do not support DLNA sources, at least as far as I can tell after a short inspection, and they are missing some vital function, like zooming. Wasted effort, IMHO.
3. My poor old AMD Radeon HD 4850 PRO video card has the same issues as it had with Windows 7 – hardware accelerated video playback results in black screen, and there is no way to fix it other than reboot. The same PC worked fine with XP – problems started with 7, and now 8 is following 7th footsteps. I understand that the video card is nothing fancy (although it is still better than 90% of video cards in today’s notebooks), but then again one would expect such a basic stuff like DXVA/DXVA2 to work. On the other hand, AMD has always had problems with video playback (and those problems were coming and going in various patterns with each new driver), so nothing new, in a way. And yes, it seems to be almost impossible to sort out the issue; while I was able to use DXVA checker to somehow “fix” media player by disabling hardware acceleration, online videos e.g. at Store still crash.

All in all, Windows 8 looks rather great, and I am big fan of Metro style in UIs (i.e. clean data-centric approach) – it is sad that they did not replace all Windows UIs, e.g. Control Panel and Explorer, with it. Most of the problems I faced are very likely related to my old-not-really-supported hardware, e.g. AMD is not supporting my video card in new Catalyst, so if you have newer hardware, you will probably have much smoother experience.

Buffalo WBMR-HP-G300N

Some time ago I have moved from my ancient arthritic ZyXEL Prestige ADSL modem to a shiny new Buffalo WBMR-HP-G300N wireless ADSL modem/router.

I am very happy with ADSL part: it has worked out of the box, and there is really nothing no complain about, as it is clearly light-years better than my old crap was. The same goes to its wireless part – got my XBOX, laptop, tablet and phones on-line in matter of minutes.

But… (here goes the famous “but”) … I cannot say that I am so excited about its other features, namely NAS, DLNA streaming a.k.a. Media Server, and BitTorrent.

First of all, NAS supports only FAT, FAT32, and XFS as file systems. We are in the age of 2-3 TB hard-disks, so FAT and FAT32 suck badly. XFS? What the fuck is this fuck? And who gives? So, this was my first disappointment, as my Buffalo USB3 hard-disk had NTFS and a lot of absolutely needed files, but there is no fool-proof data preserving way of converting NTFS to FAT32. It took me two days, and a lot of swearing, to get it to FAT32. The process involved copying files around for backup, formatting the disk to FAT32 (using Acronis tools, as Windows will not format 2 TB FAT32 partitions on its own), copying files back, etc. The speed of NAS is also abysmal – barely getting 5 MB/s on large files on the wired 1 Gbps Ethernet.

The second disappointment came when I have switched the Media Server feature on, but it did not appear on network. Buffalo’s documentation sucks big time – it is simply a collection of screenshots of the router’s web console, and some text repeating what pictures say. In 90% of cases the explanation says something obvious, but useless. For example, if there is a check-box saying “Feature XYZ Enabled” the documentation will say “Check to enable feature XYZ“, but there will be no further explanation regarding consequences and prerequisites. Kind of “try and see“. There is really no troubleshooting information on the internet either, neither on Buffalo site/forums, not in the wild. Wasted something like 6 hours trying to get the damn thing working, until finally I got an enlightenment, that the problem might be in the media itself, namely its quantity – I have something like 25,000 photos, and then some music and movies. Quick trial, and I got my movies served. So, now my media server is getting not the whole harddisk for himself, but only its Movies folder.

And, BitTorrent function just went into the same basket – that shit works only and only if the harddisk’s file system is XFS!!! :)

Good job Buffalo!

Overall, I can recommend this modem for its basic functionality, but then again, taking into account how crappy its advanced features are, there are much cheaper modems without these functions. Rating – 3 out of  5.

Alt+R

Alt+R key combination/shortcut stopped working in all my programs. I thought that I will go mad, as this shortcut is quite frequently used, e.g. in FAR Manager for seeking for files, etc. Investigation revealed that the offending program was Google Chrome, or its Chrome Toolbox extension specifically – it has Boss Key configuration, and that is “eating” Alt+R system-wide. So, remapping that one to something else, e.g. to Ctrl+Alt+Shift+R, solved that issue. Google should allow to disable these shortcuts – otherwise it is really quite painful.

ASUS Transformer TF101 Problems

My ASUS Transformer TF101 started to have problems after the Android 4.0 a.k.a. Ice Cream Sandwich a.k.a. ICS update: occasionally, when sleeping, it can automatically switch on and go into a reboot loop, or switch on and get stucked in the ASUS boot screen; only log power button press, or, sometimes, long volume down button press, can “resurrect” the poor device. This could happen with or without dock being attached. The latest update to Android 4.0.3 (got it today in the night) did not fix the problem. Annoying.

Till today I have tried to do many things to it, but without much success. Today I have made some deeper investigations, and here is what I have found.

One thing is that the device is not really going to what is called deep sleep when going to sleep mode, as touching the mousepad would wake the device up, although pressing keyboard buttons would not. This behavior was not the present in Android 3, so I started to look for a solution. Found it on a few forums (have no links to them anymore, sorry):

1. Go to Settings > Developer options. Confirm the Attention box by clicking OK.
2. Enable the Stay awake option, so that its check-box is checked.
3. Wait a few seconds (I, in fact, went out of settings and came back; just to be sure…)
4. Disable the Stay awake option, so that its check-box is disabled.

Voila – mousepad is not waking the device anymore! To verify, go to sleep, wait some seconds (30?), and then touch the mousepad.

My speculation about what is happening here is that this option is in essence a 0/1 value, 0 being disabled, and 1 being enabled, but, for whatever reason, it gets some other factual value, e.g. -1, after the ICS update, which Android treats as “Oh! I should stake awake!”, while the Settings UI treats as disabled, e.g. because it performs a check for the value being exactly 1.

Also, there seems to be some evidence that NVIDIA TegraZone might have its fingers in the problem as well, as it is checking for updates every 12 hours, seemingly even when the device is sleeping. So, I have switched its auto-update off by:

1. Start TegraZone
2. Go to its Menu > Settings
3. Switch Notifications off

Also, some people say that installing and enabling Auto Airplane Mode by DON helps reducing the symptoms or even eliminating this annoying problem. I have used it on my Transformer since the beginning, and, obviously, it was not helping on its own; but then, again, I believe that it might help in some cases, and it is not hurting to have it anyway: it is automatically enabling the airplane mode when the screen is turned off/put to sleep, thus preserving the battery when you are not using your device. I do not understand why Google did not put this function directly into Android, to be honest.

So far I did have not seen any unexpected reboots, but it is too early to say for sure – they were not happening every day.

UPDATE 2012-03-25 Well, five days went since my “fixes”, and so far so good. CPU Spy shows that the device really goes into the deep sleep mode, and I do not get reboot loops anymore. Battery seems to last OK too. To be honest, once there was a problem that I could not wake the device up from sleep, but rather it booted, but it might just as well be my problem, if I have earlier put the device down so that its Sleep button was pressed and thus it switched off completely. BUT! Today that fucking motherfucker (no other words apply here, I am afraid) again went into automatic boot and stucked at the ASUS boot screen. Fantastic! Nevertheless, five days is much more than before, so it seems that these “fixes” fix at least some part of the problem.

A-PDF Page Crop

Recently I received one PDF document with huge margins – the text itself was located in the middle of the page and its size was about 1/3 of the page. As you can imagine, reading it on-screen is really painful; also you cannot print only part of the page in Acrobat reader, which is IMHO a big omission on the Adobe’s side. Well, I thought that cropping the page would really solve the trouble, but it took me rather much time to find something affordable, simple, and inexpensive (maybe even free – who knows?) – A-PDF Page Crop. You can crop pages both visually in editor, or from command line (useful only if you have lots of similar documents; I was not trying that option) – and it does its job well. Trial version puts its mark on the front page, and that can be removed by getting a licensed version. Nothing more to say – the tool does what it is made for. That company also has many other PDF-related utilities.

IVT Elektro Standard 490 Review

Note This post is a review, but only partially. I mean, it is a review, but it is also mixed with my personal experience, and as such is sometimes emotional. But it states truth… At least how I see it. Also, this is a “living” post – I will update it when the situation changes; normally I would rather do separate posts instead, but product reviews are not my main topic, so I will do it as a single post. And last – I might have some little mistakes with dates (day here-there), but generally it is correct.

Intro

There is not that much information about the IVT Elektro Standard 490 exhaust air heat pump on the net, so in this little post I will try to somehow fill up this information void, and to tell about my experience with the product and the humble company behind it, so that you, dear reader, can perhaps avoid the problems I had/have to deal with. Also, I think, the heat pump is such a product that it cannot be reviewed on its own, but rather it should be reviewed as a complex system comprised of the device, its purchasing and installation, surrounding infrastructure, after-sales issues, etc.

What is an exhaust air heat pump? In a few words, it is a device that extracts heat from the air that would otherwise escape during ventilation, and then efficiently re-uses it to heat the house and to produce hot water for a household. Much longer technical explanation is available at Wikipedia.

When I was building my house back in 2004, I decided to use an exhaust air heat pump for heating. A ground heat pump was another option, but after calculating TCO or Total Cost of Ownership (includes costs of equipment, installation, maintenance, possible problems, etc.) for both pumps and comparing it with possible electricity savings, I could not consider the ground heat pump anymore – I do not remember all the details, but it would take something like 15 to 20 years for that pump to pay back for itself, assuming nothing serious would break during this time. So, the exhaust air heat pump came the clear winner here.

And what about brand? At that time there was not as much choice as there is now, and IVT together with Carrier were pretty much rounding it off – Nilan and others just started to appear in the Finnish market. Carrier pumps were made by IVT, so I’ve chosen IVT.

History

TS-Tekniikka was the official IVT seller in the Turku area, so I turned to them and got an offer that included plans and calculations for the floor heating and ventilation system, all needed hardware like Nereus floor heating, IVT Elektro Standard 490 exhaust air heat pump, and wall air-intake valves, as well as installation and adjustment works.

The pump itself was about 6,000 EUR with 5 years warranty, and the installation and adjustment work was another 750 EUR (plus other hardware, but it is not relevant for this story). Things went pretty much downhill after money has been paid – everything that could be delayed was delayed (delivery, installation, etc.), so I had a few extremely heated discussions with Tero Suutari, the certified asshole managing director of that circus, and his “lovely” wife Krista Suutari, the certified cunt financing manager. In the end the pump has arrived and was installed in place, but, as it turned out, not properly adjusted. (Note TS-Tekniikka is not selling IVT pumps anymore, and IVT admitted to me that they had a lot of problems with that company; so yes, I have the right to call things and people by their real names and earned titles.)

One year later I got my first year-closing electricity bill, and found out that I consumed 10,000+ kWh more during that year than anticipated and explained/promised to me by the IVT sales-guy (close to 1,000 EUR value). Shocked, I have called TS-Tekniikka about this problem, and was told that their pump is perfect and my house must be total crap (well, that was the meaning anyway). After much shouting we came to the conclusion that TS-Tekniikka will do the house checking with heat camera to see if the house is leaking heat. They did it, and guess what – the house was fine. They had to call an engineer from IVT’s Finnish head office, and the guy (he was very knowledgeable, I must admit!) immediately realized that the whole thing was not adjusted at all, but was simply connected to power and water and switched on. The floor heating was not adjusted either. He did all the adjustments, and after that I started to get normal electricity bills: today my annual electricity consumption is about 12,000-14,000 kWh per year, depending on winter and my laziness; 8,500-10,000 kWh is coming from the heat pump itself, and the rest from my own household consumption.

Of course I wanted some compensation – all in all TS-Tekniikka did not deliver according to the contract, it produced rather large extra costs for me, and it was not really pleasant to live in my house during the first year (well, no surprise here). But, in the end, I got nothing. Not even a penny. Talking (shouting and swearing too) to TS-Tekniikka, to IVT in Finland, and to IVT in Sweden proved to be pointless and irritating exercise; people there did not want to neither understand nor resolve the issue – IVT was pointing at TS-Tekniikka (all this while openly admitting that TS-Tekniikka sucks and many people complain about it), and TS-Tekniikka was simply not willing to do anything about it. Suing them for 1,000 EUR did not sound like a very good idea, so I had to “swallow”. As a consequence, my opinion about the IVT, and I mean its Swedish HQ, is: a rather ugly company that is not valuing its customers at all. This ugliness, unfortunately, is putting its mark all over the IVT’s business. Well, the fish rots from the head, they say…

That was five years ago…

Today

Fast forward till today. My heat pump broke on Wednesday, November 3 – three weeks ago. It flashes the red led, and the display says “LARM Moottorisuoja käynn.” which translates roughly to “ALARM Motor protection start-up“. In practice it means that the compressor is not starting for some reason (e.g. it broke, some kind of sensor broke, etc.), and the pump is using electricity only to heat the house and water.

Next day, Thursday, I found the official IVT reseller/repair center in the area (I do not want to say their name yet, despite I am not really happy about their performance), called it, talked with its boss, I will call him Mr. N for now, explained the situation, and he promised to send a repair guy ASAP – on Friday, or latest on Monday. The guy did not come as promised, so on Tuesday, November 9, I called Mr. N to check what is up. He answered that his repair guy thinks that some specific pipe is broken in my pump, so he has ordered it; it might arrive on Friday, and there is no point to come before that. I asked in turn, how is it possible to know that certain pipe is broken without actually seeing the pump first? Mr. N agreed with me, and promised that the repair guy will come ASAP to check it; that promise was kept – the repair guy came the same day, checked everything, did not find any faults, but managed to switch the pump on.

The pump worked well till Thursday evening, and then broke again. I called Mr. N on Friday morning, and the repair guy came on Monday, November 15. He checked everything again, but again could not figure out anything. He promised to check it up with IVT people, and then to call me back.

In the meantime, I have called IVT office in Helsinki, and talked with Mr. L, who is responsible for this part of Finland. He promised to help me to get things done ASAP, as IVT “has strict rules regarding fixing times” – but somehow, that help failed to materialize yet (will have to call him).

After the repair guy has left, the pump started well, worked for about 10 hours, and then broke once more. Not hearing anything back from the repair company, I have called Mr. N two days later, on Wednesday, November 17, to check whether my case is moving ahead; he told me that they decided to replace the compressor relay to see if it helps. They promised to get it by the next Monday, and to call me back then.

Today, Tuesday, November 23, I had to call Mr. N again, and then the repair guy came and changed the compressor relay. It did not help though, and now his (and IVT’s) best guess is that the compressor is broken. He could not say when they can replace it, nor how much will it cost, and promised that Mr. N will call me back (did not happen yet).

The net result – my pump is just 6 years old, out of warranty, and the compressor, one of the most expensive parts, is already broken. The compressor was supposed to last at least 10 years according to IVT sales speeches (but one anonymous service guy told me that it is “balooney”, as the compressor is small, and it is almost constantly on, so six years sounds like rather a realistic life-span). Additionally, I have no idea how much it will cost (numbers in 1,000 EUR to 2,000 EUR range were mentioned by some people). Thanks God my house insurance seems to cover it, so my part will be 150 EUR only. What is not covered though, is extra electricity which is burned now – it is rather cold outside, about -2C..-10C, so about 50 to 100 kWh is used daily by the pump for the last three weeks (more than 100 EUR extra already).

I will keep this post updated as the situation evolves, but so far I am rather thoroughly dissatisfied.

UPDATE 2010-11-25 (1) Called Mr. L in the morning. He promised to check why it takes so long. According to him some other mega-cool repair guy was supposed to visit me already long time ago, but… Let’s see, but so far I have not heard anything yet.

UPDATE 2010-11-25 (2) Mr. N called me! Yippy! They will replace the compressor, and the cost will be somewhere around 700 EUR plus work (guess that 1,000 EUR mentioned earlier is close enough). But, there is a catch – it will happen earliest Wednesday-Thursday next week. And we have -15C, while up to -20C is promised for the week-end.

UPDATE 2010-12-02 FINALLY!!! Today, after exactly a months of pain, repair guys came in the morning, and after about 3-4 hours of work, they declared the pump fixed! That poor machine was cut into pieces (at least some parts of it were), and then welded back together (I have been at work, so I’m writing it from my wife’s words). Gosh, I hope that it will work from now on well (although, to be honest, have very little hope and trust in it).

UPDATE 2010-12-03 First day… The pump became much louder – it makes “pumping-whistling” noise when the compressor is on. It is not very loud, but still much-much louder that before… Also, it sometimes gets slight vibration, but so far I managed to stop it by pushing the front panel a bit side-wise. Both problems were not present before the repair. I guess complaining to Mr. N will not help much, as he will probably just start laughing at me – in Finland nobody seems to bother about such “non-issues”, unfortunately. To me it seems more and more, that these pumps are kind of “single use” devices – once something breaks in it, you will have to change the whole crap very soon, as repairs will just make it behave worse and worse. Also, I have no idea if it performs well after the repair, and there is no way to check/verify it, as there were no measurements made nor any certificate was issued.

UPDATE 2010-12-19 Today came the invoice (the deadline for payment was on 9th of December – no idea where it was till now), and it is for 1,809 EUR!!! Outrageous! For example, they charge 92.25 EUR/hour for work! OMFG, they are just Incompetent Repair Guys (yes, I can repeat it in court, if needed), and not lawyers! There are totally 7 hours of work (1+1+1+4, despite first visits were half an hour long mostly), totaling 646 EUR. Then, they charge 1.78 EUR/km of driving – more than one would have to pay for TAXI! They counted each trip at 30km back and forth, so 4 trips are totaling 214 EUR (by the way, Google Maps says that there is only 26.5 km). The compressor itself is 886 EUR, which is more or less in line with what I was told by Mr. N earlier. So, quick recap before going further:

• Driving, 120km – 214 EUR
• Work, 7 hours – 646 EUR
• Compressor – 886 EUR
• “Small items” – 64 EUR

Well, I think they are simply <CENSORED> me – incompetent repair guy has to come here 4 times instead of 1 or 2, wastes 7 hours to repair the pump, and the whole process takes ONE MONTH. Additionally, I wasted about 1,000 kWh of extra electricity for heating during this time, totaling about 100 EUR or so. The issue is, that other NORMAL repair companies do not charge you for something that they cannot fix. For example, VEHO in Raisio (yes-yes, the expensive VEHO!) did not charge me a penny when they spent two hours unsuccessfully trying to figure out what is wrong with my car!

I think there is an urgent need to uncover the “cast”, for their “minute of Internet fame”:

• The repair company – IVT Center Turku (APPLAUSE)
• Mr. N, managing director – Veli-Matti Forbes
• Incompetent Repair Guy – Jussi? <somebody> (never heard his surname, unfortunately)
• Mr. L from IVT – Teemu Lehtinen

Tomorrow I will have to pay a visit to IVT Center Turku, to talk about this case. Of course, if Mr. N will be at the office. Also I will try to call IVT. Let’s see what will be the end result. END OF UPDATE 2010-12-19

UPDATE 2010-12-20 I called Teemu Lehtinen, and he promised to check what can be done in this situation (I have really no hope that he will help). Also I called Veli-Matti Forbes, and proposed him that I will pay reasonable amount for fixing. I was ready to pay for the compressor, for 3 hours of work, and for two trips totaling 53 km. Everything else is not supposed to be my issue, and he should rather account it to his own team’s incompetence. He answered that there is no way we can discuss the price. No amount of reasoning from my side helped. He said that I did not have to use his company to do the work. When I replied that there is really no choice here, as IVT Center Turku is the only company in Turku fixing IVT pumps, his answer was that that is the reason he can charge what he charges, and I should simply shut up and pay. To which I promised to not let it be and make a war out of it, and he personally can go and fuck himself. That was end of the discussion.

UPDATE 2011-January The pump stopped working with same symptoms. Now I was not wasting any time and directly called Teemu Lehtinen. He was quite helpful this time, and we agreed to skip IVT Center Turku this time. Pretty soon another person (from Rauma!!!) was sent to me, and discovered that all coolant from the compressor is gone – previous repair guy did crappy (paska) job and left holes in pipes. These were promptly fixed, and the pump is working since (now it is August). Thanks to Mr. Lehtinen, it did not cost me a penny this time, and that was nice from him.

Conclusion

First, general pros and cons.

Pros:

• The pump is definitely neater than an oil-based heater/boiler, takes much less space, is safer, and is cleaner, also environmentally. But this is true for all heat pumps, not just for IVT.
• The pump is definitely more efficient than direct electrical heating – annually I am (or should I say was?) using about 5,000-7,000 kWh less than my friend with a similar house but with direct electrical heating.

Cons:

• The whole menu system and the controlling user interface of IVT 490 is a piece of C R A P too complicated – you have to be a thermal engineer to understand what you can adjust, and what consequences it will have. Effectively it means that you are left to the mercy (knowledge) of the installation engineer, and will have really little control over things if something goes wrong.
• User Guide – very, very, very bad piece of technical writing. Or, perhaps, the problem is that it is too technical. Way too technical indeed. There are a lot of graphs that do not say anything to normal people (heck, they do not say much to me either, and I have mathematical-physical background myself). But then things interesting to normal people are not explained at all. For example, what is the difference between Normal and Saving modes? What else… Yeah – English version of the User Guide is nowhere to be found. I was not even able to get it directly from IVT.
• The pump is missing some functionality that I would consider absolutely basic. For example, the floor heating has delayed reaction to heating changes, i.e. it takes something like 4 to 6 hours for the indoor temperature to change once you change the heating water temperature. The problem is, that the pump has no way to account for this “slowness” – it is trying to do its best, but sometimes, especially in changing weather conditions, you can get not very comfortable indoor climate. Another example could be the missing possibility to set different target temperatures for different parts of the day. And so on… (is IVT interested in buying these ideas from me? :)
• I have a fireplace, and it is really a tough job to start the fire without getting tons of smoke into the room (took me a year to master!) – the pump is creating lower pressure in the house because it actively pushes used air out, so the fireplace tends to have back-draft in the beginning. In practice it means that I have to open the terrace door before starting the fire up, and keep it open for a few minutes at least, while the fire is not burning well. This is becoming an issue once you get into low temperatures, as the whole room cools down badly rather quickly (Fuck! Low temperature was the reason why I wanted to start the fire in the first place, wasn’t it?!)
• Sometimes there is not enough hot water in the evening, especially if you have visitors that use shower.
• Not directly IVT’s problem, but remember, that TS-Tekniikka was IVT’s official seller… The fresh air intake system designed by TS-Tekniikka and approved by IVT is fine during summer, but it is definitely not the best during winter: cold air sucked in through the wall vents is really not the most enjoyable thing.

The conclusion? I am not that sure anymore that exhaust air heat pumps are the best choice nowadays, but they probably are. Sales people will tell you that ground heat pumps are much better, but do your math before buying them – they might prove to be much better for seller’s profits only, and much worse for your own pocket. If you decide to go with an exhaust air heat pump, I would strongly recommend to stay away from the IVT 490 pump – IVT simply cannot handle its business well, and the product is rather average, if not inferior. Of course, there is no guarantee that you will fare better with other companies, but at least it is worth trying.

Finnish Keywords lämpöpumppu, arvostelu, ongelma, kompressori, rele.

Managed/Native Debugging in C#/C++

I had a problem debugging my solution consisting of C#, C++/CLI, and native C++ projects. I was not able to get breakpoints working in the native C++ parts – after running the solution with breakpoints set, they were turning gray with warning sign, and their tooltip stated that “The breakpoint will not currently be hit. No symbols have been loaded for this document.” despite all debug properties were set right on those projects and all debugging symbols existed. After fighting it for a few hours and almost giving up, I started to think that the problem might not be with the native C++ projects, but rather somewhere else, e.g. in the StartUp project, which is C#. Checked that project’s properties, and guess what! Debug > Enable unmanged code debugging option was off!!! Switched it on, and voila – breakpoints work fine in native C++ DLL now!!! :)

ADSL Modems and Torrents

My ZyXEL ADSL modem (some ancient Prestige model, too lazy to check now) was causing me a trouble since I remember: it frequently drops internet connection when I use uTorrent. It happens after a few minutes, and the only way to continue is to hard-reset the modem by powering it off-and-on, and then renew the IP address on the PC. Which is very-very annoying, as I use torrents for my work (we distribute software updates using torrents). Some other modems, as well as other torrent clients might have similar issue, I guess. So, today I decided to put an end to this crap and to find a solution.

The problem is very likely on the modem side – it seems to be not able to handle that high number of connections that torrent clients open. The solution is to get your torrent client “behaving nicely”. And here is what I did with my uTorrent:

1. Disabled Resolve IPs in the Peers tab’s context menu (you must have at least one active torrent to be able to get this menu) – absolutely useless feature that allows you to see DNS names of peers.
2. Set peer.resolve_country parameter in Preferences > Advanced to false. Same as above – totally useless feature that allows you to see countries of peers.
3. Disabled DHT by clearing Enable DHT Network option in Preferences > BitTorrent. This step might not be needed, but I am not using DHT now, so it is not harming, but resources are freed.
4. Disabed UPnP and NAT-PMP by clearing Enable UPnP port mapping and Enable NAT-PMP port mapping options in Preferences > Connection (of course, assuming you are not randomizing ports and using manual firewall configuration). This step might not be needed as well, but then again, that ZyXEL modem is not a UPnP/NAT device, so…
5. Reduced Global maximum number of connections parameter to 50 in Preferences > Bandwidth. You can try higher values, but for me 50 works fine. In fact, torrents’ downloading speed increased by about 10% after I reduced this parameter from 200 to 50, while general web browsing performance improved a lot (before these changes it was really painful to browse internet while downloading torrents).
6. Set net.max_halfopen parameter in Preferences > Advanced to 4. Many people say that they have problems after 5+, so the default 8 is definitely a problem.

The result – no more drops! :) To be honest, I think that the last two points should help on their own, but other points do not hurt either.

Virus Scanning: Microsoft-Recommended Exceptions

Microsoft has published on their support site interesting article Virus scanning recommendations for Enterprise computers that are running currently supported versions of Windows. It applies to the systems that experience slow-downs or instability when using virus scanning software (whatever the vendor), and proposes a solution (temporary solution?) for the issue.

Computer Security Identifier (a.k.a. SID) Myth

Would never believe this, if it would not come from Mark Russinovich:

NewSID has been retired and is no longer available for download. Please see Mark Russinovich’s blog post: NewSID Retirement and the Machine SID Duplication Myth

Sounds fantastic, isn’t it? :) I wonder, if one should use reseal on Windows Embedded images in light of this information: reseal is renaming machine (can be done other way), resets some settings, e.g. mount points (can be done other way), and changes SIDs (which seems obsolete if you are not in domain)…

Debugging MSBuild Scripts (VS2010)

I did not actively search for the subject, but once I bumped into a three-part article about the MSBuild script debugging (Part I, Part II, Part III) on The Visual Studio Blog I immediately though that it is worth remembering about it when (I am not saying if, but rather when here) I need it in the future. The blog itself is also worth checking, as there is much information about VS IDE, MSBuild, and extensibility from the Visual Studio development team.

How to Delete Locked Files…

Most likely everyone experienced this on his own skin: you try to delete a file or folder, but cannot, because Windows says that it is in use, despite you cannot figure out which application is doing that, so reboot is the only way to “fix” it. OpenedFilesView and Unlocker are two simple utilities that can help in such situation.

BlueScreenView

BlueScreenView scans minidump files created during BSOD crashes, and displays the information about all crashes in one table. For each crash, BlueScreenView displays the minidump filename, the date/time of the crash, the basic crash information displayed in the blue screen, and the details of the driver or module that possibly caused the crash. Seems to be nice utility for troubleshooting.

Catalyst 10.6 + Flash 10.1 = Kaboom! :x

Children, never ever put these two beasts together in their default configurations, or you will get spectacular crash (a.k.a. BSOD, a.k.a. blue screen; the problem is that usually it crashes so thoroughly that there is no way to even see BSOD) whenever you open a web page with Flash video in it, e.g. YouTube. The workaround is simple – disable hardware acceleration in Flash player. To do it, right click on the Flash player in browser (preferably on the page without Flash video, or BSOD will come and eat ya! just kidding), select Settings…, and then clear the Enable hardware acceleration box on the Display tab (first tab in settings). That’s it! After this fix Catalyst 10.6 + Flash 10.1 – Hardware Acceleration = Nirvana :)

Help in the Visual Studio 2010

Mmmmm… I don’t know what kind of daemon possessed Microsoft usability engineers when they were designing Visual Studio 2010 help system, but something definitely went very-very-very wrong. :( What’s wrong, you might ask?

• It opens in a web browser, which kind of sucks as it breaks my normal work-flow (I have to track help tabs in browser, and then close them, and if I open some new tabs, then help is not in place I expect it, etc. etc. etc.)
• It’s content tree (the thing on the left side) is not really a tree anymore – it just shows top-level hierarchy and then also “path” to current topic. No more, no less. And if you need to go to some other topic, you will immediately lose track of previous topic. So, if you want to “browse around” the documentation – tough luck. What’s even more ugly is that the online help can be at least configured to show full content tree, but offline help won’t allow you any customizations.
• There is no index. Full stop. Yes, yes – N-O—I-N-D-E-X. They say that there is search, but try to find some API related to paths and that you vaguely remember exists and has name PathBlaBlaBla (where BlaBlaBla is some meaningful name) and you will see that the search is as useless as it can only get.

Conclusion: new help system is VS2010 is light years behind that of VS2008 or VS2005.

Can something be done about it? There are at least two partial solutions:

1. Microsoft Help Viewer Power Tool – simply adds keyword index capability to the Help Viewer, with an option to display help in a standalone window.
2. Help 3 Viewer goes much further: it shows VS 2010 help in a traditional help viewer (similar to “old” DExplore) with full TOC and Index, and features multiple document tabs. There are other features too, and it can be set as default VS2010 help viewer.

I hope that these tools will help you with your help experience :) .

Hotfixes for WPF Memory/Resource Leaks

There are three Microsoft hotfixes (967328, 967634, and 981107) that address WPF memory and resource leaks in .NET 3.5 SP1. Two of them are included in .NET 4, but 981107 is not and it is not clear when it will be available for it.