Fixing strange hard disk noises with power management

Tux
My recently acquired Laptop made some strange noises several times a minute. With some help I was able to identify the hard disk power management as the source of the problem. This was easy to fix with some power management system scripts.

My recently bought new Laptophad had one problem: a strange noise bugged me. It came up several times a minute, even in idle mode. With some very helpful comments of some friendly readers (thanks Chris, Rui and Michael!) I was able to identify the hard disk as the main problem: the Load Cycle Count went up much too often per minute.


# smartctl -a /dev/sda | grep Load_Cycle_Count
/dev/sda: ST9160821AS: 37°C
193 Load_Cycle_Count        0x0032   099   099   000    Old_age   Always       -       2330
# smartctl -a /dev/sda | grep Load_Cycle_Count
/dev/sda: ST9160821AS: 37°C
193 Load_Cycle_Count        0x0032   099   099   000    Old_age   Always       -       2333
[...]

I went through the comments and through several bug reports and howtos to determine the best way to fix the issue: the fix should be easy, should work with suspend/resume and should be as native as possible, so should use the tools which are thought to be used for such cases.

The (for me) optimal solution was explained in the OpenSuse Wiki in the article “Disk Power Management”: a script is dropped at /etc/pm/config.d/disk which defines the parameters for the disk disk for the powersave mode and the normal mode. The explanation is easy to understand and does not use distribution-specific tools, so it should work on every distribution out there - it dead certainly work here at my Fedora machine.

The only question left now is if I should report my specific hard disk model anywhere to make sure that this problem is fixed automatically in the future.

Short Tip: Real web page zoom in KDE 4’s Konqueror

kde-logo-official
In Konqueror the font in web pages could always be increased directly by pressing Ctrl and move the mouse wheel. But since it just increased the font size, not the graphics size, and therefore destroyed the layout of the web pages totally.

With KDE 4 this behaviour has changed: the images are now zoomed as well, so the entire layout is kept and showed to the viewers as expected.

This is one of the smaller hints you do not know until someone tells you ;)

Microsoft’s ecosystem: cracks in the reliability

microsoft
The microsoft ecosystem is often mentioned to be long term reliable and therefore be a good option for companies with long term plans or needs. However, recently some parts of Microsoft’s ecosystem fall apart and became incompatible, which is not what you expect from a reliable system.

The Microsoft ecosystem consists of thousands of services and applications provided by 3rd parties. However, the foundation of this system is provided by Microsoft’s services, applications and it’s Operating System. When these are suddenly not reliable anymore the ecosystem is suddenly not stable at all.
And there have been three large cracks in the near past which should make everyone worry who bases his or her IT on Microsoft’s products: PlayForSure, Outlook Express/Hotmail, and to some degree even OOXML.

PlayForSure

PlayForSure (DRM codename: Janus) was introduced by Microsoft in 2004/2005 to mark players which have been certified to follow Microsoft’s dream list: Windows Media Player compatibility, DRM support, MTP-only support, and so on. Originally this was also an attempt to push Ogg Vorbis out of the field.
The basic idea was that every player could be used on any Windows System which had Windows Media Player 10 installed. And that every player would play any music sold in any PlayForSure store.

The shortcoming for the user would be that the players and music wouldn’t work outside of this system (Apple, Linux, Windows 2000, Windows XP without the WMP 10, etc., other MP3 players). And if the user wanted to switch to a new computer, he/she had to re-download a licence for the new computer. That’s a usual problem with DRM, but is important in this regard.

So the ecosystem was built up, and almost every new MP3 player on the market was certified against PlayForSure. Millions of songs were presumably sold to probably millions of users.

But then Microsoft decided that PlayForeSure is not as cool as the iPod, and published an iPod enemy, the Zune. But since the Zune had to be Microsoft-Only it wasn’t Play Fore sure. The ecosystem got it’s first crack. And then Microsoft thought that having two systems was a bit confusing and closed down PlayForSure at the end of 2006. In the future the label “Certified for Microsoft Windows Vista” is supposed to replace PlayForSure. But while that certificate has some elements of the former PlayForSure it - for example - requires Windows Vista and has other, additional requirements. It is a new logo.

What we have now is a broken ecosystem. And that although most of Microsoft’s partners relied on the availability of PlayForSure - many of them still advertise their products as PlayForSure certified.
So now we have hardware vendors who have to push out a new line of players which can be certified against the new logo. All their old devices now depreciate because they are not certifed against the new logo.
And we have the webshops which sold PlayForSure music. Many of their customers for sure still use Windows XP - which is of course not Vista ready. They are left alone a bit.
Last but not least we have the customers: music which was bought with PlayForSure needs licence upgrades to work on a new computer. Since these licence upgrades will not be provided after PlayForeSure’s shut down the music is essentially worthless. Think of millions of songs here - think of millions of dollars/euros here!

Outlook Express and Hotmail

Outlook Express is an e-mail client shipped with all Microsoft versions since Windows 98 and before Windows Vista. Although it was responsible for many large scaled virus attacks and spreads due to a horrible security concept and had many more, other technical shortcomings and problems it was widely used. And it was of course often used with Microsoft’s own E-Mail service, Hotmail.

However, recently Microsoft decided that the old protocol used for the communication between Outlook Express and Hotmail wasn’t what they would prefer. They wanted to switch every user to the new DeltaSync - and so they decided to drop Outlook Express support in Hotmail. As a resultyou will not be able to use Outlook Express together with Hotmail in the future any more. Instead, you have to upgrade to Windows Live - if you can, because currently it does not support Windows XP 64 or Win2k3. Also, if you used to use another client, that one will also not work anymore.

To summarizes: as a user who relied on Microsoft’s in house e-mail technology you are screwed.

OOXML

The last problem is still in development, and the outcome is not really clear at the moment. But the root of the problem is ISO’s adoption of OOXML.
Originally published by Microsoft to have a pseudo-standard which is very hard to support for other software vendors, the ISO adoption process became necessary because no one was interested in not officially standardized formats by Microsoft.

So OOXML went through the ISO process, and got several comments where it has to be reworked - Microsoft could have avoided this by working together with others right from the beginning! In the end the new format was accepted as a standard. But the necessary revision of the format resulted in a non-compatible Microsoft Office:

Such a test is only indicative, of course, but a few tentative conclusions can be drawn:

  • Word documents generated by today’s version of MS Office 2007 do not conform to ISO/IEC 29500
  • Making them conform to the STRICT schema is going to require some surgery to the (de)serialisation code of the application
  • Making them conform to the TRANSITIONAL will require less of the same sort of surgery (since they’re quite close to conformant as-is)

Given Microsoft’s proven ability to tinker with the Office XML file format between service packs, I am hoping that MS Office will shortly be brought into line with the 29500 specification, and will stay that way.

I’ve included the last part (emphasis not by me) to make the point of view of the poster clear. He thinks/hopes that MS will update the Office suite soon (an ODF test suite can be found here, btw.). However, Microsoft haven’t published any time frame, schedule or plan yet about such an update.

So in the end Microsoft first created an ecosystem around its new file Office Suite and on full purpose introduced a file format which would never become an ISO standard in that form - to afterwards apply for the said standard and change the file format again. As a result Microsoft Office 2007 still is not compatible to the new OOXML version itself atm.

Every organization and every software vendor who already jumped on the OOXML train before the ISO changes therefore has now to reconsider the software strategy: it should prepare for the switch to the updated OOXML - but also wait with it because it is unclear when the switch in MS Office will come.

However, preparing such a big update and keeping it ready consumes quite some resources. And small software vendors might not have these resources. Additionally, even smaller vendors which only had the resources to accept the new file format once are screwed.

And of course we have the users: everyone who tried to base a long-time backup solution on top of the old OOXML is screwed as well and should re-build the entire update again with the new OOXML.

Final words

The main problem behind these cracks in Microsoft’s ecosystem is of course the lack of well proved, open standards. But since Microsoft is not willing to support standards, the cracks are there now, and they are huge.

Of course from Microsoft’s point of view all these things make totally sense: everyone who strictly follows Microsoft as close as possible (use Vista only, etc.) and is ready to lose quite some money once in a while (because the music must be bought again and again or similar) will survive all these changes.

So to summarize the problem with respect to Microsoft’s position: problems only occur where users don’t have money to waste or want to have choice. Microsoft does not want the user to choose anything. The user is there to consume, not to choose - or actually think. This is best expressed by a comment from Omar Shahine, one of Microsoft’s Lead Program Managers, who tried to explain the Outlook Express/Hotmail development:

What we didn’t want to do is offer POP access and then have [...] customers use POP over DeltaSync.

If customers decide to stay with a product despite you offer them a new one - that means the customers prefer the product. Usually a company has to stick with such things in the real world.

However, Microsoft thinks it knows better. Microsoft forces it’s users against any opposition. And against any possible casualties on the consumers side.

KDE SOC projects published

kde-logo-official
The list of the Google Summer of Code projects for KDE was published. The list has several interesting entries to make the life for KDE users much easier.

The results for this year’s Google Summer of Code are out, and several KDE ideas were picked up.

Going through the list I found some items which immediately rouse my interest:

Of course this is just a small list of the accepted applications - and an accepted application doesn’t necessarily mean that the project will be a success. Additionally, this year the process around the applications and slots wasn’t perfect, but maybe the KDE e.V. can step up there to make the rejected projects possible nevertheless?
An idea would be to collect money through the KDE e.V. to sponsor other applicants, given that KDE finds their applications worth a try. I would be willing to donate some money for a stand-alone, KDE integrated browser based on WebKit, for general video input device support in Phonon or for a full featured Sonnet.

Linux (Fedora) on a Dell Latitude D630 - first impressions

Tux
The new Laptop is almost fully supported under Linux. While the install process wasn’t really easy, the hardware was afterwards detected without a problem.

The new laptop works like a charm. After few days I am already sure that my money was well spent. The hardware is almost fully supported, and everything I tested works like expected.

General hardware support

The general hardware is well supported: CPU, hard disk, RAM, screen, touchpad, bluetooth and USB controllers usually don’t make any problems on modern Linux distributions.
Other hardware also works well: the Intel WLAN card works, also does the Intel Audio. I found some articles mentioning problems in this regard, but this was fixed with recent kernels.

The NVIDIA GPU is also supported - with recent proprietary NVIDIA drivers. I would have liked to have a GPU where high quality free drivers are available, like Intel cards or ATI/AMD cards, but there wasn’t a comparable model out there featuring ATI/AMD, and Intel cards simply don’t have enough power. I just hope that NVIDIA starts a similar Open Source attempt as AMD/ATI. For a start full RandR 1.2 support would already help me to re-use all my RandR scripts I wrote over the time for different monitor setups…

Anyway, with the standard Fedora 8 system for the usual hardware and binary drivers for the NVIDIA card I can - without any problems - suspend and resume this machine.

Hardware quirks, untested hardware

There are two things I have noticed so far which are not working perfect or not at all: the first one are several multimedia keys on the keyboard: the mute/volume up/volume down keys are not working. I hope to solve this by following the guide at the hal quirk page.
Another things which is at least not working as perfect as I would it to work is the hard disk: I get a specific sound once in a while, like 2 till 3 times each minute. It sounds to my untrained ears that the hard disk just writes some small data package. The program powertop suggested to do
echo 1500 > /proc/sys/vm/dirty_writeback_centisecs when I noticed the sound once while powertop was running, but I’m not sure if these things are really related.
The sound is certainly nothing really disturbing, but it is at least noticeable, and I just wonder where this comes from and if it is something I could fix with software improvements.

Besides the here mentioned hardware there is of course hardware I didn’t test yet - like the modem or the docking station connector. I didn’t test the first one because I don’t expect to ever need it, and I didn’t test the second one because the docking station is not here yet.

Real problems

The only real problem I encountered with this machine was the installation - I struggled with it for quite some time until I found a decent way.

First I tried to boot the system using the normal Fedora 8 install DVD - that “failed” while attempting to load /sbin/loader: it just showed a message informing me that it loads the loader, nothing more. The boot up didn’t continue. However, the system didn’t really freeze because I could still restart it with Alt+Ctrl+Del. I’m not sure what the cause is/was.
The second try was an old Ubuntu 7.04 Live CD I still had somewhere - while I didn’t plan to start using Ubuntu for a longer time it would have been an option if Fedora wouldn’t have worked. But that CD failed to load as well.
The third try was an OpenSuse 10.2 Live DVD - that one worked pretty well. And since I already have a long OpenSuse past and since I am pretty used to their system it was really tempting to switch with this laptop. But then I also tried another recent Ubuntu Live DVD - and it worked as well. So I decided to also try a Fedora 8 Live CD - and indeed, the Fedora 8 KDE Live CD worked.

I must admit that I am surprised that there were so many problems. The Dell D630 is a quite old model and out there for quite some time, and I wonder how others have managed to install Linux on it in the past, as it looks to me that only new distributions successfully booth the machine.