Tobi Waves

Our department is taking part in the ETH Laptop Project. This means, we are helping our students to make better use of their laptops. Currently this means we are developing a Linux and a Windows setup tailored to the requirements of our students. These setups will make it simple for them to integrate their laptops with our Unix Environment. We also have struck a deal with IBM which offers the students IBM Laptops at competitive prices and we will put our own Windows and Linux on these boxes.

Today I have been trying to get the IBM Windows XP installation which is already on the laptop when the students buy it, into a form so that it contains all the latest security stuff and fixes from MS and updates from IBM as well as our locally developed packages. When all the stuff was in, I used the sysprep tool to 'reseal' the machine, so that when the students boot it, it will come up with the usual short setup where the user can define the admin password and has to enter the serial number. Well, that was the plan at least. When I tried to reboot after the sysprep step, Windows came only halfway up and then complained, that setup could not continue because two processes were accessing the registry. BOOM. Reboot.

Over the course of the day I tried the whole spiel in many variations, searched the web, hunted through newsgroups. As every try took about 40 minutes, this problem was really painful to debug. Eventually and counter to all I expected, it finally worked. Unfortunately I had twisted and turned so many knobs that I am not sure which one was actually responsible for the sudden success. So tomorrow I will be at it again, trying to verify my recipe for success.

I so am glad that I can mostly work with Unix systems and only have to use windows occasionally. Whenever there is a problem with a windows box I feel like I was forced to wear a thick winter gloves while trying to repair a watch, blindfolded and someone occasionally moving the watch around.

But hey, I am stronger than windows! Eventually it sits up and begs for food, but the process always is extremely annoying.

Today around 9am our main Solaris server started acting up. Its performance got patchy. We eventually found that it was suffering from excessive TCP retransmits of up to 1000%. This means that for each packet it sends out on the net it has to try 10 times until it is successful. This is an extremely hight value, or so Virtual Adrian tells us.

We started searching franticly for the reason of the problem, as performance on the server and even more on its clients was suffering badly. After about one hour of web hunting with and traffic dumping, we gave in to the pressure from the street and rebooted the beast, hoping that probably some internals of the kernel had been thrown out of whack and after a reboot all would be well. And indeed it was, at least for a few minutes. Then the server started misbehaving again, driving its TCP stack through the roof. As rebooting did not help, we went back to tcpdumping and etherealing. I did learn a lot about pcap filter syntax ...

'tcp[tcpflags] amp (tcp-rst) !0 ampamp tcp[tcpflags] amp (tcp-ack) =0'

but nothing about the reason for the retransmits. Fortunately, at this stage, the retransmit rate was not always at 1000% so work was possible for our users.

Then, in the early afternoon, Manuel found that the root disk of the server causing SCSI timeouts. As if we didn't have enough on our hands already. SCSI timeouts make the machine stop and wait for several seconds at a time. Together with the server, most people using its resources, were experience the same freezing problem on their workstation.

What a day. I have been writing emails about what was happening to our users all day long, but things were really stating to look bad. Our wonderful reputation for high quality service and superb uptime was going down the drain. It seemed though that most users were not blaming us at this stage, probably due to the fact that I kept them up to date with what was happening.

Around that time David found, that in the latest Solaris kernel patch there was a fix for some TCP stack issue which might be related to the retransmits we were still suffering from. He started to put in this patch so that we could activate it when we rebooted. This was going to be necessary anyway as I was preparing to replace the root disk with a fresh device.

Then, suddenly just minutes before the reboot, the server went back to normal, the retransmits were gone and performance was good again, no traces left.

So here I am, another day older and not much smarter about what was causing todays network problems. I can imagine things like that there is a bug in the Solaris TCP stack which can be triggered by a rouge packet and this would cause the symptoms we experienced today but I suspect, once the real reason is known, it will be way less spectacular.

Once upon a time, there was this firm, the little Toy Factory, they were building these neat and simple woodblock toys. Kids could use them in various setups. Clever kids could even create their own additional woodblocks and hoock them on. Woodblocks seemed quite simple, but the trick with these toys was their clever overall design which made it easy integrate them with other Woodblock toys and even create your own additions. The factory was really successful with the kids who knew about their toys. Partly also because each Woodblock toy also came with a complete manual explaining not only how to use it but also giving detailed account about how this particular WBT had been constructed. Woodblock toys guaranteed for hours of satisfying and creative playtime together with your friends.

Not far to the north west there was this other company, the Lolly Makers, they produced shapely and tasty lollies in many colorful designs. Kids who tried them were really taken by the great taste of the sweet lollies. The lollies sold very well and soon most kids who lived in the vicinity of the lolly factory could be seen wandering the streets with a lolly in their mouth. Interestingly enough these kids seemed to loose all interest in playing with the woodblock toys or other kids apart from talking about the latest 'inventions' of the Lolly Makers. The lollies seemed be all they needed. Rumors had it that the lolly makers were using addictive and psychoactive substances in their creations. But whoever uttered any suspicions in this direction soon got letters from a big firm in the city, who advised them to refrain from telling any further lies about the Lolly Makers.

Over the last few days I have redesigned the Website of my Department, and implemented it purely with CSS2. I had to discover the hard way that even 4 years after the standard has been published, we are not there yet. While Mozilla is shining bright with its good implementation, many other entries like Opera, Konqueror and IE are working hard to do a good job but fail in odd places. What is amazing is where they do not work. Opera, for example, can not grok, that a box which is defined by its distance from all edges of the browser window is fully defined an can be displayed propperly. It just ignores part of the settings to be able to draw in the wong place and size.

Why does this not get fixed? First I thought there must be many bugs, and they just don't get round to fix this particular one. But then I got another theory: Everybody who does serious CSS2 Webdesign is working hard to make sure that his pages work with all the players in the CSS arena (Opera, Mozilla, Konqueror and Internet Explorer). Therefore these stupid bugs don't disappear as nobody will notice them but Webdesigners who then successfully hide the bugs from the end users by going the extra mile to make their pages work with all browsers.

How about a website which collects pages that pass all CSS/HTML conformance tests at (validator.w3.org ...) with flying colors, but got axed because display problems in some browser prevented them from working fully 'cross platform'

With todays low disk prices, it becomes intriguing to use several large harddisks as backup media. On the simples level they can be used like tapse to store dump files. But having all this quickly accessible storage space, it would be way cool to have some snapshot like quality where the backups are readily accessible by the users. Today I read about a cool idea for backup which is exactly what I have been wishing for the program implementing this is called FauBackup (faubackup.sourceforge.net ...). I could smack me for not thinking off this myself. Anyhow, here is how it works.

First Day Make a full copy (file by file) of the work partition to the backup partition.

Second and every following Day Walk the work disk and compare the file i-node change time of the files with the time the last backup was taken. Anything that staid the same gets hard linked from the previous backup to the current backup. Everything that has changed gets copied over.

Additionally compressing the files in the backup could lead to significantly lower disk usage. The advantages of this approach are: a) Users can restore their own files by just going into the backup partition. What I am not clear yet, is how to prevent them from modifying the backup. b) Backup space requirements are minimal as only changed files go into the new backup and old ones are hard linked. c) if we run out of space old backups can simply be removed as all backups are complete and independent of each other.

Now I just have to decide which computer is going to be my first victim. And I have to figure out if I am going to be stubborn and write my own implemntation implemntation of this idea or if I can convince myself to use faubackup.

In 14 days we will be moving out of our good old flat, to a new house. Or rather we will move out, and after Easter we will move into the new house.

Yesterday we finally chose the new table and chairs we were going to buy for the house. We spent a great deal of time figuring out what table we were going to get and where we would get it without spending all the money we have left after buying that house.

What we sort of ignored all the time, was the chairs. We did not ignore them completely, no, but we did never discuss their color. So when we finally were ready to order everything, yesterday at the furniture place, the guy wanted to know which color the chairs should be (the part which is not wood).

We have a pitch black floor, a fair beech wood table, white walls, transparent light gray white curtains. The chairs should contrast to their surroundings without taking all the attention. Aargh ... what a decision 20 minutes before they are going to close the shop for the night.

We went back and forth amongst all the colors on offer and finally settled on a fair grayish color with a hint of blue. With the order sent and the price agreed, we are left to worry if this was really the color which is right for us. Will not be to delicate? Will it fit with the rest of our furniture? I see six to eight weeks of worrying to worry ahead of us. Because of all our special wishes, it will take them that long to produce and ship the things.

Don't say it ... just don't! I'm going to bed now.