Search Results: "ag"

24 January 2007

Wouter Verhelst: Ian Murdock on Solaris...

If you want another example, one from a company who any developer would agree is an outstanding engineering organization, here's one: "Sun has maintained binary compatibility between operating system releases for nearly a decade, enabling existing Solaris applications to run unmodified on Solaris 10. This means that Solaris applications developed 10 years ago will run on Solaris 10 unchanged, taking full advantage of new and advanced Solaris features."
Yes, it's true that there is outstanding backward binary compatibility in Solaris. However, if you're going to try use that as an argument in favour of Solaris, try again. Have you ever actually used a Solaris system? If you're going to say that backwards compatibility is good and that we should incorporate backwards compatibility into Linux somehow, then please do not point people towards the two ugliest operating systems on the earth with a lot of garbage baggage for the sake of backwards compatibility...

Martin F. Krafft: A good interface for selection?

I am working on a script to help me switch mail profiles from within vim, because mutt's profile abilities are seriously limited (even with muttprofile). The script will have three modes: manual, menu, and auto. In manual mode, you tell it something like switch-profile debian and it converts an existing mail to use my Debian profile. In menu mode, it displays a menu of available profiles, and in auto mode uses heuristics defined by the profiles to determine which one to use. My problem at the moment is menu mode. What I want is a good selection interface that (a) works over most any terminal, and (b) allows quick and dependable keyboard-driven selection. So far, I've had multiple ideas:
  1. A simple shell script that would assign [0-9a-z] to the profiles and then expect me to press one of the letters to select the profile I want (also see ssft). I don't particularly like this because if I added a profile to the presumably alphabetically sorted list, I'd have to relearn which keys to press when doing quick (subconscious) selection.
  2. debconf, which comes with multiple interfaces, but doesn't really seem made for the task. In the end, it would probably end up giving the same experience as (1) above.
  3. A method which I've implemented in the script (run it with the --menu option), which is reminiscent of "autocomplete". I need to add arrow selection and I have to figure out how to actually access $result from outside the while loop (this method doesn't work, because I need stdout for the interface and I have not managed to figure out using file descriptors 3 and above), but I already like it very much, although it seems rather brittle so far.
Do you have any other ideas? Or do you know of a tool/library I could use for this? I'd much rather have this work without having to install a separate package, but there's also no point in reinventing the wheel. PS: I am using POSIX shell for this even though Perl or Python would be much better suited, simply because it's faster and available everywhere. NP: Man on Fire / Habitat Update: Jan Hudec points me to Vim's console-menus and now makes me ponder whether I want a vim-only solution, or one that could be used by friends of other editors too. And Simon Huggins is trying to get me to use htag, which I'll have to check out.

23 January 2007

Lars Wirzenius: Debian: Project Ummikko: Final report

Last July, I started an experiment, called Project Ummikko, with a friend of mine to see how well Linux is suited for a new user who has used Windows for many years, but is not a techie. I installed Debian, with GNOME, for her, and she went cold turkey: she started using Linux only, unless there was something she had to do, but could not achieve under Linux. It has been half a year, and it's time to report on the results. Translating and abbreviating her own words:
The good part of using Linux is that I don't need to worry about viruses or defragmenting the hard disk and so on. I would be happy to remain a Linux users for things like that. In basic everyday use I'm happy. The things that make me want to return to Windows are gaming, flash and other multimedia, and a feeling of helplessness. I want to play Alpha Centauri and adventure games I can borrow from friends. Flash doesn't work, and installing software from Debian via Synaptic is difficult, I'm not comfortable with it. With Windows, I'm used to being able to go online to find a program to solve some problem I have, or some plugin I'm missing, or some toy program I want, and after a few mouse clicks, I have it installed. There's many things that just don't work with Linux, like Second Life, or when my ISP would give two free movie downloads this month and next, but it only works on Windows and Mac. I have a feeling of being left out. I could manage with Linux, with the assumption that I have someone like Lasu to do installation and tweaking. The best thing about this experience? It was interesting to be able to make the experiement, the meta level has been the best thing. The best thing about Linux is the kind of maintenance things like no viruses or defragging, Linux doesn't rot like Windows does.
After we had the above interview, I upgraded her Debian, and this fixed at least most of the problems related to flash. Synaptic remains uncomfortable, and there's not much that can be done about Windows-only software. I was pleasantly surprised by the results. I had expected much worse problems, but I'm happy to report now that Linux really does seem ready for the average Windows user. The uncomfortability of Synaptic may be partly due to the fact that it is a different model from Windows: you get all your programs from one place, instead of anywhere on the Internet, partly due to just not using Synaptic much at all, and partly due to Synaptic not being perhaps as easy to understand as possible. I personally find Synaptic to be very nice, and use it regularly, but I am biased by having used Debian package managers for over a decade. Even if she had learned to use Synaptic, it would, of course, not have been enough: there's lots of Linux software that hasn't been included in Debian. My friend therefore feels that it would be nicer, more liberating, to be able to just download any old Linux software from a web page and install it easily. My conclusions: package management for Debian based systems could do with some work targetted towards neophytes. I am sure there are things that could be done to make package installation easier, less daunting, without dumbing down the system.

Martin F. Krafft: Haunted by a password

I restarted my laptop last night only to discover that I could not piece together the password I used to encrypt my harddisk, and it did not help that I knew it was pwgen-generated (a sequence of random digits or letters which try to be "pronouncable"). I had left my palm pilot with the password database at home, and the emergency key I created for such cases was on a USB medium with a read error. I beat my head on the table and tried and tried and tried, but to no avail. So I started pondering how to convince my friend Hanspeter to walk to my flat again and help me out, having resigned myself to the idea that I'd have to do more reading tomorrow (today) at the office. As it turned out, Hanspeter would stay at my flat that night anyway and switched on my machine at home to allow me to fetch the emergency key. But I didn't end up needing it. I slept and dreamt and suddenly had some sort of a nightmare in which a string of letters came haunting me, tormenting my consciousness in much the same way that the Guitar Hero notes come flying at you. I woke up and had ten characters in my head, and I was confused. It took me at least two minutes to remember that I had been scrambling for the harddrive password, and when I figured it out, I wasn't surprised that the string I dreamt would decrypt my drive. I really should gain some distance to these frickin' computers! NP: Man on Fire / Habitat

Erich Schubert: Boycott adea solutions

... for being stupid spammers. Obviously they polled some blog search engine (maybe Google's blogsearch) for some keywords like Java, tried to guess the name from that and grabbed email addresses they could find. I just received this stupid "job application offer":
Good Afternoon Planet,

I represent a major Airline in Dallas, TX. We have 3 openings for C Developers with Java, Unix, Sybase and Oracle skills. Basically our client is doing a 2 year migration from C to Java and needs C language Guru's.
This "major Airline in Dallas" (Southwest? Amerian Airways? I don't know.) should maybe be looking for a different company to do their development. The company they are using right now, Adea Solution, obviously: One more quote:
**I apologize if you are not looking. We have a system here that pulls information from the web and does not distingish between those who are current and those who are not.**
If you know that your system is crap, just don't use it. Instead of spamming. If I were looking, I wouldn't put "I'm currently not looking for a job" on my homepage. And I'd probably try to get in touch with someone professional.

Erich Schubert: Boycott adea solutions

... for being stupid spammers. Obviously they polled some blog search engine (maybe Google's blogsearch) for some keywords like Java, tried to guess the name from that and grabbed email addresses they could find. I just received this stupid "job application offer":
Good Afternoon Planet,

I represent a major Airline in Dallas, TX. We have 3 openings for C Developers with Java, Unix, Sybase and Oracle skills. Basically our client is doing a 2 year migration from C to Java and needs C language Guru's.
This "major Airline in Dallas" (Southwest? Amerian Airways? I don't know.) should maybe be looking for a different company to do their development. The company they are using right now, Adea Solution, obviously: One more quote:
**I apologize if you are not looking. We have a system here that pulls information from the web and does not distingish between those who are current and those who are not.**
If you know that your system is crap, just don't use it. Instead of spamming. If I were looking, I wouldn't put "I'm currently not looking for a job" on my homepage. And I'd probably try to get in touch with someone professional. [Update: their homepage appears to be unreachable... looks like someone else was even more annoyed than I...]

Erich Schubert: How Skype cheats

Ever noticed the following with Skype: Go offline while several of your friends are online. Suspend your computer. A few hours later, resume your computer and go back online with Skype. Notice how it claims all the buddys that were online when you disconnected just came online? And a few seconds later, some of them all go offline? And several minutes later, they even come back? Skype (well, at least version 1.3.0.53, YMMV) obviously caches which users it has seen and just assumes they are still online. When it then has actually connected to the Skype network, it notice that none of its peers has them and decides that they maybe aren't online after all. Then it actually searches them throughout the network, and when it has found them puts them back online again. Thats what I call cheating. Other IM services are much more reliable here. And other IM services - at least ICQ and many Jabber servers, though not Google Talk - also have offline messages. I.e. you can write messages that will reach the receipient once he goes online again. Skype again cheats on these: if you send such a message, it will after some time display the notice that sending it failed. The next time both of you are online, it will actually be delivered. This can result in suprisingly old messages, and makes connecting to people much harder: for putting someone onto your list, you also have to be online at the same time. Assuming that you are using Skype only during work hours and going offline inbetween, this makes communicating with people in different timezones a pain. File transfers with Skype are also a game of chance: If both of you are somewhat firewalled (e.g. behin a DSL-modem-router combo), it often can't transfer directly. Instead it picks a random "super node" with large bandwith for transferring the files. However, the super nodes bandwidth isn't always as good as expected by Skype, so sometimes file transfers with Skype are awfully slow, even when your line isn't busy. MSN for example (though I do NOT suggest using MSN), is using it's own, well-powered servers to handle such file transfers (or maybe all; thats why I don't really suggest using MSN, since I've only seen it transfer my files through microsofts servers...). But usually file transfers via MSN - especially in firewalled situations - are much faster than with Skype. (A drawback of them is the short timeout) So consider using XMPP/Jabber (= Google Talk) instead. They are more reliable, and an open, well-defined, extensible standard.

Erich Schubert: How Skype cheats

Ever noticed the following with Skype: Go offline while several of your friends are online. Suspend your computer. A few hours later, resume your computer and go back online with Skype. Notice how it claims all the buddys that were online when you disconnected just came online? And a few seconds later, some of them all go offline? And several minutes later, they even come back? Skype (well, at least version 1.3.0.53, YMMV) obviously caches which users it has seen and just assumes they are still online. When it then has actually connected to the Skype network, it notice that none of its peers has them and decides that they maybe aren't online after all. Then it actually searches them throughout the network, and when it has found them puts them back online again. Thats what I call cheating. Other IM services are much more reliable here. And other IM services - at least ICQ and many Jabber servers, though not Google Talk - also have offline messages. I.e. you can write messages that will reach the receipient once he goes online again. Skype again cheats on these: if you send such a message, it will after some time display the notice that sending it failed. The next time both of you are online, it will actually be delivered. This can result in suprisingly old messages, and makes connecting to people much harder: for putting someone onto your list, you also have to be online at the same time. Assuming that you are using Skype only during work hours and going offline inbetween, this makes communicating with people in different timezones a pain. File transfers with Skype are also a game of chance: If both of you are somewhat firewalled (e.g. behin a DSL-modem-router combo), it often can't transfer directly. Instead it picks a random "super node" with large bandwith for transferring the files. However, the super nodes bandwidth isn't always as good as expected by Skype, so sometimes file transfers with Skype are awfully slow, even when your line isn't busy. MSN for example (though I do NOT suggest using MSN), is using it's own, well-powered servers to handle such file transfers (or maybe all; thats why I don't really suggest using MSN, since I've only seen it transfer my files through microsofts servers...). But usually file transfers via MSN - especially in firewalled situations - are much faster than with Skype. (A drawback of them is the short timeout) So consider using XMPP/Jabber (= Google Talk) instead. They are more reliable, and an open, well-defined, extensible standard.

Bastian Venthur: Allow myself to introduce . myself!

Ok then, this is my first serious blog entry. To break the ice, maybe an introduction would be appropriate: My name is Bastian Venthur I’m currently 26 years old and live in Berlin, Germany. I’m studying computer science at the Freie Universit t Berlin and I am Debian Developer since October 2006. I maintain a few packages, mostly KDE related and am quite active in reporting bugs. One of my current hobbies is to juggle. I started a few months ago and already master some neat tricks like Mills Mess, Over the Head and the famous four ball Async Fountain.

MJ Ray: So, what happened at Cooperatives SW?

I left home a bit late and rushed to the train stop, but the journey was easy after that. There was quite a good turnout at the meeting: Chris (chair), Alex (clerk), Paul (treasurer), Brian (secretary), Malcolm, Paul, Andrew and myself IIRC. Continuing: A promo leaflet is on hold until after funding is sorted out; No mutuals responded to a funding request letter; The reply to the BASIS grant bid is expected some time in January; The database update is waiting on Plymco as far as I could tell. Financial: the clerking contract went about 700 quid over budget last year and has been reduced to a skeleton service until funds are sorted out again. Donations from the Coop Group and Midcounties were forecast but aren't in the bank yet. AGM: the aim is to return to more of an event next year. Holding it in October should hopefully avoid the transport and weather problems of the 2006 event. Fund-raising: Chris to draft a bid for the cooperation fund to try to secure enough money for a support worker. The priorities would be clerking and event organisation, with possibly marketing, communication and news-production tasks. Numbers to come from the support tracker that Alex has been using. Need some sort of corporation to apply. Alex, Brian and myself to assist. 12 February deadline. Reply expected in March. Future work: newsletter deferred to mid-March. News items to Alex. Paul (I think) mentioned that some Healthy Eating Partnerships seem to be promoting their box schemes as 'Food Coops' when they are not cooperatives at all. We need to contact partnerships and find out which is doing what. Alex handed out a few Principle Six speed-networking packs. Interesting idea. Paul mentioned Cornwall Switch which will find the best green energy supplier for you (even if you're not in Cornwall). Next meeting 16 April 2007.

Yves-Alexis Perez: Xfce 4.4 released.

Ok, so now everybody knows, Xfce 4.4 has been released, the final version is now out since sunday.

During sunday I worked to package it, and by the time the release was announced, packages were (mostly) ready in our svn. It runs fine (no real changes since 4.4rc2 so there's no urgency).

4.4 won't be uploaded to unstable soon, because of the freeze status. It may be uploaded to experimental in the future.

I've made/uploaded preliminary packages to my repository. There are there packages for ppc and i386. ppc packages are built against gtk 2.10 from experimental, and they run fine (except mousepad with line numbers activated, known bug in gtk 2.10.7). i386 packages are built against gtk 2.8 except gtk2-engines-xfce, built against gtk2.10 because it now uses in the packaging a call to a module only present in gtk 2.10.

If you want to install Xfce 4.4 on i386 via my repository and don't wan't to switch to experimental gtk 2.10, you can do as follow:

apt-get source gtk2-engines-xfce
dpkg-source -x gtk2-engines-xfce_2.4.0-1.dsc
cd gtk2-engines-xfce-2.4.0
vi debian/rules


Now you just have to remove the call to dh_gtkmodules.
Then:

vi debian/control


In the Build-Depends: line, change the dependency from gtk 2.10 to gtk 2.8. (libgtk2.0-dev (>= 2.8) should be ok).

Then just rebuild the package as usual:

apt-get build-dep gtk2-engine-xfce
dpkg-buildpackage -rfakeroot -us -uc


As alway, don't use those packages if you want stability. Current i386 packages are not up to date (as I can't rebuild quickly for this arch), and they can change at anytime. If you still want to test, don't report bug on Debian BTS but rather on IRC (#debian-xfce@freenode).

Happy testing!

Raphaël Hertzog: More fun with Linux and serial ports on slow hardware

This is a never ending story for me. The first time I’ve had problems with Linux’s handling of serial UART dates back to 2005 (see my previous blog post on buffer overruns). At that time I could improve the situation by applying two patches (kernel-preempt and low latency). One year later, I have a situation where the buffer overruns are again easily reproducible at slower baudrate (54 kbauds), arguably there’s more than a serial application running this time, and it looks like the load generated by other processes (mainly watching digital I/O) renders the system less reliable with respect to its handling of serial ports. This time I follow the advice to try out the 2.6 kernel because many “real-time improvements” (coming from the -rt branch, check its wiki) and “embedded improvements” (coming from linux-tiny) have been merged. So I tried the stock kernel with very bad result. Results are better than the stock 2.4 but they are worse than the patched 2.4. So I decide to try the -rt patch on 2.6, but this patch doesn’t work on my CPU card and my bugreport didn’t lead to any fix (nobody responded even though I tried hard to include the necessary information and I was ready to do whatever I would have been asked to try). At the same time I explain my problem on the linux-kernel mailing list. The discussion doesn’t answer my questions but still brings two ideas to try out. In the end, with two simple tweaks to the stock 2.6 kernel (mainly configuring the UART to send the interrupt as soon as the first character arrives, instead of waiting for 8 chars to accumulate in the FIFO) I have been able to get something better than the patched 2.4. And it turns out my first choices for the 2.6 kernel configuration have not been very wise so the comparison between stock 2.4 and stock 2.6 above doesn’t mean much. Unfortunately, even if better than the patched Linux 2.4, it still doesn’t give good results in some conditions. So my primary question remains: is there a way to patch my kernel so that it will handle the serial related tasks (servicing interrupts from the UART mainly) as its primary job ? I don’t mind if such a change impact negatively the speed of the system if I can make sure that my serial exchanges are reliable. And by reliable I mean of course no buffer overruns, but there’s a second similar problem that has been discovered: when using software handshake, the system can send out between 10 and 25 characters after the partner has sent its XOFF. Sending up to 6 characters after the XOFF is ok, more is asking for troubles because the UART on the other side will probably encounter a buffer overrun… If you have any idea on how to resolve my problems, by any means, let me know.

Zak B. Elep: Smarty enough for Smart Bro, and 3G

Finally, I’m now on broadband! :D I made the move today to Smart Bro as they have been touting a promotion offering free installation and a 384 Kbps connection via their Canopy system. As I’ve been mostly offline (or slightly online, counting the occasional GPRS modem connection on a GSM network) I figured that this is the best way at the moment to get broadband connectivity, as the only DSL choices available locally have been either seriously damaged (due to the last supertyphoon) or excessively priced (with unpredictable foreign-currency adjustments.) So, needless to say, I’m now enjoying myself with this new investment. Adding to this, I also got a new Netgear WGR614v7 wireless router so I can share the connectivity with perlis. While not exactly as flexible as the famous WRT45G from Linksys, it surely does the job to sharing the Net, so I’ve no complaints about it. And I’ve no complaints about the service offered by Smart Bro, either; the installer did an absolutely good job of making sure that the Canopy SM is on the most receptive direction and even gave me some pointers on what to expect with such a connection. Actually, I’ve already made arrangements about the peeves experienced by other Smart Bro users by taking a cue from Joey Hess by setting up the right service caches and by being not too abusive with my expectations. But, I still have a lot to expect with this new connection. After all, I still have a few more items on my TODO that needs doing, namely writing about what the hell happened to me since last month until now, and my future directions (erm, resolutions) for this new year. Ah well… … but first, I need to do some monster dist-upgrades… :P BTW: a UMTS (3G) signal has finally been broadcast in Daet, giving me a reason to also have smart-3g as a backup service in case the Canopy goes down. Yipee!

Erich Schubert: DBus inspector

I've been working a bit more on DBus inspector. Python is so much fun and PyGTK just rocks. I love you, guys. :-) Ok, not everything is working right yet. I tried making the app multithreaded, and putting the DBus handling into a second thread, but failed at that so far. Getting threading work right with Python and GTK and DBus is just too much, I guess. :-) Here's a screenshot: DBus inspector screenshot You can grab the code temporarily from this location. It's rather crude, undocumented (fortunately Python code is quite easy to read) and still lacks some interesting parts (e.g. icons to differentiate between services, interfaces, methods, signals, ... and of course actually displaying method signatures or even allowing you to invoke them...) - but it's a start. Note that DBus using applications support introspection at varying levels - NetworkManager doesn't have a useful introspection description, and gaim goes the other way by apparently exposing every single method during DBus, so you can do the craziest stuff (but it's hard to find the method you are actually looking for).

Erich Schubert: DBus inspector

I've been working a bit more on DBus inspector. Python is so much fun and PyGTK just rocks. I love you, guys. :-) Ok, not everything is working right yet. I tried making the app multithreaded, and putting the DBus handling into a second thread, but failed at that so far. Getting threading work right with Python and GTK and DBus is just too much, I guess. :-) Here's a screenshot: DBus inspector screenshot You can grab the code temporarily from this location. It's rather crude, undocumented (fortunately Python code is quite easy to read) and still lacks some interesting parts (e.g. icons to differentiate between services, interfaces, methods, signals, ... and of course actually displaying method signatures or even allowing you to invoke them...) - but it's a start. Note that DBus using applications support introspection at varying levels - NetworkManager doesn't have a useful introspection description, and gaim goes the other way by apparently exposing every single method during DBus, so you can do the craziest stuff (but it's hard to find the method you are actually looking for).

Steve Kemp: I dont mind not knowing what lies down the track

I’ve got my first real rails application working correctly - a simple online bookmark system, which is scuttle-lookalike. (This is part of my master-plan to remove PHP utterly from the few remaining places where it exists upon my server(s), so it isn’t as pointless a project as it appears to be!) It is pretty neat, and it handles most of the things you’d expect. Fetch it from CVS and install it with the following recipe:
cvs -d :pserver:anonymous@cvsrepository.org:/home/cvs login
cvs -d :pserver:anonymous@cvsrepository.org:/home/cvs co bookmarks
cd bookmarks/
mysql --user=root -p -e "CREATE DATABASE Books_development"
rake db:migrate
ruby script/server
Once that is done point a browser at http://localhost:3000/ - and you should see something like this. (If there is no admin user setup it will display a “signup” link, once there is an admin defined this will go away. When you’re an admin you can view/edit/modify all the bookmarks which are set to “private”.) I’m new to rails, and ruby for that matter, so if this is packaged badly, or is missing something to make it work please do let me know. Updated: Changed from download to the CVS repository. I also have a collection of tools for exporting from Insipid & Scuttle if they are useful let me know.

Ian Murdock: How do you compete against phenomena ?

Knowledge@Wharton: “Today [Steve] Ballmer sees two major competitors for Microsoft — the open source software movement and advertising-supported software. According to Ballmer, the threat comes not from specific companies, but from the business models represented by these two trends. ‘Right now, the emblem of the first one is Linux and the emblem of the second one is Google. But it’s not the companies, it’s the phenomena’ that present the greatest challenge to Microsoft.” Microsoft is very good at competing against companies, but how do you compete against “phenomena”? That isn’t as clear. (Jonathan Schwartz wrote about this in 2004.) And at least, from Microsoft’s perspective, the second phenomena is emblemized by a company. Linux isn’t even a company. The usual tactics won’t work.

Ross Burton: Drink The Lemonade

Dave Cridland posted to maemo-developers again today, spuring me to spend my lunch time checking out his email client instead of doing something productive like eating. It's very interesting. Basically he edits the RFCs for Lemonade, which is a IETF standard for email on mobile devices. Basically it is a set of extensions for IMAP and SMTP that improve their performance over high-latancy low-bandwidth connections (like bad wi-fi or GPRS), such as forward-without-download and efficient re-synchronisation commands. Existing extensions such as a useful IDLE implementation are also used to push new mail notifications to the client. To test the ideas in the real world there is Polymer. Polymer is a simple IMAP client that uses the Lemonade extensions as much as possible, so is ideal for mobile use. There is more: it uses ACAP to store the settings on the Internet (Dave also provides free ACAP accounts) so it's configure once, run anywhere. I've just tried it on my laptop, and it works nicely. But it gets better. There is also Telomer, an email client using Lemonade and designed for Maemo. I've just tried it out, it may be seriously lacking in features at the moment, but it works. Features can be written over time, but it takes serious thought to get something as fast and usable as Polymer on mobile devices. I'll quote Dave:
That all said, once it's started, it connects to the mailserver over GPRS in about 10 seconds and pulls up the summary listing in 15, with a mailbox size of 2,882 messages. With a mailbox with 33,732 messages in, it's a little slower at around 25 seconds. That's still way faster than a desktop client on a LAN, of course, unless that desktop client is Polymer - in which case it'll seem remarkably slow.
This is what we need for a mobile email client. I know I've found my email client of choice on my new N800. NP: Wish You Were Here, Pink Floyd

Holger Levsen: SLUG meeting

Yesterday I attended the Sydney Linux User Group january meeting and gave a talk about the DebConf video-team and the differences in our setups compared to LCA2007 (slides available now, video soon).

It was good to see a bunch of people from LCA again and the 2nd talk about slack, googles internal maintainance tool which was just freed, was also quite interesting. For me there was nothing really new, as slack works quite similar as FAI, the interesting part was to compare them :)

Unfortunatly I didn't see the report about googles summer of code, as it was held in parallel. But you can't have icecream with fries :-)

Fortunatly I also could attend the Sydney linuxchix meeting before the SLUG meeting, where a good introduction to Ruby on Rails was given.

Relocating to Melbourne now - cheers and thanks to everyone in Sydney contributing to the good times I had here! :-)

Erich Schubert: Limiting inbound traffic

... is far from trivial: you are not on the senders side, and not every protocol (in fact very few) allow rate limiting. Not to speak of limiting the total of multiple connections. Basically you'll have to accept that the sender is sending the data at it's own pace. However, the TCP part of "TCP/IP" is designed in a way that it does a fairly good job at detecting bottlenecks. The basic principle is this: when a packet is received at the destination, a confirmation will be sent back. By adjusting the rate of packets going out to match the rate of incoming "received" message (and to minimize the number of lost packets that need to be resent), TCP/IP connection can adopt to the slowest link in the chain. This is how incoming traffic shaping works: incoming packets are held back (or even dropped; however holding back packets might also lead to dropped packets somewhere up the stream), which in turn will limit the speed "ACKs" go out, and thus (hopefully) the speed data is pushed down to us. It's not perfectly reliable, and by design won't work well with short-lived connections and non-TCP-traffic. But it can still come in handy. We have a shared internet connection with 2 mBit. That usually is fine, but our router isn't very smart. When I'm using up all bandwidth for a longer timeframe (say, apt-get upgrade with new openoffice and tetex), the other users will suffer from a serious performance impact. One of them is often playing games online, and eventually gets kicked out of a game server because of his bad roundtrip times. By limiting my incoming traffic rate, I can hopefully keep the connection useful enough for him. In my tests, it worked pretty good at keeping my bandwidth useage down. The remaining unused bandwidth should allow his game data to pass through without larger extra delays. Here's the short script I'm using:
tc qdisc replace dev ethWIFI handle ffff: ingress
tc filter replace dev ethWIFI parent ffff: protocol ip prio 50 \
  u32 match ip src 0.0.0.0/0 police rate 1500kbit \
  burst 10k drop flowid :1
tc qdisc replace dev ethWIFI root tbf \
  rate 1500kbit latency 25ms burst 10k
Don't forget to adopt it to your needs, you probably havn't renamed your interface to ethWIFI and might want a different valye than 1500kbit.

Next.

Previous.