Search Results: "ron"

6 May 2006

Joey Hess: day at the dilab

Today I went over to Brian's to work on some problems with the dilab. The remaineder of this blog entry is unlikely to be interesting to anyone except possibly sparc and mips porters, but I wanted to write it all down. First thing I noticed was that the internal network was very slow and lossey. I isolated this to its switch, which must be broken. Brian lent me a 7 port hub, which I put the more demanding machines on, leaving the others on the switch, and hopefully this will clear up some of the network issues I've been seeing during tests, until I can install a proper new switch. Unfortunatly, this still didn't clear up the proliant's failure to use its emulated network CD drives, which I'd hoped it would. Still working on figuring out why that's broken. As I was testing out all the machines after that to make sure they still worked, I heard a very strange sound from one of the X-10 appliance modules (#3). Sounded like a CD drive failing to open. I think its relay must be fried, after 1.5 years of turning on and off half a dozen times a day. So I had to move the raq 1 on that over to a different power module (#1), and moved the ia64 that was on that one over to a non-controlled power port. The raq 2, that handles d-i daily mipsel builds, was hung with "booting Debian" on its LCD, and power cycling didn't help, unfortunatly it lacks a serial console to see what's wrong. So I pulled it and took it home, will have to investigate it and try to get a stable system on it. Until then mipsel d-i dailys are still down, until ths takes them over. Finally, I took a look at the 2.6 kernel issue on the sparc ultra 5, which worked with 2.4 but not 2.6. Luckily Brian knows a lot about suns and a lot about serial (used to be my sysop, and still is in a way), and he diagnosed it as having the port running too fast, dropping it down to 9600 fixed that problem. So, shopping/todo list from the day:

Martin F. Krafft: Travel medicine

In preparation of the Debconf6 conference in Mexico, the question has to come up how to prepare for the hygienic circumstances in developing countries. While Mexico was also described to be on the top end of the developing countries, certain issues are still left to be addressed. I have travelled to developing countries a bit and have gathered some experience, which I would like to share. However, keep in mind that I am not a doctor, just a globetrotter with a strong interest in pharmaceuticals. Let me address the most common problem first: "Delhi belly", as travellers' diarrhoea is sometimes referred to (because almost no traveller makes it through India without dysentery). I know of three preventative and four reactive treatments. The best preventative in my experience is to stay well above the necessary level of hydration by drinking lots more than you usually do, especially when it's hot. Prefer bottled water and juices, stay away from ice and fresh juices and shakes (as hard as it may be). Hygiene (i.e. hand disinfectant) is important, but you cannot control what's in the pot that way, especially not in restaurants (though upscale ones are generally safer, so don't be greedy). In extreme cases (when you have no other choice), another approach worth a shot is to drink enough high-percent alcohol with anything you take in orally. It's no guarantee and often not a very preferable solution (especially when it's hot), but it's kept me running a bunch of times before. Instead of growing paranoid about it, consider the treatments: the most common two treatments I have seen among other travellers were Loperamide ("Immodium") and Ciprofloxin (or Ciprofloxacin), an antibiotic. Both are suboptimal and to be used as emergency medication only (and many traveller's don't know that). While Loperamide regulates your urge to go to the toilet by blocking and binding that which comes out the behind, it also prevents the body from ejecting the bacteria which caused the digestive system to malfunction. Thus, Loperamide should only be used when you cannot get to a toilet as often as necessary (e.g. on a bus ride), and it must not be used for more thantwo or three days without medical advice from a doctor. Ciprofloxin, the antibiotic, is almost always a sure-fire cure against diarrhoea, but it must be taken for around seven days, usually up to four days after the symptoms faded, or else the body can build up a resistance to the chemical. During this time, you should not consume alcohol (or other drugs), and limit the amount of physical exercise, which makes it not suitable for e.g. hiking or trekking trips. Note that Ciprofloxin is a prescription-only medication. Apart from staying hydrated, my favourite cure is medical coal (tablets at 250mg each; "Kohle Compretten"), which have negligible side-effects and can thus be used on-demand. If I feel my stomach going a little wobbly, I pop two of them 2-4 times a day, and that settles it after 24 hours. If the diarrhoea does come through, double the dosage. It'll help regulate the need, but at the same time won't block it. If you do get unlucky, try to rest, switch to food known to be safe (e.g. packaged food), and drink a lot, ideally electrolyte solution, but a slight salt-sugar solution will also do. I actually like fresh orange juice with sea salt, which will do the trick. I am generally not a big fan of taking pharmaceuticals, but I always have some important ones with me:
  • Ibuprofen: mixes well with alcohol and is thus a perfect small-scale pain-killer against pre- or post-party headaches, or headaches that would ruin a nice dinner with friends. Make sure to drink a lot of water, and even more if you also take in alcohol
  • Paracetamol: medium-scale pain-killer, helps reduce fever. Does not mix well with alcohol. Primary symptom treatment for malaria and dengue fever, influenza, and colds. Drink lots of water.
  • Diclofenac: against swellings and inflammation. Does not mix with alcohol, and the tablets may seriously affect your digestive system. Consider combining with Pantozol, even prophylactically.
  • Valium, or preferrably something less strong: on the bus or on the plane, if you cannot sleep because it's too uncomfortable. Make sure to have the dosage right or you won't make it off the bus, even if someone wakes you.
  • Ciprofloxin: whether it's diarrhoea or any other sort of infection, or bacterial infestation, it's good to have Ciprofloxin with you. If in any way possible, I recommend calling your doctor back home before you start taking it, just to be sure. Antibiotics are really not suited for experiments.
  • Hydrocortisone: insect bytes and sun burns. Use sparingly because it attacks (thins) the skin.
  • Loperamide: see above. Try to avoid it unless you really need it.
  • Bisacodyl: congestions can be quite painful.
Other stuff you may want to carry along includes wound disinfectant (Dexapanthenolum; also any high-percent alcohol would do), fruit/grape sugar (Dextro-Energen), stuff against sickness on cars/boats/etc. (Vomex), bandages, Diclofenac gel, ice spray, a fever thermometer, a bunch of wet hygienic towels, single-use eye drops, steroid-based ear drops, and so on. Your pharmacy will have a complete list. Anyway, given the proximity of medical assistance to the conference campus in Mexico, most of the above won't be necessary. Nevertheless, I think it's important that travellers to countries with lesser hygienic standards than their home countries are familiar with the above, and if only to consciously decide for or against a treatment. In general, I think it's always best to try to isolate and treat the cause rather than the symptoms. And surprisingly often, dehydration is (one of) the cause(s), so electrolytes are among the most important to have with you, in my experience. But again, I am not a doctor.

Christian Perrier: Poland turning right hardly

So, the right-wing Polish Prime Minister has nominated ultra-nationalists and anti-european ministers. One of these ministers has even sustained the actions of the Belarusian dictator, who is currently keeping his country as the last dictatorship in Europe. It seems that such issues are to be faced by most of the new members of the EU. Nationalism and fear of the unknown are pretty common issues around. Even us French people who are so good at giving lessons to others are constantly tempted with such close-minded attitude (remember last year's vote about the Constitution....most Debian friends know about my own opinion about our NO vote, even if it mostly came from supposedly left-wing minded people). I already fear what will happen for our presidential election in 2007. The ideas from the "Front National" are still creeping out and the right wing parties are still tempted to integrate them in their program. I hope that all my friends from the recent EU member countries won't lose hope. We will succeed in building a great european community and I'm confident we will go over these nationalist feelings. It will maybe need another generation but I know my kids will, some day, live in a federal country named Europe. PS: I do not post often on political issues. I posted on that one probably because it has unexpectedly hit my quite hard. I really *hate* all those ultra-nationalist (and ultra-catholic, in the polish case) people.

5 May 2006

Erich Schubert: More on your Google SoC applications

I'd just like to point out that using Debian is not sufficient for sending us your application. Apparently some projects were submitted with minimal wording changes to Debian, Python, Postgres, Gentoo and Apache. If you send us your project, it should have some extra relationship to our project... it should be something our users or our community needs and that is somehow related to Debian. There is nothing wrong with being useable by other projects as well, but the relationship to Debian needs to be a bit more than just "Debian is my favourite distro." For example the SELinux policy module dependency handling is obviously something all distributions would like to have to ease installation. But when integrated with some nice magic of autoselecting policy modules based on the Debian packages installed: go for it. I've been asked about details on the Debtags SoC ideas. Here are some mails for you to read:

4 May 2006

Martin F. Krafft: Back to Bangkok

I spent my last evening in Yangon at "Sandy's Myanmar Cuisine", a wonderful restaurant (albeit a little pricey, I paid $10 for appetiser, main dish, and two beers) overlooking the Kandawgyi lake, attached to the Yangon Kandawgyi Palace hotel; dinner was duck eggs filled with minced prawns, followed by boiled snake-head fish, wrapped in moringa leaves. Quite delicious, although I would have really liked to try their pig ear salad, but they rightfully alerted me that their pig ears weren't fresh. Afterwards, I treated myself to perhaps the second-best back massage I've ever received (the first one was delivered by a massage therapist from Berkeley, whom I had met on the boat down the Mekong to Luang Prabang last year. Before the airport the next morning, I made a stop at the Yangon MICT, specifically the Myanmar Info-Tech company, to meet William, the Myanmar Linux Guru #1. Unfortunately, he was busy at the 5th Myanmar ICT week, so I sat in one of the talks instead and was very pleased to learn that there is an active (but small) group of people working hard on localisation of Pango to the Burmese and Pali languages (the latter being the ancient Buddhist language, and the acceptance of the Burmese character set into Unicode 4.0. I exchanged email addresses with a few of the folks involved, who were well familiar with Debian and Ubuntu, and rather interested in teaming up. I'll forward their addresses to the localisation people as soon as I get back. With still some time left before check-in, I enjoyed my last Myanmar-style breakfast -- Mohinga -- which is a spicey fish soup with noodles, and definitely the best way to start the day; I'll miss it. At the airport, I was getting mildly annoyed once again at their security procedures, which in part required everyone to surrender their gas lighters. Of course, I had four of those on me and in my hand luggage, but when I refused to give them all of them, pointing to the smoker's lounge on the far side of the security check, they held their heads low and gave in. I bet they do this primarily to sell the lighters for some cash on the side, given that nowhere else in Myanmar, or flying to Myanmar, I was asked to do the same. My chaotic self managed to also leave the info brochure from the Info-Tech company, as well as the slides from the talks at the immigration counter, which promptly got me into a questioning by the military as to what kind of documents these were (they cannot speak English in general). With the help of a local, I told them all about localisation and they seemed impressed, but when I finally got on the plane, I guess I felt somewhat of a relief to be outside this tightly-controlled country again. I landed in Bangkok and took a taxi to the hotel, only to discover that thanks to a power outage and the crappy Internet connection back in Mandalay, I had managed to book the wrong one. Instead of central Sukhumvit, I am now staying at Soi 57, which is 4 km further outside of the city. That wouldn't be a problem, for the sky train station is right in front of the hotel, but the staff have not had a lesson in friendliness and generally barked back rather than answering my requests. Plus, since my flight tomorrow is only at 23:30 and I requested to take a shower at 20:00 before taking off (having otherwise checked out and cleared my room), they simply shoved a sign into my face which stated that check-out after 16:00 hours would cost a full day's room rate, and that they did not have any showers available otherwise. Screw that. I'll make sure to let asiarooms.com know in the form of a review. Having settled, and with a bunch of Thai Baht left to spend, I headed for the center, bought myself another Murakami book ("Hardboiled Wonderland and the End of the World"; I seriously love this author), and then proceeded to spend money as fast as I could -- and I am not good at that. I got another massage, bought some bamboo plates and one of those gravity-defying wine bottle holders, a bunch of DVDs, had dinner, and then didn't feel like anything more, so I went back home. As usual, coming back to Bangkok is quite a disappointment (while arriving at Bangkok at the beginning of a trip is a thrill). The city is busy, dirty, and noisy, it's all comparatively expensive (even if you bargain hard), and people are so much more into making money that it's not really possible to get in touch with locals. Then again, I cannot figure out whether my disappointment stems from the shopping mode (into which I was put by several requests to bring back stuff), this being the end of my vacation, or just being back in a metropole after so many rural days. Almost no plans for tomorrow. I might have to make my way to the dreadful Khaosan Road to pick up some student IDs for friends, I would like to (finally) see the snake farm, and I have a lunch appointment with the retail manager of Jim Thompson (the famous silk exporter), who happens to be a friend of Adam, the guy I met in Mandalay, and who introduced me to her -- no worries, I am not getting into the silk business, nor will I spend money there, but the lady is supposed to be quite a character and fun to talk to. Why not? If I wake up, it'll be Tai Chi in Lumpini Park at around 5:30, and if there's time in the afternoon, I'll make my way up north to the biggest market of Bangkok (don't have the name handy). In general, not such an exciting day, waiting to return to Europe. Sigh.

Jonathan McDowell: Xinit considered harmful.

Last year Black Cat decided they wanted to be more organised about offering dedicated servers. We already had a rackful but they were all in use. So we looked around for a suitable supplier. Originally Dell were the front runner; although they have a bad reputation we already had a bunch of machines from them and they'd been reasonable reliable, did serial BIOS properly and were competitively priced. Dom pointed out that having hotswap drive bays would be a really good plan; we'd already specced the machines with dual drives to enable RAID1, but if they were hotswap then we could easily move drives into a spare chassis in the event something went wrong, meaning the customer could be back up and running again quickly while we investigated the old machine. Good plan. This started to push Dell into ridiculous pricing, so we had a look around. We had some machines from Xinit, which again we hadn't had issues with. They were able to give us a price that wasn't that different to the non hotswap Dells, so we decided to go ahead with them. Bad move. On September 6th I informed them we wanted to go ahead. They'd quoted us a 10-14 working day lead time. As we didn't have a credit account already set up with them and it was likely to take a while to sort out we said we'd pay up front in the interest of getting things moving as quickly as possible. By September 22nd we still hadn't heard anything from them, like even confirmation that payment had been received and an ETA for delivery. So I contacted them, received confirmation they'd received full payment by September 13th and was told we should see systems by September 30th or October 3rd at worst. On September 26th they emailed to say there was a problem with their shipment and we'd be receiving a part shipment with the rest to follow. 4 servers were delivered on October 12th. It turned out none of them had been configured as requested (serial console, 9600 8n1, all set to auto power on after power failure), so I ended up having to plug a monitor and keyboard into each machine before connecting it up to the serial console server and power cycler. I chased Xinit again on 24th October. They'd claimed a 5th server was to be delivered before that and still hadn't told me anything about the remaining 15. I was not impressed. A new contact got back in touch and made all manner of apologies and promises about keeping in better contact and keeping me up to date with what was happening. Suffice to say this didn't happen, and calls to the main Xinit number were of no use as no one could help, and the mobile number I'd been given for this contact was unanswered more often than not. On 1st November we received another 11 machines. These were at least configured as requested, but during our testing we discovered they had been misassembled. 5 machines had CPU heatsinks put on incorrectly; visually misaligned and such that any load caused a machine crash. Very unimpressive from a company who supposedly does full burn in tests on machines before shipping them (and indeed used this as an excuse for not shipping us machines sooner on occasion). Further chasing didn't have much effect, but we eventually received the remaining servers on December 5th. Nearly 3 months after we initially placed the order. While Xinit have happily provided many platitudes they have failed to address the numerous failings in their organisation. They are appalling at responding to phone calls or email. They continuously fail to make good on any of the promises they make. They can't provide realistic delivery dates and don't have the decency to get in contact and let you know when they're going to miss them. They have proved unable to assembly machines to any level of competence; most of what they do is buy in chassis and fit parts to them - this is not complicated and yet they managed to get it wrong on several of the machines we ordered. We were promised a machine or two in compensation, but of course these never arrived. In addition, they're so badly organised that in the middle of March we received a spam from one of their sales people who was completely unaware we'd been customers of them! I cannot stress how much you should never consider using this company. I've spoken to others who've bought machines from them or had dealings with them and I know I'm not alone in this. Really. Don't do it. You'll regret it. I suspect it would have taken less of my time and energy to buy all the parts and assemble the machines myself. I wish I'd done so instead. This does leave me lacking a hardware supplier however. We're in the position of needing more machines, but given recent power pricing increases Xeon LVs (the new Sossaman cores) look like the way forward. Except no one appears to have any stock. And even if they do, how do I know I won't get my fingers burned again?

Joey Hess: country

Living in the country is odd sometimes. People come by, take the keys out of the key holder[1] in your truck, drive it up to Washington DC, joyride around hauling trailers, then half a week later, return the truck in dead of night with a new jack added in. At least that's my experience..
[1] That's the little drawer in the dshboard that for some reason has the cigarette logo on the front of it.

Holger Levsen: debconf6-video and network notes...

So J rg has kindly send a request for streaming servers and we got quite a number of replies yet. Thanks! But... we're absolutly not there yet, so if you happen to know icecast and be around during debcamp, please help the video team. (Because we have other things to prepare as well, but we lack icecast experience the most...)

Also, if someone could prepare a ffmpeg2theora backport from sid to sarge, this would also help a lot. Actually, it's libavformat-dev (from sid) which doesnt compile in a sarge environment.

I brought a debian mirror (plus non-free, contrib, security and volatile) for i386 and powerpc, co vering i386, powerpc and sources. If you have an amd64 mirror, please bring it along.

I'll stop blogging now and go to actually see Oaxtepec... See you there or

Daniel Stone: [tech] theo de raadt considered harmful

xf86Init.c in OpenBSD:
Fri Mar 10 17:29:51 2006 UTC (7 weeks, 5 days ago) by deraadt
proper geteuid calls because suse hires people who mistype things

I don't know what went wrong in Theo's life to make him such a whiny, raving sociopath, but I, for one, have had about enough with the sort of behaviour I wouldn't accept from my seven-year old sister, let alone someone who's allegedly old enough to know better. Grow up, Theo.

Erich Schubert: On system init

I've been investigating alternative system inits from time to time the last few years. With the advent of bootchart, there was quite a hype around improving boot times with parallelization and such. Recently this topic has manifested itself in the initscripts-ng group on alioth, and some to-be Google Summer of Code projects. I've been using a modified version of minit for a long time, just lately I was to lazy and switched back to regular init. Since I use suspend2, I don't really reboot my system often anyway. However, minit has more features than being parallel and small. It offers service monitoring and respawning. Which brings me to something missing totally from current linux init systems: service state tracking. Right now, when you start a service, e.g. the web server, you pretty much hope it will never die. Agreed, apache does a good job here. Others may or may not. Mysql uses a really hackish shell script to respawn. And then there are of course tools like monit that check your services and respawn them when needed. There are a couple of things wrong with starting services via SysVInit style scripts from the shell. For example, in order to start a service you need root rights. And then this will "leak" file descriptors from your shell to the daemon (thats why some apps need to be started with "nohup"). Which also causes extra errors with SELinux, for example. And this is just one of the differences from services being started by manually running their init script from the script being run by runlevel changes. Therefore, I think we should move to a new init that can offer full state tracking for services, and is actually used for all services (and not just initial start and getty respawning...) Here's a fairly extensive state diagram I came up with:
state model for sysvinit Some features I'd like to have in a new init: Sorry, no comments on my blog, intentionally. Please use the initscripts-ng mailing list for discussion, thats what the list is for!

3 May 2006

Martin F. Krafft: Eventful times in Yangon

The last 18 hours have been quite eventful, including a mugging, an accident involving first aid, two more confrontations with the military, a visit to the university, and a monk drowning a fly. Fortunately, the heat has gone down quite a bit, the thermometre at the pharmacy now measuring 44 degrees in the shade. When I left the Internet cafe yesterday in the early afternoon, thick clouds covered the sky and it was starting to become quite bearable. Thus, I decided to visit Shwedagon Paya that afternoon, but not before getting a massage (for $3), which was duly necessary after the two hours of almost continuous typing during the hottest hours. I reached the pagoda sometime around 16:00 and almost got into a fight with the ticket office. Admission is $5, and since I am keeping my dollars until all the Kyats are spent, I was asked to pay 8000 Kyats. The current exchange rate (which fluctuates +/- 50 Kyats per day, it's very unstable) is 1450 Kyats to one dollar (it was 850 just one month ago), so I'd have to pay a little more than half a dollar more if I paid in local currency. The lady tried to tell me that she's paying 1550 or 1600 for $1, so I tried to tell her that she better find some other exchange place because she's being ripped off. After ten minutes of a heated discussion, I turned around to leave and she finally settled for 7500 Kyats. You may wonder why I make such a fuzz about 500 Kyats, but the reason is simply that the money goes to the government, which I don't want to support unless I have to. The pagoda itself is quite impressive, rising 100 metres into the sky from atop a sixty metre high. In as such, you can see this pagoda from many places in the city, unless enclosed by tall buildings that usually line the streets. It's said to contain the hairs of Buddha, from when he shaved his head to go and contemplate under the tree in the forest for six years at the age of 29 years, while in Bagan I was told that Buddha threw his hair into the sky, where it hovered and bring luck as long as it remains up there. I suppose he had to shave more than once, so the theories can both hold. I had a guide show me around, but I ended up knowing more about Myanmar history and Buddhism than him, so it was kind of a joke when he asked me for $5 afterwards. I gave him a lecture that he needs to announce his price beforehand and gave him $3, then left him protesting. Again, I am not here to rip people off, but I also won't pay $5 for a half hour tour, when the guide should instead be at home and reading books about the stuff he told. Still, I found the pagoda quite impressive, even though the overcast sky prevented me from seeing the sunset, for which it is famous among tourists. I would come back again at night to watch the supposedly beautiful lighting after dusk, and headed back to the city, with a number of detours thanks mostly to hopping on the wrong busses. As I mentioned before, "yes" apparently means "no" and "yes" and "I don't understand" and "thank you" and almost anything else, if used as a response by a not-so-fluent Burmese. For dinner, I treated myself to a somewhat expensive Sushi dinner, came back to the pagoda to watch the evening life of families and couples strolling clockwise around the stupa (resembling a little the evening family strolls in small Italian cities), found the lighting too bright and still boring, and made my way to the hotel, on foot because I only had a $20 note on me, which is useless when trying to pay for a taxi. ... and then I was mugged; not by some evil Burmese or one of the many muslims living in the city, but by two American rednecks, who apparently ran out of money (and it's very hard to get more without going across the Thai border) and were desperate enough to threaten me with a knife. I really didn't feel like a hero and prepared to hand over the $20 bill, when I saw a military truck drive up the road behind their backs, and remembered the Burmese call for help ("Keh-ba!"), which I shouted just as they were about to pass. The rednecks tried to make a run but had no chance (the military did not even think about using guns, they just called out to a police man at the end of the street, in the direction the rednecks were running). There was not much discussion, the two were loaded onto the truck, and the police officer then explained to me that those guys will be held in prison for two weeks before handed over to their embassy for further prosecution in the US. He asked me if I wanted to press charges against them, in which case they could get sentenced to up to ten years in prison in Myanmar for use of a weapon against civilians (with no right to a lawyer or help from the embassy), but I declined. Curiously, just in the morning, I had read a bit in a doctorate thesis on Myanmar prison life in the bookstore on Merchant Street at the corner of 38th, and it was obvious that the next two weeks for them were going to be hell on earth and enough of a punishment for their misdeed. While handing over my passport, I was assured that the "prisoners" were not going to find out my identity (and I doubt they'll find this blog), and with still somewhat shaky legs and a slight feeling of guilt, I climbed in the back of a police truck to take me to the hotel, wondering a little about how friendly and cool the police were towards foreigners. The next morning, I rose early to confirm my flight reservation and make some arrangements for tomorrow's departure, then headed off to the university by public bus in the hope to talk to the folks at the computer science department about Linux, and to hand over some of the Ubuntu CDs I have left. In Mandalay, it was not possible for me to set foot on the campus, and here in Yangon it would not have been much different, had I not met a zoology professor on the bus, who took me to the campus and refuted everyone's claim that I could not enter. Instead, she called the computer science head professor, who then explained to the officials that she had been waiting for me and would come and pick me up. While waiting, I witnessed a Burmese student just falling to the ground and ran over to help -- nobody else did. With the military folks chasing me (I had not thought about that), I didn't actually have to explain a lot when I knelt down next to the guy, obviously trying to help. The officials just stood close by and watched my every move, as I scrambled to remember my first aid training, diagnosed the guy with a heat stroke, or something in that direction (no sweat, very hot forehead, rapid pulse, white face, mild seizures, uncontrolled movement of the eyes, and delirious lulling), and gave orders to the bystanders to help me carry him to the shade, get water (I had some electrolyte rehydrating solution left), sugar, to have someone fan his head, and someone else call a doctor. Nobody understood anything of what I said until another student with very good English skills translated. Nevertheless, noone wanted to get a doctor because it was too expensive, so I offered to pay the charges and off they went, and I handed over 10'000 Kyats (more than enough) to the translating student with instructions. In the meantime, the computer science professor had arrived, and took me to her office, where I found out that the university is already using Red Hat to teach some of their courses, that she's the vice president of the Myanmar Computer Professionals Association, and that she's also a member of the Myanmar Linux Users' Group. She set up a meeting for me with the founder of this group tomorrow morning, and I proceeded to show off Ubuntu's Live CD to her, and to explain the benefits of Debian over Red Hat, which she understood very well and seemed delighted, because "dependency hell" has been a major problem for them in the past. I left her with half a dozen CDs and information on how she could obtain more, then had lunch with two of the teachers (where I witnessed a monk drowning a fly into his soup while laughing! Imagine that!), who wouldn't let me pay no matter how hard I tried. Upon leaving the campus for the city centre an hour or two later, I was informed that the guy had recovered and was okay, and the officials gave me 3000 Kyats, which I suppose must be the return money from the doctor. Another instance of honesty. I'll spend the afternoon strolling over the market to spend some of my excess Kyats (the rest of which will go to an orphanage), probably get another massage, and finally take in my last Burmese dinner, together with the Burmese white wine I had picked up at the vineyard on the way to Taunggyi, possibly accompanied by the Portuguese traveller I had met before in Bagan and Inle, who's now in Yangon as well... ... which is good, because I found a new argument pro beggars the other day, as I was resting at a street tea shop: I witnessed plenty of Burmese people handing money to a bunch of beggars on the other side of the street. One benefactor then came over to the tea shop, sat down at the table next to me, and answered my question as to why he just gave them money with a simple "because I have some at the moment, and they do not." To me, it could not have been more convincing, and by now, the only argument against supporting beggars is the one raised by the Portuguese: if begging is lucrative, people will give up their self-sufficieny in the countryside and come to live on the streets in the city, in the hope for more money. I'll be looking forward to tonight's discussion. Tomorrow's my day of departure (after meeting the Linux Users' Group founder at the MICT park on the way to the airport), and I'll be in Bangkok for 24 hours before going home. And already the thought of Bangkok makes me sad, because it surely was a great trip in this lovely country, even though I wasn't always happy with how it turned out. I'll be back.

Joey Hess: confession

Hi, I'm Joey. You seem to know who I am, here at this (future) DebConf. I'm familiar with your work in Debian, I probably know broadly what packages you maintain; you've fixed some bugs I filed, and vice-versa. Maybe we collaborated on something or other, exchanged some emails (I'm joeyh@debian.org), worked on something in svn. I might even sometimes hang out in a channel with you on irc (I'm joeyh there), and with occasional difficulty, I can probably connect the nickname you use there with your email address. Maybe we met once or twice at previous DebConfs or elsewhere. Maybe I signed your GPG key. Hell, maybe we saunaed together. And I'm trying to get a look at your nametag, and it's too far away; I need to update the prescription on my glasses. Yes, I might know you on email, irc, and other tcp based protocols, but I have to confess, right now, standing in front of you, I have no idea who you are. And Real Life lacks both google and grep-available. Please, please don't take this the wrong way. Please try to see it as an opportunity to help me connect the you I know with the real you, to merge those two currently completly separate concepts in my head together into one heady cocktail. Some of my best friends here started out this exact same way. Please mention your name, irc nick and/or two or three of the things I should know you for. And please don't be offended if it takes me two or three tries before the association sticks. Once I finally really do make the connection, it'll be unbreakable (that's why I'll be so happy to see Manoj after 6+ years!) but in all other ways my wetware stinks at this.
(I'd print this out on a card, but it's too long, so I'll just write the URL to the blog entry in on my nametag, I guess..)

2 May 2006

Andres Salomon: openswan vs openvpn speed tests

I’ve been using OpenSWAN for a little while now, and have been quite unhappy w/ it. Aside from being poorly documented and a pain in the ass, we’ve run into kernel bugs and stability issues; the various SWAN implementations feel very cobbled together. I also firmly believe that network stacks should be implemented in userspace, and linux’s IPSec stack confirms that belief (Van Jocobson’s network channels display a nice example of how an efficient userspace tcp stack would work). Layering GRE tunnels and using OSPF over the tunnels simply adds to the complexity. Playing around w/ OpenVPN was such a pleasant experience. It has excellent documentation, and was a breeze to configure/set up. I was concerned with speed issues, however; OpenSWAN uses the kernel’s IPSec stack, while OpenVPN has to copy packets between kernelspace and userspace, and uses SSL + Linux’s tun device. I experimented w/ two machines; a dual opteron and a single processor 2.4GHz p4, both w/ Debian’s 2.6.15-1-686-smp kernel. Doing a straight netcat of /dev/zero between the two machines yielded 98mbit/s (according to iptraf). Using an OpenSWAN tunnel and a GRE tunnel [0], I got about 81mbit/s sustained between the two machines. The load on the slower machine hovered around 0.30 while this was happening. Using OpenVPN in UDP mode, I got about 15mbit/s, w/ a load avg of 1.84. Very disappointing results. I was curious if using OpenVPN in TCP mode made a difference; it did. With it, I got about 17mbit/s, w/ a load avg of 2.10. This wasn’t very satisfactory; replacing OpenSWAN w/ OpenVPN when it could only do 17mbit/s over a 100mbit connection, and drove the load way up while doing it was not going to work. Looking at the config, I tried another change; disabling compression [1]. This made all the difference. With OpenVPN in UDP mode and compression disabled, I got 98mbit/s between the two machines, w/ a load avg of about 1.90. This made me much happier; looks like I’ll be replacing OpenSWAN. No need for GRE tunnels over top of the OpenVPN tunnels, either; OSPF works just fine between the tun devices. [0] There are a number of reasons to use GRE tunnels over OpenSWAN IPSec tunnels; Ken Bantoft’s Linux Symposium paper on the subject summarizes them nicely. [1] Most of the traffic that goes across our tunnel links is compressed already, anyways. No sense recompressing it.

Simon Richter: Windows sucks, film at 11

Argh. I need a specific thread to call a function that is guaranteed to be safe to call at any time. Under Unix, I'd have the thread register a signal handler for one of the realtime signals, and use pthread_kill to deliver the signal. Windows has a pretty elaborate mechanism for doing stuff from another thread, named Asynchronous Procedure Calls, however for that to work, the thread that is supposed to call the function needs to be in "alertable" state, which it only gets into by means of a few functions, all of them ending in -Ex, i.e. there are ways a thread may sneak around becoming alertable all the time.

Bdale Garbee: Gender Research

Back in mid-February, I was invited to participate in a review of EU government policy support research on gender issues in Free Software communities at the University of Cambridge. The meeting was fascinating to me not only because of my level of interest in the content, but also because the reviewers were a wonderfully diverse group of genuinely interesting people. I also must admit that being a guest of Kings College for two nights was quite an experience in and of itself... As someone who has been a strong proponent of the idea that approaches and technologies adopted by Free Software developers are something other communities might want to borrow from or emulate, I found the results of this research both disturbing and enlightening. The disturbance centers around just how strongly male-biased our current behaviors seem to be in practice, particularly in the Debian project. My moment of enlightenment was the realization that many of the ideas we discussed about how to encourage and enable more female participation in our communities could simultaneously make our projects more appealing and rewarding to all of our contributors! I've talked to various people about this research in my travels since the meeting in February, and many have followed up asking for pointers to the findings. I was pleased to learn this morning that the integrated report of findings and recommendations from this work are finally available online. I hope at least some of you will read this report, and will then be as interested as I am in looking for ways to change the status quo. Perhaps a BOF at the upcoming Debian developer's conference to discuss these results and brainstorm specific actions we can take in the Debian project would be a good place to start? I expect to arrive late on the 15th, and must depart early on the morning of the 21st...

Martin F. Krafft: The Yangon cinema experience

I did leave the midday heat today for the cinema, but even though I wanted to buy the ticket already more than an hour in advance, it was completely sold out (and the cinema has 500 seats...). I ended up getting an upstairs seat for 1700 Kyats ($1.20) instead of 1000, and went to watch "The Myth" by and with Jackie Chan. The movie itself didn't strike my fancy. Imagine Jackie Chan and Indiana Jones tossed in a tumbler, and out comes this flick, a tale of ancient warriors, gravity defiance inside hard to reach caves, and love, of course, all interlaced with some ridiculous fighting scenes. Call me cheap or boring, but I certainly do prefer Jackie Chan in the streets of Hongkong, battling the mafia or other gangster clans, running up walls, and fighting Kung Fu as opposed to lances and swords. What made the movie an unforgettable experience nevertheless were the 499 Burmese in the room with me (or at least I could not spot another foreigner). The film was in English without subtitles, and most of the people (whole families, couples, little kids, elderly people) came for the fighting scenes and did not understand anything in between: most talked almost all the way through, the crowd growing louder as talking replaced the fighting, and their talk turning into rooting and laughing when Jackie picked up the sword again for the next butchery. The couple next to me were seriously making out, skin contact included, while the three guys diagonally in front of me were obviously drunk and got even more so during the flick. So in the end, I can't really say that I liked the movie, for I barely saw it. But I'll consider doing the cinema thing again, this time on the cheap seats down below, where the real party was apparently going on. "The legend of Zorro" strikes me as a good choice because it's probably pretty crap, judging purely from preview, actors, and director.

Martin F. Krafft: Yangon

I am nearing the end of my trip, and have arrived in Yangon late yesterday. Fortunately, as I got to the airport, there was yet another change in schedule, and I only had to make one stopover in Mandalay as opposed to a second one in Bagan. And if it hadn't been for the French package tourists, wo were nasty (= unwashed, or with a really bad perfume), smelly (I think they all suffered from diarrhoea and couldn't control their exhausts) and rather obnoxious (= loud, and unwilling to speak English, aggrevating themselves why the Burmese wouldn't talk French to them), I might well have enjoyed the trip. I did finish Murakami's book as we touched down in Yangon, and I really enjoyed it. Give it a week, then I'll know whether it has replaced "Wild Sheep Chase" as my favourite of his books. One problem overshadowed the entire day, nevertheless, and I think it must relate to bad food the night before at the Four Sisters Inn in Nyaung Shwe (not meaning to badmouth them, such things can happen): Around noon, I started feeling nausea every now and then, and my stomach apparently meant to tell me that something's not right, my belly feeling hot to the touch. I was then sure I'd come down with another food poisoning, but so far (24 hours later), my body has not revolted yet and the symptoms are fading, so maybe it's strong enough to cope without sending me to bed for a day. Which would be rather unfortunate. The hotel, which I prearranged over the Internet, is nice, and in a very good location. Upon arrival, I tried to score new reading material at one of the bookstores but found them closed already, so I ate some food at the market (not giving my stomach a break, really), had a couple of beers at a small bar, and finally fell asleep to a crap action movie involving a kidnapping and many cellphones on the TV, with the intention not to get out of bed until I could not sleep anymore. And that I did. When I finally went to breakfast, it was almost 9 o'clock, which is averagely three hours later than the last two weeks, and tried to figure out what to do for the day. Of course, I am haunted by bad spirits because even here in Myanmar, which is not socialist (anymore), the first of May is still a holiday, meaning that most sights and shops (including the ENglish bookstore) are closed for the day. I'll thus consider just roaming the streets, getting out of the heat at noon to watch a movie at the cinema (which is supposed to be a big event on holidays), write some more blog entries, and then possibly hire a guide for Shwedagon Paya towards the evening, 2-3 hours before sunset. Thanks for reading along.

1 May 2006

Gustavo Noronha Silva: 1 May 2006

So, long time no blog entry. Work has been sucking my blood these days, but I managed to do some stuff recently that I feel like sharing. First of all, I've been enjoying TurboGears a bunch. Unfortunately I didn't get around to playing with it as much as I would have liked. Some time ago I became the maintainer of cherrypy, and that became an important piece of TG some time later. Now TG has been using something called Python eggs, and its versioning and dependency information. We fought a bunch on the debian-python mailing list trying to figure out how to handle that stuff in Debian packages and it seems like the best solution is the one which ended up on the new python-cherrypy package, which packaged CherryPy 2.1, and right now contains CherryPy 2.2. Most of the efforts of the Debian Python people has been documented on the Debian Python FAQ. Also, let me say that python-support is really cool; Thanks Joss! Also, I'm now a member of the Debian Python Modules maintainers team! Team maintainership is a really good idea, and I'm happy that it is becoming more and more the rule, not the exception; and I am very happy to be part of many very good teams. In other news, I've been working on the new version of libgksu/gksu. The API for libgksu has been greatly improved and simplified. The new design also allowed me to handle some situations in a much saner way; some things that bothered me for a long time are now gone. Packages are on their way to experimental, but if you want to take a look at what is happening, and comment on the new API (please, do!), you can look here: http://people.debian.org/~kov/gksu/libgksu2/ I'll upload some packages to http://people.debian.org/~kov/gksu/debian/ while the packages which are on NEW are not processed.

30 April 2006

David Welton: Rails Annoyances

First, let me say that I'm actually enjoying Ruby on Rails quite a bit. I agree with the large number of people who, like me, are finding that Rails hits a sweet spot between too much structure, configuration and overhead, and anarchy, on the other side. It's good code, fun to use, and for the most part I enjoy Ruby as a language. However, being a bit of a perfectionist, and since Rails is "opinionated software", I thought I'd fire mine right back at it regarding a few of the things I don't like. Mostly minor nitpicks, but I'll likely add to them as time goes by and I use Rails more (I'd like to use it more at work, for instance) - after all, I'm opinionated myself.

Ian Murdock: Linux Standard Base (LSB) 3.1 released

I’ve been rather quiet lately, but that’s because I’ve been busy helping put the finishing touches on LSB 3.1, which was released early last week. New in LSB 3.1 are the integration of the ISO standard LSB Core (ISO/IEC 23360) and the addition of the LSB Desktop platform above the core (which stops at the desktop toolkits in 3.1, i.e., GTK and Qt, but that will move up the stack to encompass more of GNOME and KDE in 3.2 and 4.0). LSB 3.1 also lays a solid foundation on which to build LSB 3.2 and 4.0. Key milestones here include alignment of the LSB roadmap with the roadmaps of the major Linux distros and more direct participation of the key stakeholders in the standard (distro vendors, ISVs, and upstream project maintainers). Much, much more to come on these points as soon as I’m fully recharged. In the meantime, I’m pleased to report we had a very successful launch last week. The release was supported by a who’s who of the Linux industry: Dell, HP, IBM, Linspire, Novell, RealNetworks, Red Flag, Red Hat, Ubuntu, and others. Importantly, no fewer than ten major Linux distributions have committed to LSB 3.1 certification, including upcoming versions of Asianux, Linspire, Mandriva, Red Hat Enterprise Linux, SUSE Linux Enterprise Server and Enterprise Desktop, Sun Wah Linux, Turbolinux, Ubuntu, and Xandros. In short, if you’re an application developer, and you want to easily target all of the major Linux distributions, LSB 3.1 is a great way to do it. More details here: BusinessWeek (and dozens of others via the Associated Press, including ABC News, the Boston Herald, Forbes, FOX News, MSNBC, the San Francisco Chronicle, and the San Jose Mercury News), eWEEK, InformationWeek, InfoWorld (and others via the IDG News Service), Slashdot (twice, here and here), and many other places.

Next.

Previous.