Network messed up completely

"How about let's do it today?"

I have attempted to complete the move from folk to western yesterday; and while I was at it, moved some other hard disks and computers around, too. Osamu Aoki had given me a computer he didn't need anymore – a dual Celeron @ 433Mhz, which is probably faster for some tasks than my PentiumIII @ 650Mhz – so I used rock, my desktop, to replace pop, my parent's box, and used the new box to replace rock.

Of course, nothing works as it should when you attempt to do such a thing.

rock didn't like the hard disk from pop, and when I attempted to install windows, it found a lot of errors for some strange reason. It took me a while to discover that this was out of the ordinary, but of course by that time the damage had been done -- the entire 30G partition containing most of the data on pop was corrupted. With (almost) no backups. Aargh!

western didn't like the second network card I put in the machine, so I only have one (X less) box with an Internet connection at home currently. Grmbl.

I put pop's hard disk aside for the time being, and will try to find out what went wrong sometime this weekend. The Internet connection isn't as important. But for the record, I'd just like to say that hardware sucks. Majorly.

NM graphs

Christoph Berg created some NM graphs, but has a problem with one of them where stuff overlaps too much.

I don't know neato at all, but I do know that there is a tool called 'springgraph', which supports a subset of neato syntax. Not much, but I think enough for the graphs he's using; and since springgraph calculates node distances dynamically, using an algorithm based on the number of connections, it might solve his problem.

Shitty day

Today isn't my day.

• I had an argument with my parents over something silly.
• I got a splinter in my hand, which is giving me pains, and nothing to remove it.
• My USB stick containing my GPG key broke down (yes, I still have a backup, but it's 40km from my current position).
• A customer that owes us a lot of money still didn't pay.
• On top of that, I got a bit of a headache.
• I fucked up my firewall, and can't access my mail.
• I forgot my laptop's adapter at home, so now that the battery is empty I'm forced to use this dog slow desktop box.

Grmbl.

Darn.

"This looks like a nice place for a camp fire."

There was an item on the news about a failed bank robbery attempt in some south american country (I think it was equador, but don't take my word for it). Originally, there were three robbers, but a security guard killed two of the three when they tried to enter the bank. The one remaining robber managed to take some customers and staff of the bank hostage. His demands? 40.000 dollars and... food.

What happened to the getaway car?

Printing

Lars blogged about printers, mainly about how he's dissatisfied with the current state of affairs. I agree that a state of affairs as he describes it would suck. Luckily, reality is a bit better

You really should try installing CUPS, and the foomatic-filters-ppds package¹. Once that is done, edit /etc/cups/cupsd.conf on any machine having a printer connected to it, and find the option "BrowseAddress". Enable it (there are enough comments in the file explaining you how to do that; make sure it doesn't send out stuff to machines on other networks). Once you restart cupsd, it will start to broadcast a list of printers available to it, and other machines in the network also running cupsd will pick it up (unless you disabled browsing on those machines) and make the printers available to the local system.

Using this setup, I do not have to reconfigure my laptop every time I want to print; both at home and at the office, I have systems broadcasting their printers; and all applications that directly support CUPS (rather than using lpr) will show me a list of printers available on the network. So, the network part of Lars' problems are already solved.

That leaves the manual setup of printer queues. Cups allows you to do this via an easy web interface (which, by default, only listens on localhost), but there are other options. Unfortunately, though, I don't think it's possible (yet) to make it as easy as Lars would want. But it's not as if it's hard...

¹based on the views expressed in that packages' description, one could say that the maintainer of said package would probably disagree; but I find it much easier to just install a full list of prebuilt files rather than having to manually run foomatic-ppdfile and having to remember where to put the file.

Pope getting (too much) email

Those campaigns to get people of age on the Internet seem to be working.

According to a press release by the vatican press agency, the pope has been so smart to post his email address on his website in 6 different languages, and now gets over 20.000 emails a day, all wishing him well. Wonder how he'll be able to read all of those, if at all.

And how long it'll take before spammers 'accidentally' put the Pope's email address on one of their lists, and he starts getting ads for penis enlargement pills or generic vi/\gra.

I'd love to learn about his reaction on that one.

Received a mail from the PGP Global Directory server

Some people have received it ages ago, but mine only appeared today. Which isn't entirely unreasonable, given that there are thousands of PGP keys on the key servers.

In itself, the PGP global directory is a good idea. They want to have a directory of only active keys, instead of a directory containing all keys of all people all over the world, including those keys issued by people who just tried playing with pgp or gpg for two days and forgot about it, people who are dead and didn't issue a revocation certificate before their dead (or didn't have an expiry date on their key), etcetera.

Only it would've been nice had they actually thought about what they were doing.

Some of the people who received a mail from the PGP Global Directory already mentioned that the mail from the directory completely bypassed the PGP trust model, so it isn't actually something one should trust.

This one is even worse, though: I received a mail for a key with id 0x4B5A3DFD. This probably means nothing to you; but let me tell you that about nine months ago, my laptop was stolen, and that the key has been revoked ever since...

If they want to check whether keys are still active, they could at least have checked whether a key has some sort of revocation certificate on it before proceeding... idiots!

What the hell is this?

Script started on za 12 mrt 2005 15:41:52 CET
wouter@country:~/debian/eID/belpic-2.3.13$ dh_strip -plibbelpic0-dev
	strip --strip-debug debian/libbelpic0-dev/usr/lib/libbelpic.a
strip: debian/libbelpic0-dev/usr/lib/stkApu2z/libscconf.a: Invalid operation
dh_strip: command returned error code 256
wouter@country: /home/wouter/debian/eID/belpic-2.3.13
Script done on za 12 mrt 2005 15:42:05 CET

I'm not sure what prompts that error, and I'm not really into the ucky details of how toolchains are supposed to work, either. Hints are welcome.

It's funny to see people who know the word 'SPAM' as 'unsolicited email' find out that this is not the original meaning of the word.

There's supposed to be a scetch in Monty Python's Flying Circus (I never saw it myself, unfortunately) with some guys that keep singing SPAM, SPAM, SPAM all the time, and unsolicited. Very much like the SPAM mails we receive today. Someone – a fan of the show, no doubt – saw the similarity, and gave UCE the name we mostly use today. The company that makes SPAM doesn't mind, apparently – although it insists it has nothing to do with unsolicited email. Naturally.

Unhappy about the release meeting's outcome

Very unhappy, in fact.

Due to some circumstances, I actually knew about this meeting for quite some time, about one or two weeks before it actually happened; but I understood they were going to talk about the release schedule for etch, which is something I'm interested in, but not so much that I really, absolutely, want to have to say something about it.

Then they suddenly decided to discuss the viability of having 11 ports. Now that, I do care about. It's disturbing that people can discuss this without involving those people actually working on the port. And no, it doesn't appear as if they actually talked to any of the m68k buildd maintainers (or any porter of any of the other architectures, for that matter) before deciding that it might be a good idea to essentially kill some architectures. Yes, that is what it'll do; providing unstable snapshots is a joke as a release strategy.

This is depressing. After fighting customers to pay me and European legislators to not institute software patents, I now suddenly have to fight ftp masters not to kill off stuff they're not involved in (anymore, in some cases), too.

I'm not sure I still see the point of all this.

Evolution and large mails

Part of being a buildd admin is that one, at times, gets pretty large mails in plaintext. Such as this one:

  20 r   Mar 15 Quickstep Build (5,1M) Log for failed build of tetex-bin_3.0-1 (

Currently, I mostly read mail in evolution, and only do a few bits in mutt. One of the things I do in mutt is signing successful mails, because evolution isn't as easily scriptable as mutt is. Failed mails are usually replied to in evolution, though.

Not this one. When I hit the 'reply' button, it almost choked on the mail, and required a lot of processing time just to open the window. Really bad...

Don't get me wrong, please.

Daniel Silverstone seems to think that, in my opinion, any reduction in the support of minority architectures is something I do not want to talk about.

That's not correct.

But this proposal, as it currently stands, will kill off m68k. If there are no stable releases with security support, I know for a fact that people won't be using it. I upgrade my unstable m68k box enough to know why.

I have no problem with us being sent to a corner of the Internet, one you have to look very careful for to find it. We are a minority architecture for a reason, and there's nothing wrong with that. I also have no problem with moving the strain of supporting the port onto the porters themselves, if there are problems for other people. However, the port should always remain a viable one; and only allowing us to do unstable doesn't do that.

Daniel also makes a few comparisons, ones that aren't actually fair.

Even though amd64 has proven that it is possible for an architecture port to survive outside Debian, I don't think it is possible for m68k to do the same. Amd64 is shiny, new, and hot; m68k is not. I don't think it is hard for the amd64 port to find new interested people should they have more work to do; but that doesn't mean that the same is true for m68k. In that light, I have no problem with us doing most of the work is that is necessary; but I do have a problem with us duplicating work that is already been done elsewhere. There's no point in that, and it might kill our port off, too.

It is also not quite fair to compare the m68k port team to Ubuntu. The latter consists of a staff of ~30 full-time paid employees; the former exists of a staff of (currently) ~7 people who manage this in their free time. Sure, if you've got ~30 full-time people, it's completely feasible to take an unstable snapshot and turn it into something nice and stable. The same probably isn't true for 7 hobbyists.

It's not that it doesn't make any sense to reduce the support for minority architectures–we knew that would one day happen–but when the day had come, we should have been part of the discussion. Even though it is officially a proposal, and not a given thing, it certainly didn't feel that way; and the feeling that the work I have been doing for the past four years would go down the drain at someone else's whim, well, made me very unhappy. With a proposal like this one that came totally out of the blue for the people involved, don't you think some of us are tempted to throw in the towel in disgust? Gee.

Don't want to read -devel

I have 247 messages in my -devel mailbox, after I spent most of monday and yesterday reading it.

I don't want to. Need courage. Need courage.

...

Okay, let's go. In an attempt to concentrate on this, I switched off everything now. No IRC, no phone, no IM, no nothing. Should work.

I hope.

"Barbie OS"

This is cool.

... if it's true. Can't find anything about it on the mattel site.

NBD: not enough magic

There's a bug report that I keep receiving against NBD upstream, one that is important for people trying to use it in high availability situations, but also one I cannot locate the source of. This is rather annoying.

When people try to use NBD over a rather fast line (gigabit or above) it will work okay for a while; at some point, however, it suddenly stops working okay, and exits with a message 'Not enough magic'.

I know where that message comes from; the NBD protocol requires every packet to begin with a 'magic' number, and the connection is dropped if the magic number isn't correct. However, at the point where this error message appears, the connection has been up for quite a while already, and the magic number is the same for every packet; it's unlikely that it would suddenly be wrong. Thus, I can think of only two reasons why it being wrong would be the case:

1. The network somehow mangled a few bits in the data; as a result, a packet's data is corrupt and the magic number isn't correct anymore. I would be surprised if this were the case, but then it isn't entirely impossible. This would mean that gigabit lines aren't entirely error-free[1], and that the TCP stack in the kernel doesn't handle corruption very well. I would find that quite strange, and don't think this is the case.
2. There's a bug somewhere in the kernel (the TCP stack or the nbd client code) or in nbd-server that's only exhibited under high stress, and that results in it not properly handling the magic number or so. This would be even weirder.

If there's anyone with hardware knowledge enough to hint me as to whether the first is likely, I'd appreciate that.

[1]in contrast to (at least) 10Mbit, which is what I used to test it for quite a while, and which does not exhibit that behaviour; I have a 100Mbit network hub since recently, though, and I'll start testing using that thing as well soonishly.

Now reading: 'Ik was twaalf en ik fietste naar school' ('I was twelve, I took my bike and went to school')

The book by Sabine Dardenne, about her horrible ordeal at the hands of paedophile Marc Dutroux. This isn't the original French book, because my French is horrible. But I'm assuming it's going to be an interesting read, although not really comforting.

Nice weekend

Had a nice – or should I say wonderful? – weekend.

Saturday morning, we left for Heverlee, to St. Lambertus' Chapel, a very nice Romanesque church from somewhere between the 11th and 13th century. Arriving there, we helped mom set up the flower decoration she had been preparing, and Roel and I tried out the piece we had been practicing for the last two months. There were more musicians there, of course — another one of the cousins and a few friends of Marian and Thomas, the bride and groom. Most of them practiced their piece, too.

At 13:00 (I think), the marriage service started. Afterwards, we went to my cousin Koen's place (who lives nearby) and had a chat and a meal, waiting for 18:00 when the evening feast would start. Went there and had a lot of good food, and equally much of fun.

On Sunday, got up late and had brunch. After that, went to Berchem where the piece dad had been directing was being performed. As this was the last time they would perform, helped them break up afterwards. When that was over, it was about 7 pm, so we got ourselves some food at a local takeway Chinese restaurant, and ate that (the food, not the restaurant).

And it's monday now. With over fourhundred messages in ~/Maildir/debian/devel. Grmbl.

DPL vote

It's that time of the year again, the DPL vote is near. I've been thinking for a while about proposing a constitution amendment to increase the time where we have to vote from one year to (at least) two, as I think the yearly campaignings aren't exactly doing the release much good; but let's not go into details on that one right now.

As seems common on Planet Debian, here's how I voted: Matthew Garrett at one, NOTA at two; and I didn't order the others.

Why Matthew? I've seen Matthew at FOSDEM, and some people who know how he led another volunteer-based organisation based in the UK, and who had only positive things to say about him. They know who they are, I'm sure they'll speak up if they want to. Additionally, I read some of the campaigning on -vote, and among all candidates, his answers made most sense of them all, IMO.

I didn't place any of the other candidates above NOTA for a few reasons.

Jonathan Walter would surely have been first on my ballot had it been a vote for 'Who should we kick out of the project', but this vote is a different one.

Anthony Towns is a great guy with a lot of skills, but I don't think he's the perfect guy for a DPL; IMHO, the most important skill of a DPL is to be able to disconnect oneself from a discussion (even if they are interested in the subject), to post clear, concise summaries, and/or to try to steer the discussion away from flame and into 'the right' decision, whatever that may be. If Anthony has that skill, I surely never saw him use it—he especially seems to be prone to post long, entangled mails rather than clear and concise ones, which isn't exactly a good thing.

Angus Lees may think that he's perfect for the job because he hasn't really participated to discussion on many mailinglists before applying, but I do think a prospective DPL should at least be involved into Debian to more extent than that. Also, and this goes for some other candidates as well, I don't think it's healthy for a DPL to jump around positions as easily as they did on the SCC proposal; only a few days before that proposal, they express their opposition to the general idea of dropping architectures from Debian, but when a concrete proposal to do so is there, they agree to be listed as supporters of that proposal. How these two can combine isn't exactly clear to me, and it smells of 'no character'.

As to the two 'Project Scud' candidates, I'm not very fond of that idea. Having a small 'inner circle' of people who have no real powers according to the Debian Constitution isn't exactly a good idea in my view; I tried getting the position from the 'project scud' members on that point, but didn't get an answer to my satisfaction.

I find it especially unfortunate that I feel compelled to not rank Branden Robinson above NOTA. I have done so on all previous DPL votes where he was a candidate, because I think he has a clear view on what Free Software in general, and Debian specifically, should do to face the future; but there are a few ideas he supported recently that I do not want to see implemented.

Let's call it 'nybbles'

There's been some discussion in the SCC thread about how 'Second-Class Citizens' is a bad name. I happen to think most of the idea is bad, but if we're going to implement it, we should call it 'nybbles.debian.org'

Screenshot

There's a meme going on over at Planet Gnome about what your desktop looks like. Looks like fun. here goes:

Yeah, sue me, so I have a company.

The fun part about the screenshot, though, is that the above image (the one I resized to 640x480) is about 50k large, whereas the original 1024x768 one is about 46k. Heh.

IPv6 back up

A few weeks ago, I removed folk from my network, because what it had been doing was now mostly migrated to western, added an extra network card to the latter box, and configured it as gateway to my network. It mostly works (except when there's a huge amount of mail, which seems to happen more than it used to, now), but I had forgotten one detail: IPv6 routing hadn't been migrated yet. Fixed that, now.

While reading the radvd.conf(5) manpage, however, I remembered something interesting: 'Mobile IPv6 extensions'. This has 'something' to do with laptops, and has intrigued me ever since I first heard about it, but I have no clue what it's supposed to do, or how it's supposed to work

Searching the RFC Index doesn't reveal anything. Most likely because it's not there yet. Googling does reveal an index of draft documents by some IETF working group related to that subject, but the link to the document about "Mobility Support in IPv6" is dead. It also turns up some site by Lancaster University (whatever that is), but I get a connection timeout when trying to look at that site. And since they talk about Microsoft IPv6 stacks, I don't think it'll be of much help.

Grunt. And I don't even know whether it'll be able to do what I want it to do.

Update: I got an email from Rui Tiago Matos, stating that the mobile IPv6 stuff for Linux is over at Mobile-IPv6.org. Nice. That site also included a working link to the current RFC draft. Reading that thing now, let's see what it can do.

Rui also told me about the Daidalos project, which seems to be a government-sponsored project, with an interesting preset:

The Daidalos vision is to seamlessly integrate heterogeneous network technologies that allow network operators and service providers to offer new and profitable services, giving users access to a wide range of personalised voice, data, and multimedia services.

Cool.

I'm a dumb fuck.

see why.

And today, when I wanted to put a second installation on my laptop, I was confused and wiped out my 50G main partition instead of the 10G one I keep around for stuff like this. Yes, I have backups, but I probably won't be able to fix it 'till monday.

Darn.