Dude is trolling. And we have taken the bait.

Since when expression of a criticism over what I believe is likely to be the worst server swap in the history of Internet, is called "trolling"?

You guys have been stroking each other heads for two weeks and managed to announce a success several times already... yeah, great success!
If you think that I am trolling, then I strongly advise you to look in a mirror. Because I only express my criticism, over how this process has been handled. And maybe if some of you did the same two weeks ago, instead of thanking for nothing, the problem would have been solved already. Negative reinforcements often work better

Piotr, we had about 18 Virtual Machines Physical Servers, running different parts of maemo.org plus two bare metal servers running the builders, all interconnected with one of them serving the static assets and another serving the database in DMZ, not taking into consideration tmo which was completely separate. Definitely more than a lightbulb.

Nokia decided thankfully to help, but as a big corporation, it thinks differently from us so the help while welcome added some more difficulties to the mix, for example the possible need for a double migration.

Most things went according to plan with some small outages. The repositories however were a different beast. Back in the day, there were so many N900's that Nokia had been using Akamai server farms for the mirroring of repository.maemo.org. That means that tens or maybe even hundreds of servers around the world had mirrors of repository.maemo.org and data was served by the nearest to you using DNS wizardry.

When the repositories were moved, we tested them again and again, and everything was working alright. Changing the DNS is a bit of a lengthy process (until they propagate throughout all the caching servers in the world) and something that was of a paramount important to us. To have the DNS is having maemo.

After successful testing of the new servers we requested the DNS change from Nokia and when that happened the ~X00 Akamai servers attacked our little new VM thirsty for updates, taking the whole system down. The only way to be able to get the rest of maemo.org up was to shut repository.maemo.org down.

Until today there have been many attempts to turn this machine on but it still drags everything down for various reasons, mostly related to traffic shaping, firewall configuration and other technical aspects far away from my realm of knowledge. Reverting the DNS to the old server, assuming that it still exists, is NOT an option, we may not be able to get it back for ever.

However, because of the efforts of people here in the community, 99.9% of the things you may or may not need are already backed up and ready for you to use, in servers that are paid by those peoples' pockets.

Instead of wasting your and other people's time with uninformed accusations and flamage, please use that time to visit this thread which is also already prepared for you with nice screenshots and is, at the time of writing, in the active topics on the right side of the screen, to solve your problem until our repositories are back up.

tl;dr; Not only servers were checked, backup was already prepared and live.

Thanks for explaining, @qwazix.

Yes, I know about the mirrors, but don't feel quite comfortable connecting my phone to the servers that are paid by those peoples' pockets.

And from what you guys are saying, this is exactly what I was thinking; all the delays are caused by Nokia, their paid employees. Which is quite obvious if you see that the process never moves forward during a weekend, while volunteers usually work during weekends.

So I only wonder how many more weekends do they need to get it back up..

Don't take it personal.

So it's better you throw your device into the sea...

...i think we have a winner.

Oh, here we go: this would be a perfect example of "trolling", as for my dictionary

But going back on topic.
Personally I don't believe Nokia would have switched off the repos, at least for several more years. Keeping the servers running wasn't a big cost for them so I seriously doubt that they would risk spoiling their PR by switching off the servers. Not to mention that there would obviously be some legal concerns, because there is a significant difference between end of support and making 40 thousands of 2-3 years old devices, $400-500 each, practically unusable.
So maybe that is the reason why I am not as enthusiastic and thankful about this, as you are.

Anyone knows when repositories will be working? 2 weeks maybe?

I am sorry for you, but the facts are there. Please familiarize yourself with that before mouthing your opinions.

For Nokia, Maemo is end-of-life, there is nothing here to see any more. As for the current state of the company, N900 and it's ilk are embaressment for them. Currently it's all WP in the company, and they would rather forget their Linux legacy. (and indeed, have worked hard to eradict it totally...)

The way they see things, it causes no bad PR to leave the Maemo devices without support. As was stated before, the legal oblications have been filled. As for moral obligations... please

Well IIRC, the Nokia paid hosting contract ended on 31. 12. 2012 - after that the hosting company was free to just shut everything down & delete all the data. Even if they on a whim decided to just leave it running for a while, there was severe breakage (wiki, autobuilder, etc.) without anyone being paid to fix it while at the same time out of reach of the community.

So the migration was needed, was the only option (other than letting everything just die). And I hope everybody that feels to be offended by community volunteers helping to keep the Maemo infrastructure running to the best of their ability in their free time without even being paid a penny to thing a bit before verbally attacking them.

Basically: Put up or shut up.