<p>Thanks for all your hard work, you can rest assured that nobody will be throwing rocks :-)</p>
<div class="gmail_quote">On Jun 7, 2011 10:11 AM, "Alexandre Oliva" <<a href="mailto:lxoliva@fsfla.org">lxoliva@fsfla.org</a>> wrote:<br type="attribution">> On Jun 6, 2011, Alexandre Oliva <<a href="mailto:lxoliva@fsfla.org">lxoliva@fsfla.org</a>> wrote:<br>
> <br>>> On Jun 5, 2011, Alexandre Oliva <<a href="mailto:lxoliva@fsfla.org">lxoliva@fsfla.org</a>> wrote:<br>>>> It's still not clear how or even whether it did. I *might* be related<br>>>> with it: both mentioning in public the vulnerabilities of the system<br>
>>> and bind-mounting the repository in /tmp.<br>> <br>>> Bernie (FSF admin) tells me the VM host can no longer see the external<br>>> SATA disk that held storage for our VM guest. He's going to the colo<br>
>> site on Monday to see what gives. This suggests the problem has to do<br>>> with disk failure, rather than any software issue. Hopefully<br>>> reconnecting the disk will bring it back up, but we cann't rule out the<br>
>> possibility that the disk died. So, please cross your fingers and pray<br>>> to your favorite deity or to the Big Bang ;-)<br>> <br>> And so it turned out that the disk could be plugged back in and our<br>
> machine booted up fine.<br>> <br>> However, tmpwatch had already done its thing, cleaning up everything<br>> that hadn't been accessed for 10+ days within /tmp. Because of my<br>> bind-mounts of /mnt/blag_repo in the chroot I used for testing inside<br>
> /tmp, this amounted to a lot of the ftp repository.<br>> <br>> I'm deeply sorry for not thinking of the possibility that tmpwatch would<br>> cause this.<br>> <br>> As for recovery of the lost data... I'm waiting for an answer from<br>
> Bernie on whether they do have backups of our system, for he implied<br>> they might have backed it all up recently.<br>> <br>> Regardless, I recommend those who know how to rebuild/complete at least<br>> the 140k repostory out of our upstream sources to do so. I have torrent<br>
> seeders running for the final isos, and I have local copies of the<br>> 140k-pre isos as well, as well as BLAG-90001-i386.iso, and if it would<br>> help, I'd be glad to upload them. However, my upstream bandwidth is not<br>
> exactly great, so it would be a while.<br>> <br>> Again, please accept my apologies for my deeds towards this mess.<br>> <br>> Weyasey, it is now time to reach for those rocks. Please aim carefully.<br>> ;-D<br>
> <br>> :-(<br>> <br>> -- <br>> Alexandre Oliva, freedom fighter <a href="http://FSFLA.org/~lxoliva/">http://FSFLA.org/~lxoliva/</a><br>> You must be the change you wish to see in the world. -- Gandhi<br>
> Be Free! -- <a href="http://FSFLA.org/">http://FSFLA.org/</a> FSF Latin America board member<br>> Free Software Evangelist Red Hat Brazil Compiler Engineer<br></div>