Woke up a bit later, but still managed to get in time for the presentations.
I started the morning with Jeffrey Moyer's talk about Automounter, then I Catched up on some mailreading, figured out the default gw
for wireless so I can actually read my mail in the talks :) Then continued with the Logfs talk from Jorn Engel , the thing that's
probably still the most annyoing about specialised filesystems for Flash is that we might be using them in some Linux based projects but the rest of the planet will continue using fat16 or some other antique type.
Our camera's and the USB disks we share with other people still are in a wrong type. I've been using different filesystems on my usb stick but the moment you want to copy something you have to create another filesystem on it.
Then went on to Harald's update on Netfilter which I should actualy have skipped since I knew lots of stuff from it David Millers Keynote at the UKUUG already well, at least it all sounded familiar :), I should have gone to Ben Martins talk :) I learned about nfsim , the Netfilter Simuation Environment however.
I opted for Volker's Clustered Samba talk. Lots of stuff in Clustering samba seems like it has already been done before in other
Clustering environments however the moment you realise that this stuff also has to work on about some totally other platforms such as Solaris,
HP/UX and AIX and also on top of different filesystems. Quoting Lars(MB) on that topic: "You have my sympathy" Always interresting to see
audience feedback from people such as Lars and Alan who actually give valuable input to a speaker who learns during his talk.
Everybody seems to be testing stuff on GFS/GPFS but not that many people are using OCFS2 yet, at least that how it sounded in Hamburg. Volker has been doing GFS / GPFS
testing, Philip told us yesterday that he didn't have time yet to do OCFS2 testing, he acutally mentionned that it was because he couldn't get it to work in 1 day, so he dropped it.
The conference closes with a Future of Linux Distributions discussion , opening with the question "Are there to many distributions and isthis a problem ?" Joey, of debian security fame, notes that we might need to look at why people start working on different Distributions, are the currentdistributions not listening enough ? Off course some distributions have their own specific goal, but a fork just for the reason of forking or just to be able "I maintain my own linux distribution (wich is used by me, my grandmother and my niece)" , that shouldn't happen.
There was some discussion going on about what's still missing in Linux, with suggestions ranging from Process migration , which obviously already exists for ages as in the form of the openMosix project trough automated system management tools. (I`m still wondering on which planet the guy who asked for those has been living for the past say .. 5 years.)
Headed back to the hotel and then into Hamburg City to eat .. a Hamburger with some other .be people