I had not upgraded from F7 to F8 because I was waiting for this, which I had heard about. I downloaded the F7 version and found that the only upgrade option it gave me was an upgrade to F8. I figured that this would be OK, for F8 would give me the ability to upgrade to F9. It downloaded files OK, but hung up during the phase where it was looking for dependencies. That’s the bad news. The good news is that I was able to reboot into F7 with no hitches. I’m going to try to do this again but with more space in the /boot partition…

Here’s to moving in the right direction!

MVP

greetings, eMBee

greetings, eMBee

>Not with LVM. It won’t be efficient enough to do a lot of snapshots like you want. Other solutions are either not accessible (different licenses, proprietary, patent issues etc) or alpha level (btrfs) currently.

Thanks for your input, very valid points, really but let me hilight that I pointed just the direction we should be working on (IMHO) and thats the snapshots with copy on write capability. All other solutions seem to be overly complicated long term and don’t even provide very much needed easy cutback possibility.

I do know well current licensing problems, patent issues and the technical limitations of LVM etc., but without vision ‘there must be better way managing filesystem’ Jeff Bonwick et al. would not have not invented ZFS, right?

I’m already quite old unix/linux/bsd *art and just like to remind you that we’ve seen quite astonshing development since what the filesystem state were with AT&T System 7. I don’t think we would be with all these features if we had given up and not even tried.

Butter fs (btrfs) and NILFS both seem good start. ZFS got head start, but I bet that competition can make even ZFS better and if Sun doesn’t see the light (pun intended) by double licensing or whatever needed to get ZFS widely abopted with Linux there something else will replace it with competent features.

I was just encouraging and pointing out which direction we will have to in long term go.

Not that we have all the needed tech available quite yet, unfortunately, but it will happen sooner or later. Also it will likely happen sooner if we all understand the benefits and are able to push together, right.

Knowing that giants like Oracle and NEC have taken on the challenge this is much more likely to happen compared to the early times when Linux was young and simple, so simple that I could even get it and patch features or fix broken issues to kernel fairly easily :)

Now it’s so complicated that I don’t have time to delve in, but certainly will still be able to have a clear vision which direction the solution will be found.

Cheers, and keep up good work with Fedora

:-) riku