So sprach Ron Heron am Fri, Nov 17, 2000 at 06:38:35PM -0800: > This is a matter of how you look at it. If you have > an ext2 partition, you can make that partition > reserfs. The issue is that you copy the contents of > the partition to a temp location, run mkresierfs, and Okay, if you have enough space available, then you are right. But basically you don't _convert_ the ext2 fs, you totally scrap the fs, and put a reiserfs on the same partition. But I suppose that's only nitpicking. > is great, however for mass storage. Ext2 takes 22 > minutes to do e2fsck on 2 15G drives, reiser takes 22 > seconds! Plus reiserfs doesn't waste as much space on big systems as ext2 does, and you can't run out of inodes - you can't because reiserfs dynamically creates them. > If you don't have a commercial system, and want that > nervous, fuzzy feeling about using bleeding edge > developments, go for it. Otherwise, just stick to the > old ext2 and be happy. Hmm, I'd rather say it the other way around. Granted, I haven't tested it on a heavily loaded production system (because I don't have access to such a system), but I would try it even there. Simply the reason that it is a journaling system would be a good enough reason for *ME*. Plus there's a speed advantage, and that I haven't had any problems with reiserfs in the last, hmm? year?, or so that I use reiserfs exclusively (well, about at least. I still have a 100meg ext2 boot partition). What exactly are the reasons that you would not trust reiserfs for a production system? Do you know of any scenarios where reiserfs always fails? I know that there are some issues with NFS on a reiserfs fs, and I know that there are some troubles with kernel software raid. Anything else? Alexander Skwar -- How to quote: http://learn.to/quote (german) http://quote.6x.to (english) Homepage: http://www.digitalprojects.com | http://www.dp.ath.cx GnuPG ID: 59F6A6F5 FP: DC8AFA56C492EE6058D5 BAA62EEE3AD559F6A6F5 ICQ: 7328191 Uptime: 3 days 7 hours 7 minutes
Keep in touch with http://mandrakeforum.com: Subscribe the "[EMAIL PROTECTED]" mailing list.
