As far as I know there aren't any official, fully tested stable releases of Zenwalk any more so this is probably beta quality. Unfortunately Zenwalk also does not seem to have any users these days so there isn't much scope for testing anyways. Nevertheless, as a former Zenwalk user, going back about 12 years or so, and somebody who enjoys using Slackware I thought it's time for catching up with this lovely, minimal and fast distribution. At least that's how I remember it.
While in the installer I launched cfdisk to resize the partition which worked a treat, rebooted, installed the old Zen and off we went. Yeah, until I tried booting into the other distribution on sda1, which btw uses the ext4 file system by default for new installs. Which I always thought was a bad choice but had gone with this time. It had been a "give it another chance, what can possibly go wrong, surely after all this time it has matured" decision. Gparted confirmed the partition table was corrupted and simply trying to fix it did not work. Next step: SystemRescueCD. A truly fine piece which I can only recommend to anyone in a similar situation. It has about anything one could want onboard for file system errors and data recovery and it actually works. It wasn't able to fix my partition table - even testdisk complained that "it is too corrupted for me" - but recovering worked great. As it was still able to read inside my partition I was able to transfer all relevant files, i.e. the ones deemed worth keeping, to another drive, several gigs of it. That's right, all the data was still in my corrupted partition and just waiting to be read.
So far so good. A format later a backup of that distribution is back on and it's almost as if nothing happened, only this time we're using ext3, a file system I have always found utterly reliable in the face of power failure and other adversities. Unlike it's successor. Or maybe it's just me doing the wrong thing at the wrong time to the wrong type of file system and my actions are driving it up the wall. Maybe I should have used gparted in the first place to partition the drive further and not cfdisk on a beta quality distribution release, who knows. But a bad taste remains as this is not the first time ext4 shows it's precarious nature.
Ext4 is of course the newest in the ext-family of file systems and was introduced into the Linux kernel in 2008, so it is still relatively new. Fedora adopted it early on as it is a experimental testing ground distribution by nature and is among the first if not the first to adopt new technologies, like it did with early switches to pulseaudio,grub2, systemd and virtually everything else in the linuxsphere. Sometimes I wonder if other distributions only adopt technologies because they are in Fedora already. Any quick internet search reveals a plethora of results for corruption in ext4 file systems.
Due to similar past experiences I have been using reiserfs, xfs and the older ext file systems exclusively for about six years and have never had any problems with file corruption or unreadable partition tables on these machines in this time. The reiser4 implementation is also a younger file system but seems far more robust than ext4. If you want to do yourself a favour and not run the increased risk of having to salvage your partitions, use jfs, btrfs, zfs or what ever else, but with all this choice around there really isn't any reason to use ext4 or to have it as the default file system in distribution installs.
Testdisk and SystemRecueCD are very recommended.
HowTo Repair Broken ext4 Partitions.
Interesting post! This is really helpful for me. I like it! Thanks for sharing!
ReplyDelete- spotify web player