On Wed, 20 Sep 2017 13:02:22 +0200 Dave Plater <dplater.list@gmail.com> wrote:
On 20/09/2017 12:36, Carlos E. R. wrote:
On 2017-09-20 08:59, Dave Plater wrote:
On 20/09/2017 08:50, Per Jessen wrote:
Dave Plater wrote:
I think all that is happening is openSUSE is dropping support for reiserfs (Which I fondly remember running SUSE 6.4 on recycled hard drives, even one that was clunking, using reiserfs) but I don't think that it will be maliciously removed from the kernel.
Agree completely. There is simply no reason.
That would prevent yast from converting the partitions in the first place. If you can't read a partition you can't convert it.
This conversion idea seems a pretty odd and risky thing to attempt - fiddling people's data, needing extra space, backup etc.
It sounds like a lot of work for the yast/installation team as well, as you said it's risky, I don't know the structure of the reiser file system but it would need to be converted into a similar structured file system or it would require a lot of spare space to backup the system and then copy over to a new partition. This would be a show stopper for users without enough space if a reliable partition converter hasn't been created. Converting live, no, I mean, on site, a reiserfs partition to... what?
Small files it stores where a normal filesystem stores the file name, without reserving data sectors to it. If I recall correctly, as the data sector block is not determined, it might store two files to
Only by copying the files elsewhere. Creating a small filesystem then grow it.
[1] Wikipedia explains it better. «ReiserFS stores file metadata ("stat items"), directory entries ("directory items"), inode block lists ("indirect items"), and tails of files ("direct items") in a single, combined B+ tree keyed by a universal object ID. Disk blocks allocated to nodes of the
«By contrast, ext2 and other Berkeley FFS-like file systems of that time simply used a fixed formula for computing inode locations, hence
Reiserfs structure is a nightmare to understand and code, IMHO because it is brilliant. Write a migration tool in-house? Wow. the same sector.[1] How can you code conversion of that? tree are "formatted internal blocks". Blocks for leaf nodes (in which items are packed end-to-end) are "formatted leaf blocks". All other blocks are "unformatted blocks" containing file contents. Directory items with too many entries or indirect items which are too long to fit into a node spill over into the right leaf neighbour. Block allocation is tracked by free space bitmaps in fixed locations.» limiting the number of files they may contain.[21] Most such file systems also store directories as simple lists of entries, which makes directory lookups and updates linear time operations and degrades performance on very large directories. The single B+ tree design in ReiserFS avoids both of these problems due to better scalability properties.»
https://en.wikipedia.org/wiki/ReiserFS Everything is doable but the complexity of the task would increase the possibility of bugs. Anyway it seems that fstransform claims to convert between all the major filesystems. I dropped reiser for xfs way back when it's demise was first announced. My computer has to permanently run on a ups connected to a large battery because my mains supply never exceeds 210v (230v supply) and lives at 170v during peak demand sometimes drops below 120V. I've a 60AH deep cycle battery on the ups and a voltage tolerant power supply to keep it topped up. So my prime requirement for a file system is power failure tolerance and xfs seems to be the best for that. Saying that ext4 appears to be ok lately. The only problem I've had for ages is a corrupted file on a usb drive with an ntfs partition. I had to mount it in an ms VM to fix it. Dave P
Dave, please would you fix your mailer so it does quoting in a readable way? Try hitting return before typing your contribution or something. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org