On Wed, 2003-04-02 at 13:05, joe wrote: > Shawn wrote: > > >Suse may have "pulled it off", but with the occassional data corruption > >here and there. > > > Hmm, problem with that theory is that > suse is pretty solid - you've got stuffy > european banks depending on it, etc - > > How much data corruption do you think > that bank would stand before they get > rid of the offending OS? Actually, Hans Reiser stated in Febuary that there are a few issues left in the current release they are working on, including one related to unlink. He also said he "wished he could say they were bug free". There were also several recent reports on the mailing list of corruption when using write caching, which seems to be a bane of most journaling filesystems apparently. The mailing list for the past few weeks seems to be filled with commentary and discussions about possible data corruption. Admittedly, to me some of the more "unstable" aspects that people are seeing with reiser may be due to using Kernel 2.5 in many instances, or an untested combination of some random Kernel version + ReiserFS patches. I personally look for many of the remaining issues to be shaken out when the 2.5 codebase becomes the mainline stable kernel. I hope ReiserFS becomes as "bug free" as Hans Reiser desires, alternatives are always nice. Does this mean I think ext3, etc. are bug free? No. I'm not that naive. But for me, ext3 is the way to go for now, I can' tell you how many times the ability to mount an ext3 partition as ext2 has saved my bacon so to speak or been very convenient... Anyway nothing here, move along, just my insigificant 2 cents, -- Shawn <drevil@xxxxxxxxxxxx> http://www.warpcore.org/
