Re: Reiserfs stability??
Eduard Bloch said:
> Who cares? It is Yet Another Hey-People-I-Show-U-How-To-Use-EXT3-FAQ, one
> of dozens available in the net.
I care. it is the first thing that comes up when I do a search for
"ext3 linux 2.2". it has the look of the "official" ext3 faq, infact
none of the other links on the first set of hits for ext3 on google
seem to show anything other then diffs or mailing list archives. Reiserfs
has a central archive where I can go to get patches info, contact
> *scratch*, *scratch* Why didn't you consider that Ext3 for 2.2.x has not
> needed upgrades for a year since there are no important bugs? On the other
> hand, ugly things in ReiserFS code come to daylight again and again, so
> upgrades are important.
Because it is new, despite it being based on ext2 code. reiserfs
has it's share of problems too probably. I found a pretty serious,
but obscure bug in reiserfs about a year ago, spent a day tracking
it down, submitted it to hans reiser, within 24 hours I had an update
that fixed the problem. And note this was an obscure problem that
occured ONLY on 3ware raid 5 arrays(not raid1, not raid10, and not
raid5 on any other systems that I have). If I were to have such
a problem on ext3, I would be screwed.
Also I take into account future kernel upgrades, if a critical bug
is discovered in 2.2.19(my main kernel) and I was forced to upgrade
to 2.2.20(or later), at which point ext3 may break(fail to patch,
compile, or have a bug that is tickled by the new kernel) again I
am screwed. The possibility may be small, but I much prefer to use
code that has a maintainer(even if its not activly developed) over
code that has no apparent maintainer).
If you can point me to information that contridicts the FAQ I pointed
out which says EXT3 under 2.2 is under development I'd be happy
to re-evaluate the situation for my systems.
till then, reiserfs is my one and only journalling filesystem choice