On Thu, 29 Aug 1996, James Graham wrote:

> 
> At this point, we could have stably released 1.2 and been well on our way
> to putting 1.3 together; instead we're still sitting here running 1.2B which
> isn't likely to change by the time the next -current gets underway.
> 
> Why don't we just punt at this point, call it 1.2 and unlock the tree so we
> can get going?  It's certainly stable enough to do so.

I'd like to second that notion, though I do appreciate thatit's a
volinteer effort. :-)


> Curiously, why is there an "fsck_ffs"?  Do we plan on fsck'ing any other
> kinds of filesystems?  Do we currently support "fsck_msdos"?  "fsck_xdos"?
> "fsck_ntfs"?  For obvious reasons we don't "fsck_nfs" or "fsck_null"...
> what's that leave us?

Ooo! Oooo! Mr. Cadder! Mr. Cadder!

I'm reading the red book (D&I 4.4BSD), and I'm just finishing the LFS
section. If it worked, it sounds like LFS is pretty cool (I've heard
rumblings about its completeness in the 1st 4.4 release). I know our RS/6k
at work, which uses the Journalized File System, IBM's version of LFS, is
fairly robust about dealing w/ power loss, reset button presses, and other
such drastic proceedings. Like the time the janitor accidentally unplugged
the computer, then plugged it back in. It came up fine.

When using LFS, you'd want an fsck-esque program for boot consistancy
checks. :-)

Take care,

Bill