On Mon 24 Mar 2003 09:34, Duncan posted as excerpted below:
> On Mon 24 Mar 2003 06:56, Pixel posted as excerpted below:
> > Duncan <[EMAIL PROTECTED]> writes:
> > > ..  I was pretty disgusted with DiskDrake as well, after I'd put a
> > > bunch of comments in detialing the various mount options, ran
> > > DiskDrake, and not only had it delete or mix up the comments and order,
> > > but actually uncomment a bunch of stuff, so that when I rebooted, I had
> > > a bunch of root mount points like /this /option /is /pretty /useful,
> > > etc!
> >
> > please send the non-screwed fstab
>
> That was an entiremajor Mdk release ago now (8.1).  I'm hoping something
> like that isn't still around, altho like I said I haven't tried it lately.
>
> I'll try it again, tho, and post if a similar problem still exists..  Thx.

OK.  Tried it.  It isn't uncommenting stuff and creating weird root 
mountpoints with them, any more.  However, it still reorganizes things, 
removing comments I've put in there.

The original fstab is attached as fstab.jed.  (JED are my initials.  I often 
use that for a backup, when I'm afraid something might overwrite a .bak 
backup.)  The comments should be self explanatory.  Note that I use the long 
/dev/ide/host... paths as an urpmi upgrade at one point left devfs 
unworkable, and all my /dev/hdX partitions unmounted!  Using the longer 
native paths might not look so neat, but it works better in the event devfs 
doesn't.

I then used diskdrake to create a new partition out of some free space on 
/dev/hdb, made it ReiserFS, and set the mount point as /test, for this test.  
The resulting fstab is attached as fstab.munched.  Not only are all those 
notes about options gone, but so is my nice ordering by function.  Everything 
is tossed about pretty much randomly, or so it looks to me, with the 
categorizing comments still there, but no longer applicable to what's under 
them.  Furthermore, all my devfs-failure-safe paths have disappeared, to be 
replaced by totally nonfunctional paths, should devfs fail to load, for 
whatever reason.

What I EXPECT to happen, is that diskdrake ONLY makes changes to entries I've 
made changes to, in this case, appending a single new entry at the end of the 
file, leaving everything else as it was.

Should I file a bug report on this. or is this expectation to fancy, and it 
would just get marked WONTFIX or similar (or will this posting get it looked 
into)?

While I'm at it, last time I checked, XFdrake did something similar..  I run 
three monitors off of two video cards, so my XF86Config file is 
understandably a bit to complex for most automated tools to parse very well.  
However, making it so it didn't entirely foobar things if I did run it, would 
be nice.  Is something like that worth pursuing, or is that again a WONTFIX?

In both cases, saving the existing file to a backup of some sort, with a 
screen telling folks exactly what command to use to reverse the damage, if 
something went wrong, would be quite useful, IMO.  (Perhaps it already does 
this, which is why I used the .jed extension for my manual backup, as I also 
keep a .bak copy, but I can't tell if that's my .bak copy or diskdrake's, and 
there were no instructions telling me about it and how to reverse the damage 
in the event something went wrong, in any case.)

-- 
Duncan
"They that can give up essential liberty to obtain a little
temporary safety, deserve neither liberty nor safety." --
Benjamin Franklin
# fstab file.
##############################################################################
# Common MntOpts
# (no- prefix negates atime, auto, dev, exec, suid, user)
# (a)sync                                               (async default)
# atime         update access times                     (default)
# auto          mount at boot and with -a option
# defaults      auto, async, dev, exec, rw, suid, nouser
# dev           interpret char/block special devs on fs
# exec          permit execution of binaries
# _netdev       net-access req'd to mount
# ro/rw         read-only/read-writeable                (rw default)
# suid          allow suid/sgid bits to take effect     (default)
# user(s)       user may mount, impl. no- exec suid dev (default nouser)
#               (unless overridden) -s allows any unmount, else only mounting user
##############################################################################
# Dev/Part, MntPnt, Type, MntOpt, Dump, FSCK
##############################################################################

# /, /boot /swap
/dev/ide/host0/bus1/target0/lun0/part5 / reiserfs noatime 1 1
/dev/ide/host0/bus1/target0/lun0/part1 /boot reiserfs noatime 1 2
/dev/ide/host0/bus1/target0/lun0/part7 swap swap defaults 0 0

# removable
/dev/fd0 /mnt/floppy auto user,sync,exec,noauto 0 0
/dev/cdroms/cdrom0 /cd auto user,ro,noauto 0 0
/dev/cdroms/cdrom0 /cdx auto user,ro,exec,noauto 0 0
/dev/scsi/host0/bus0/target0/lun0/cd /cd auto ro,noauto,user,exec 0 0

# special
none /dev/shm tmpfs defaults 0 0
none /proc proc defaults 0 0

# other Linux
/dev/ide/host0/bus1/target0/lun0/part11 /var reiserfs noatime 1 2
/dev/ide/host0/bus1/target0/lun0/part10 /tmp reiserfs noatime,nosuid,nodev 1 2
/dev/ide/host0/bus1/target0/lun0/part9 /home reiserfs noatime 1 2
/dev/ide/host0/bus1/target0/lun0/part8 /usr reiserfs noatime 1 2
/dev/ide/host0/bus1/target0/lun0/part13 /mail reiserfs noatime 1 2
/dev/ide/host0/bus1/target0/lun0/part12 /news reiserfs noatime 1 2
/dev/ide/host0/bus1/target0/lun0/part14 /mm reiserfs noatime 1 2
/dev/ide/host0/bus1/target0/lun0/part6 /opt reiserfs noatime 1 2

# Windows
/dev/ide/host0/bus0/target0/lun0/part8 /win/arch vfat uid=503,gid=505,umask=007 0 0
/dev/ide/host0/bus0/target1/lun0/part1 /win/jed vfat uid=503,gid=505,umask=007 0 0
/dev/ide/host0/bus0/target0/lun0/part1 /win/main vfat uid=503,gid=505,umask=007 0 0
/dev/ide/host0/bus0/target0/lun0/part7 /win/media vfat uid=503,gid=505,umask=007 0 0
/dev/ide/host0/bus0/target0/lun0/part6 /win/news vfat uid=503,gid=505,umask=007 0 0
/dev/ide/host0/bus0/target0/lun0/part5 /win/temp vfat uid=503,gid=505,umask=007 0 0
# /, /boot /swap
/dev/hdc5 / reiserfs noatime 1 1
/dev/hdc1 /boot reiserfs noatime 1 2
/dev/scd0 /cd auto user,ro,noauto 0 0
none /dev/pts devpts mode=0620 0 0
# special
none /dev/shm tmpfs defaults 0 0
/dev/hdc9 /home reiserfs noatime 1 2
/dev/hdc13 /mail reiserfs noatime 1 2
/dev/hdc14 /mm reiserfs noatime 1 2
# removable
/dev/fd0 /mnt/floppy auto user,sync,exec,noauto 0 0
/dev/hdc12 /news reiserfs noatime 1 2
/dev/hdc6 /opt reiserfs noatime 1 2
none /proc proc defaults 0 0
/dev/hdb5 /test reiserfs defaults 1 2
/dev/hdc10 /tmp reiserfs noatime,nosuid,nodev 1 2
/dev/hdc8 /usr reiserfs noatime 1 2
# other Linux
/dev/hdc11 /var reiserfs noatime 1 2
# Windows
/dev/hda8 /win/arch vfat uid=503,gid=505,umask=007 0 0
/dev/hdb1 /win/jed vfat uid=503,gid=505,umask=007 0 0
/dev/hda1 /win/main vfat uid=503,gid=505,umask=007 0 0
/dev/hda7 /win/media vfat uid=503,gid=505,umask=007 0 0
/dev/hda6 /win/news vfat uid=503,gid=505,umask=007 0 0
/dev/hda5 /win/temp vfat uid=503,gid=505,umask=007 0 0
/dev/hdc7 swap swap defaults 0 0

Reply via email to