Your message dated Mon, 14 May 2007 18:55:12 -0400
with message-id <[EMAIL PROTECTED]>
and subject line Bug#423863: e2fsprogs: Option "-L" without option argument
gives bogus "invalid blocks count" error
has caused the attached Bug report to be marked as done.
This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.
(NB: If you are a system administrator and have no idea what I am
talking about this indicates a serious mail system misconfiguration
somewhere. Please contact me immediately.)
Debian bug tracking system administrator
(administrator, Debian Bugs database)
--- Begin Message ---
Package: e2fsprogs
Version: 1.39+1.40-WIP-2007.04.07+dfsg-2
Severity: normal
I have made the admittedly stupid mistake to omit
the argument to the "-L" option for mke2fs, yet
instead of pointing out this simple fact it did
insist on an "invalid blocks count" for the device
and trying to tell it about the blocks count did
not actually help it.
This error message is bogus because instead of
telling about serious problems like "invalid blocks
count", mke2fs should either abort and report the
missing / empty argument to the "-L" option it got,
or issue a warning and just continue, leaving the
volume label string empty.
Giving a proper argument to the "-L" option did
of course fix it, but people who see messages
about block counts will probably investigate
and not have a clue about the argument they
dared to omit. Not unless it strikes them as
odd to repeatedly request a "-L"abel but no
actual identifier string on the plate.
Oh well - proof that the punishment is always
worse than the crime, so please have mercy and
be more forgiving with tired e2fs creators who
may not be able to recognize their stupidity
in the early morning... ;-)
-- System Information:
Debian Release: lenny/sid
APT prefers unstable
APT policy: (500, 'unstable'), (500, 'testing'), (500, 'stable')
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.20 (SMP w/2 CPU cores)
Locale: [EMAIL PROTECTED], [EMAIL PROTECTED] (charmap=ISO-8859-15) (ignored:
LC_ALL set to [EMAIL PROTECTED])
Shell: /bin/sh linked to /bin/bash
Versions of packages e2fsprogs depends on:
ii e2fslibs 1.39+1.40-WIP-2007.04.07+dfsg-2 ext2 filesystem libraries
ii libblkid 1.39+1.40-WIP-2007.04.07+dfsg-2 block device id library
ii libc6 2.5-7 GNU C Library: Shared libraries
ii libcomer 1.39+1.40-WIP-2007.04.07+dfsg-2 common error description library
ii libss2 1.39+1.40-WIP-2007.04.07+dfsg-2 command-line interface parsing lib
ii libuuid1 1.39+1.40-WIP-2007.04.07+dfsg-2 universally unique id library
e2fsprogs recommends no packages.
-- no debconf information
--- End Message ---
--- Begin Message ---
tag 423863 +wontfix
thanks
On Mon, May 14, 2007 at 05:22:50PM +0200, Dirk Ritter wrote:
> Package: e2fsprogs
> Version: 1.39+1.40-WIP-2007.04.07+dfsg-2
> Severity: normal
>
> I have made the admittedly stupid mistake to omit
> the argument to the "-L" option for mke2fs, yet
> instead of pointing out this simple fact it did
> insist on an "invalid blocks count" for the device
> and trying to tell it about the blocks count did
> not actually help it.
You didn't tell me exactly what option string you gave, but I'm pretty
sure there's not much I'm going to be able to do for you.
The problem is that artificial intelligence is in its infancy, and
mind reading interfaces are even more so. So mke2fs can't necessary
figure out that -L was missing an option. Indeed, when I was first
trying to figure out what you did, I tried some things and didn't get
the invalid blocks count error message:
mke2fs -L /dev/foo.img
Gives a usage message, since /dev/foo.img is assumed to be the
requested label (to a computer, that's a perfectly fine label name,
and if you're expecting decent artificial intelligence, prepare to be
disappointed), and so there is a missing device name --- hence the
usage message.
mke2fs /tmp/foo.img -L
Gives an error message, "mke2fs: option requires an argument -- L",
since the -L is at the very end so, getopt() tell that -L was missing
an argument. HOWEVER, you can specify options after the fixed
argument is a GNU getopt() extension, and won't necessarily work on
other Legacy Unix systems, such as Solaris. So I generally don't
advise that users or system administrators get comfortable with this
kind of option ordering, since it violates classical Unix conventions,
and won't work anywhere.
I finally hit upon what you did, which was probably some variation of:
mke2fs -L -b 4096 /tmp/foo.img
So what happened is that mke2fs assumed that you wanted a label of
"-b" (which is a valid label name), with the pathname of the
filesystem to be created to be "4096", which is also potentially quite
valid, and a block count of /tmp/foo.img, which unfortunately, isn't a
valid block count. So the error message given by mke2fs is perfectly
valid. I'm not really going to fix this, given there's not much I can
catch all possible variations. For example:
mke2fs -Lcb 4096 /tmp/foo.img
Maybe the user really wanted to use a filesystem label of "cb"; maybe
the filesystem was intended to "Charlie Brown" videos.
Besides, the last time someone tried to add artificial intelligence to
a program, what we ended up with a REALLY ANNOYING dancing paperclip
that would often completely wrong advice. I refuse to add that kind
of bloat to mke2fs.
Although maybe it would make for a good April Fool's joke, except that
someone has already implemented "vile" (i.e., adding a paperclip
advisor to the "vi" editor).
- Ted
P.S. And if we do invent real artificial intelligence, the danger
would be that our computers might start issuing responses such as,
"You're too stupid/tired to run mke2fs; I'm changing the root password
and locking you out of the system until you get eight hours of solid
sleep. Go home, young man!" :-)
--- End Message ---
