[Bug 199407] mkuzip(8) verbosity change request to help with makefs(8) filesystems
bugzilla-noreply at freebsd.org
bugzilla-noreply at freebsd.org
Mon Apr 13 01:43:06 UTC 2015
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=199407
Bug ID: 199407
Summary: mkuzip(8) verbosity change request to help with
makefs(8) filesystems
Product: Base System
Version: 11.0-CURRENT
Hardware: Any
OS: Any
Status: New
Severity: Affects Only Me
Priority: ---
Component: bin
Assignee: freebsd-bugs at FreeBSD.org
Reporter: kwhite at site.uottawa.ca
Created attachment 155531
--> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=155531&action=edit
requested enhancements to mkuzip and makefs
After an Easter weekend's head scratching over a frustratingly
intermittent problem...
mkuzip's default verbosity level hides a message that might help
end users (me) when disk /dev/ufs/... labels on mkuzip filesystems
"mysteriously" don't appear at boot. mkuzip provides useful messages
when "-v" is added, but they're lost amongst the noise. I propose
adding verbosity levels: "-v" "-vv"
Patch for mkuzip(8) attached. The mkulzma(8) derivative could be
modified similarly.
sys/geom/label/g_label_ufs.c expects well behaved filesystem sizes.
Unfortunately, mkuzip(8) may need to add some padding when compressing
a filesystem, and the provider size will then no longer match the
superblock. Particularly for a makefs(8) filesystem.
One workaround for "file size is not multiple of XXXX, padding data"
would be for the user to specify a bsize the same as the intended
mkuzip blocksize. e.g.:
$ ZBLOCKSIZE=32768
$ FSIZE=`expr $ZBLOCKSIZE / 8`
$ LABEL="ALabel"
$ makefs -o label="$LABEL" -o bsize=$ZBLOCKSIZE -o fsize=$FSIZE ...
$ mkuzip -v -s $ZBLOCKSIZE ...
A patch (attached) for makefs(8) that pads the resulting filesystem
to fit on a mkuzip cluster_size boundary, might be useful as well.
$ ZBLOCKSIZE=32768
$ LABEL="ALabel"
$ makefs -o label="$LABEL" -c $ZBLOCKSIZE ...
$ mkuzip -v -s $ZBLOCKSIZE
...keith
~
--
You are receiving this mail because:
You are the assignee for the bug.
More information about the freebsd-bugs
mailing list