mdmfs(8) lacks the ability to populate throwaway memory filesystems from an
existing directory.
This features permits an interesting setup where /var for instance lives on
a device where wear-leveling is something you want to avoid as much as
possible and nonetheless you don't want to lose your logs, ports metadata,
etc. Here are the steps:
1. Copy /var to /var.bak;
2. Mount an mfs into /var using -k /var.bak at startup;
3. Synchronize /var to /var.bak weekly and on shutdown.
Note that this more or less mimics OpenBSD's mount_mfs(8) -P flag.
PR: 146254
Submitted by: jlh (many moons ago)
MFC after: 1 week
Mentioned in mdconfig(8), malloc-backed md(4) can be unstable unless
required memory is allocated up front with -o reserve. Furthermore, panics
have been observed with md used in fstab on 12.0-RELEASE. Choose the stable
route and pass -o reserve.
Submitted by: Paul Vixie
MFC after: 1 week
This code allocated a correctly sized buffer, read past the end of the
source buffer, writing off the end of the target buffer, and then writing
a '\0' terminator past the end of the target buffer (in the wrong place).
It then leaked the buffer.
Switch to a statically sized buffer on the stack and update the source
pointer and
length before use so the correct things are copied.
Fix a logic error in the checks that the format of the line is as
expected and move on out of an assert.
Remove an unneeded close(). fclose() closes the descriptor.
Found with: CheriABI
Obtained from: CheriBSD
Reviewed by: kib, jhb, markj
Differential Revision: https://reviews.freebsd.org/D19122
Exiting with a signal should not be treated the same as successful exit with
zero status.
Return signal exit information to the callers via negative integers, to
enable distinction from normal exit statuses. (All consumers that check for
errors don't care what the exact non-zero exit value is -- in such a case
they print a diagnostic message and either continue or bail.)
Additionally, check for unexpected sources of waitpid() wakeup and bail if
we encounter them.
Reported by: lev@
Reviewed by: kib, lev, markj (earlier version)
Sponsored by: Dell EMC Isilon
Differential Revision: https://reviews.freebsd.org/D17035
Mainly focus on files that use BSD 2-Clause license, however the tool I
was using misidentified many licenses so this was mostly a manual - error
prone - task.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
No functional change intended.
Existing scripts and associated config such as rc.initdiskless, rc.d/var,
and others, use mdmfs to create memory filesystems. That program accepts a
size argument which allows SI suffixes and treats an unsuffixed number as a
count of 512 byte sectors. That makes it difficult to convert existing
scripts to use tmpfs instead of mdmfs, because tmpfs treats unsuffixed
numbers as a count of bytes. The script logic to deal with existing user
config that might include suffixed and unsuffixed numbers is... unpleasant.
Also, there is no g'tee that tmpfs will be available. It is sometimes
configured out of small-resource embedded systems to save memory and flash
storage space.
These changes enhance mdmfs(8) so that it accepts two new values for the
'md-device' arg: 'tmpfs' and 'auto'. With tmpfs, the program always uses
tmpfs(5) (and fails if it's not available). With 'auto' the program prefers
tmpfs, but falls back to using md(4) if tmpfs isn't available. It also
handles the -s <size> argument so that the mdconfig interpetation of
unsuffixed numbers applies when tmpfs is used as well, so that existing user
config keeps working after a switch to tmpfs.
A new rc setting, mfs_type, is added to etc/defaults/rc.conf to let users
force the use of tmpfs or md; the default value is "auto".
Differential Revision: https://reviews.freebsd.org/D12301
specifying rotational-positions, reflect that in the command arguments.
PR: bin/110178
Submitted by: Alex Kozlov <spam at rm-rf dot kiev dot ua>
Approved by: imp (mentor)
Rationale:
We are interested in the current (last) element of the argv array
there, not in its first element. The if construct is there because
we want to avoid adding empty (zero-length) arguments to argv, so
we just don't advance argvp if the current argument is empty, and
it gets overwritten at the next iteration. Note that strsep(3)
doesn't treat consecutive delim characters as a single separator,
it returns empty fields between such characters, and it's up to the
caller to handle them this or that way.
Also add a comment that the argv array ends up null-terminated in
any case (it's due to the design of the for loop) as an answer to
a possible question why the whole argv isn't zero-filled.
Submitted by: yongari
Tested by: yongari
MFC after: 3 days
unit w/o suffix is specified. It had better be an empty
string as it will be passed to a printf-like function that
builds the command line to run.
PR: kern/109863
MFC after: 1 week
compatible, it would have to (at least):
- support the "compat-compat" -T option,
- *not* support the -l, -O, and -v options,
- default to soft updates being disabled.
Worse, the compatibility mode makes it impossible to mount_mfs(8)
a file system from fstab(5) with soft updates disabled (-S). [1]
Now, the only difference when called as "mount_mfs" or "mfs" (as
opposed to "mdmfs") is that the file mode of the mount point is
set by default to 01777. All options available to mdmfs(8) are
also available to mount_mfs(8); the -C option is still recognized
but ignored for backward compatibility.
PR: bin/98860 [1]
MFC after: 2 weeks
disk. Apparently some people want to use mdmfs as mount_* as a
shortcut for mounting existing file-based file systems.
Note that unlike in the patches from the submitters, this option is
not available in compat mode. Compat mode was supposed to support only
things that mount_mfs used to support. To use this option from fstab,
mdmfs should be called mount_md, not mount_mfs. This distinction has
not always upkept for new options, and those can't be fixed now
without breaking people's systems, but new options should not usually
be allowed in compat mode. (Not sure why -F is allowed there at all.)
PR: 57641
Submitted by: Ruben de Groot
Submitted independently by: Wojciech A. Koszek, for Urzad Miasta Czestochowa
autonumbering is requested, set the unit to 0 instead of -1. This
option is just for output, and "/dev/md-1" looks disconcerting.
Submitted by: Wojciech A. Koszek
<group> is a numeric user/group ID instead of a user/group name (as
explicitly intended to be allowed by both the manual page and the
implementation).
Before this fix, mdmfs(8) aborted:
| # mdmfs -s 32m -w 0:0 md /var/tmp/foo
| Assertion failed: (mip->mi_have_uid), function extract_ugid, file /usr/src/sbin/mdmfs/mdmfs.c, line 555.
| Abort trap (core dumped)
The "mi_have_[ug]id" fields were only set in case a name lookup was
successful. Instead they also have to be set in case the string to
integer conversion was successful.
Additionally, as a result of this fix, two assertions at the end of
the function are now always true and hence can be just be removed. It
is guarrantied that both the UID and the GID are set when the function
returns regularily, else it would have been already bailed out with
usage()/exit(3) or errx(3) before.
Spotted by: Christoph Schug <chris@schug.net>
MFC after: 3 days
created with the multilabel flag from inception. This simply
passes the "-l" flag on to newfs(8).
Obtained from: TrustedBSD Project
Sponsored by: DARPA, McAfee Research
return for getopt() and comparing to -1, ditto with fgetc() and EOF,
and using the kg_nice value from <sys/user.h>
Submitted by: Stefan Farfeleder <stefan@fafoe.narf.at>
Reviewed by: obrien, bde (a while back)
Tested lightly on: ppc, i386, make universe
of an argument name collision with -O, use -v, and default to whatever
the newfs default is for the platform (generally, UFS1). This is
required to support diskless workstations that use UFS2 for their
mdmfs file systems.
Reviewed by: dd, bmah
Approved by: re (bmah)
Obtained from: TrustedBSD Project
Sponsored by: DARPA, Network Associates Laboratories
name is "mount_mfs" or "mfs". Previously, the condition was that
the program name must start with "mount_", but this both missed
the case where mount(8) invokes mdmfs with argv[0] = "mfs", and it
included cases such as "mount_md" where compatibility is not
required.
Reviewed by: dd
default if the executable is named (called as) "mount_*", or can be
enabled with the -C option. This allows users to leave their old
fstab entires unchanged (modulo symlink'ing mdmfs to mount(md|mfs))
and have things behave the way they should (by emulating mount_mfs
silliness), while still allowing mdmfs to be used as a generic
make-an-md-and-mount-it type thing.
Right now, the only effects of this option is to set the mount-point
mode to 01777 as if "-p 1777" was given, and to complain about getting
command-line options that mount_mfs didn't take (e.g., -X, -L, et al).
The latter is mostly to try to catch operator errors.
Also implement -U, which turns on soft-updates. It's redundant (since
softdep is the default), but implement it anyway for compatibility.
