Add FUSE support directly to the API. Instead of needing to use the
external 'guestmount' command, you can mount the libguestfs filesystem
space on a local mountpoint using an API call from any language.
Note that although mount-local-run is marked as Cancellable, the
current implementation does not support it, but it would be relatively
simple to add it.
fuse_main can return any non-zero value on error. In particular on
certain sorts of error such as unsupported -o options it returns 1
(not -1).
Therefore make sure any non-zero return is turned into EXIT_FAILURE.
(Thanks Pádraig Brady)
Remove the hack that let you run ./fish/guestfish or
./fuse/guestmount. You now have to do:
./run ./fish/guestfish
or
./run ./fuse/guestmount
to run these programs without installing.
* df/df.c: As above.
* df/main.c: As above.
* df/output.c: As above.
* fuse/guestmount.c: As above.
* inspector/virt-inspector.c: As above.
* rescue/virt-rescue.c: As above.
The documentation for the getxattr and listxattr calls is not very
clear and as a result we were always returning something different
from that which the Linux kernel would usually return.
This fixes these calls, at least far enough that both the 'getfattr'
and 'getfacl' programs now work fine on FUSE-mounted filesystems.
Note that SELinux attrs are *not* passed through. This appears to be
a known bug between SELinux and FUSE. For more information see:
http://www.spinics.net/lists/selinux/msg09460.html
The other programs have the variable, but the flag is not enabled
either because it doesn't make sense or because the implications are
not well understood.
This adds the guestfish --rw option, intended in future
to be required for writing to disk images.
At the moment this does not change the default and so does
nothing. This patch is intended for backporting to the
stable branches so that we can start to introduce scripts
which use 'guestfish --rw'.
This feature is also available in guestmount because of the
shared option parsing code.
You don't need to do anything to enable it, just using -i
will attempt decryption of encrypted partitions.
Only works for simple Fedora whole-disk encryption. It's a
work-in-progress to make it work for other types of encryption.
In guestfish, factor out the processing of the options -a, -c,
-d, -i, -m, -n, -r, -v, -V, -x into a separate set of files:
options.c, options.h, inspect.c, virt.c.
Change guestmount so that it uses these same files (from the
../fish directory) to process the same options.
This unifies the handling of these options between the two programs.
It also adds the useful inspection feature to guestmount, so you
can now do:
guestmount -d Guest -i --ro mnt/
If this string was non-empty, then it broke a lot of things because
autoconf and other parts of the build system were expecting this
string to contain a simple MAJOR.MINOR.RELEASE version number.
This requires changes to guestfish and guestmount so they use the
guestfs_version API to fetch the version from the library. (The
Perl tools were already doing it this way). In a way this is more
accurate, because it's no longer hard-coded in the binary, but
fetched from the dynamically linked libguestfs.so.
guestmount.c: In function 'main':
guestmount.c:899: error: implicit declaration of function 'setlocale'
[-Wimplicit-function-declaration]
guestmount.c wasn't including locale.h.
guestfs_mount adds -o sync implicitly. This causes a very large
performance problem for write-intensive programs (eg. virt-v2v).
Document this as a "gotcha".
Change the tests, guestfish, Sys::Guestfs::Lib, guestmount to use
mount-options instead.
(Note that this gotcha does not affect mount-ro).
The source of the performance problem was first identified by
Matthew Booth.
This commit adds the calls to setlocale &c to all of the current
C programs.
It also adds l10n support to hivexget and hivexml which lacked them
previously.
To test this, try:
LANG=pa_IN.UTF-8 guestfish --cmd-help
(You can only do this test after installing the package, or at
least the 'pa.mo' mo-file in the correct place).
(NB: The API / ABI doesn't actually change here - it's just made much
simpler to use).
The API for RBufferOut functions was unexpectedly hard to use in the
case where a zero-length buffer might be returned. For discussion on
this see:
https://www.redhat.com/archives/libguestfs/2009-November/thread.html#00115
This commit ensures that in the zero-length buffer case, the return
value is never NULL. Thus code is now able to just check if the return
value == NULL to indicate an error, which is simpler for all concerned.
The implementation of this is, however, more complex because we have
to be careful about this case inside both the daemon and the library
code, which is what this commit does.
This has passed a full round of tests.
This implements FUSE filesystem support so that any libguestfs-
accessible disk image can be mounted as a local filesystem.
Note: file writes (ie. write(2) system call) is not yet implemented.
The API needs more test coverage, particularly lesser-used system
calls.
The big unresolved issue is UID/GID mapping between guest filesystem
IDs and the host. It's not easy to automate this because you need
extra details about the guest itself in order to get to its
UID->username map (eg. /etc/passwd from the guest).
