(Not entirely sure whether using Gnulib to replace standard functions
is a good idea at all.)
link with libgnu:
CCLD libguestfs-test-tool
libguestfs_test_tool-test-tool.o: In function `main':
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:103: undefined reference to `rpl_getopt_long'
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:113: undefined reference to `rpl_optarg'
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:125: undefined reference to `rpl_optarg'
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:126: undefined reference to `rpl_optarg'
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:109: undefined reference to `rpl_optarg'
libguestfs_test_tool-test-tool.o: In function `set_qemu':
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:382: undefined reference to `rpl_perror'
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:366: undefined reference to `rpl_perror'
libguestfs_test_tool-test-tool.o: In function `make_files':
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:416: undefined reference to `rpl_perror'
/home/bengen/src/deb/pkg-libvirt/libguestfs/debian/build-default/test-tool/../../../test-tool/test-tool.c:428: undefined reference to `rpl_perror'
CC guestfsd-9p.o
In file included from ../../../daemon/9p.c:32:0:
../../../daemon/daemon.h:33:34: fatal error: guestfs-internal-all.h: No such file or directory
This involves adding a new test API which crashes the appliance in the
middle of a simulated upload, then a test which uses that API to test
for the libguestfs (library-side) crash.
Instead of the segfault you now get a more informative error:
*stdin*:0: libguestfs: error: connection to daemon was closed unexpectedly.
This usually means the libguestfs appliance crashed. Please enable
debugging (LIBGUESTFS_DEBUG=1) and rerun the command, then look at the
debug messages output prior to this error.
libguestfs: error: /dev/stdout: write: Broken pipe
libguestfs: error: file receive cancelled by daemon
It should be possible to add the guestfs___free_string_list to
dllmlguestfs.so, but I cannot work out exactly how to do this. As a
result we end up using src/utils.c directly.
The libutils convenience library is a place for code shared between
the main library, language bindings and virt tools. Note that the
code is statically linked into both the library, each binding and each
tool, but this is an improvement because (a) the source is shared and
(b) libguestfs.so can export fewer private functions.
Currently it contains the cleanup functions, and the functions
guestfs___free_string_list function and guestfs___for_each_disk.
guestfs___for_each_disk has changed so that it no longer
unconditionally sets the error in the guestfs handle. Instead callers
can control error handling.
These are internal functions. Very old versions of libguestfs used to
export them, but they haven't been (and shouldn't be) exported for a
long time.
Also remove the unused guestfs_error function.
Earlier versions of automake complain if they get a configuration
parameter which they don't understand. The error is:
configure.ac:27: error: option 'serial-tests' not recognized
Use some m4 hackery to work around this.
It was failing to clean up and close the 'struct command' buffer,
because commit 94d90f03e2 added an
incomplete patch to use CLEANUP_CMD_CLOSE.
This fixes commit 94d90f03e2.
btrfs_subvolume_list can now take either the path of a mounted btrfs
filesystem, or a mountable describing the location of a btrfs
filesystem, or one of its volumes. In the latter case, the filesystem
will be automatically mounted outside of /sysroot before running the
btrfs tool, and unmounted afterwards.
There are now two forms of the 'debug progress' command:
(1) debug progress <n> (the original form) generates ordinary
rate-limited progress messages for <n> seconds.
(2) debug progress <n> <rate> generates progress messages every <rate>
microseconds for <n> seconds.
The second form omit the usual rate-limiting, and so wouldn't
be generated like this from an ordinary API call. However this
is useful for testing events (see RHBZ#909624).
Note that you have to use this as follows:
#if defined(__GNUC__) && GUESTFS_GCC_VERSION >= x0y0z /* gcc >= x.y.z */
since GUESTFS_GCC_VERSION is not defined unless it's GCC.
A Mountable is passed from the library to the daemon as a string. The daemon
stub parses it into a mountable_t, which it passes to the implementation.
Update all implementations which now take a mountable_t.
This change updates the api style of all apis which should take Mountable
descriptions rather than block devices. It also updates the documentation
accordingly, but doesn't implement any functional changes.
Because of evolution of the code, if the user used the --qemu or
--qemudir options, libguestfs-test-tool would segfault because
guestfs_set_qemu was being called before the guestfs handle was
opened.
Change the code so this doesn't happen, and also remove the global 'g'
variable to make the code a bit more robust.
Bug found by Amit Shah.
The second parameter passed into getline(3) is the size of the
allocated buffer, *NOT* the length of the returned line. This can be
confusing, so rename this parameter as 'allocsize' consistently
throughout the code.
This is just code motion.
This was disturbing to some users, but other users saw the warning
when it was really something they should pay attention to. Make it
clear that it's a warning, unless people are really using SELinux +
sVirt.
Thanks: Richard Harman.
For example:
{
CLEANUP_FREE_STAT struct guestfs_stat *stat = guestfs_stat (g, "/foo");
// stat is freed automatically when we leave the scope.
}
The new header file "guestfs-internal-frontend-cleanups.h"
includes definitions for these. It should not be included
directly.
Partition number was being passed to guestfs___check_for_filesystem_on
based on an index in list_partition. However, this ignores the
possibility of multiple block devices.
This change makes guestfs___check_for_filesystem_on examine the
passed-in device directly to determine if it is a whole device, or
what its partition number is.
The output of btrfs subvolume list has changed in F18 to include generation,
which breaks the parsing in btrfs_subvolume_list. This change replaces sscanf
with a more robust regular expression. The new regular expression should also
handle the addition of future unexpected columns.
The visibility field in action replaces in_fish, in_docs and internal.
The defined types are:
VPublic:
A public API. This is exported and documented in all language
bindings, and in guestfish.
VStateTest:
A public API which queries the library state machine. It is exported
and documented in all language bindings, but not guestfish.
VBindTest:
An internal API used only for testing language bindings. It is
guarded by GUESTFS_PRIVATE in the C api, but exported by all other
language bindings as it is required for testing. If language
bindings offer any way to guard use of these apis, that mechanism
should be used. It is not documented anywhere.
VDebug:
A debugging API. It is exported by all language bindings, and in
guestfish, but is not documented anywhere.
VInternal:
An internal-only API. It is guarded by GUESTFS_PRIVATE in the C api,
and not exported at all in any other language binding. It is not
documented anywhere.
Replace selected calls to 'perror (filename)' with:
fprintf (stderr, "syscall: %s: %m\n", filename);
so that more information is available about precisely which syscall
failed.
Note this is *not* reply_with_perror. These messages are only printed
in verbose output, for the benefit of debugging.
Previously I noticed that bfan used this command without any error:
><fs> hivex-open /WINDOWS/system32/config/software write:ture
^^^^^^^^^^
This was because the code allowed any string to be evaluated as a
boolean.
The new code is stricter. It allows the following strings only case
insensitive (everything else is an error):
1
true
t
yes
y
on
0
false
f
no
n
off
Not to be confused with the libxml2 macro 'BAD_CAST' which converts
from 'signed char *' to 'unsigned char *'.
The 'bad_cast' function was defined and used all over the place as a
replacement for a '(char *)' cast. I think it is better to make these
casts explicit, instead of hiding them in an obscure function.
This commit rearranges the internal header files.
"src/guestfs-internal.h" is just for the library, as before.
"src/guestfs-internal-frontend.h" is for use by all library, bindings,
tools C code, but NOT the daemon.
"src/guestfs-internal-all.h" is for use by all C code including the
daemon.
This is just code motion, but it has some important consequences:
(1) We can use the CLEANUP_* macros in bindings and tools code.
(2) We can get rid of TMP_TEMPLATE_ON_STACK.
(3) We will (in future) be able to stop bindings and tools code from
using the safe_* allocation functions (which are NOT safe to use
outside the library alone).
qemu-wrapper isn't regenerated if QEMUDIR is changed, so just
delete it and force regeneration.
Additionally we can drop the silent binary check, since
check-with-upstream-qemu-1 already does a similar test with --version
that will actually report an error to the user.
Certain functions are intended to be internal only, but we currently
export them anyway. This change moves them into a separate section of
guestfs.h protected by a GUESTFS_PRIVATE variable. This change also
enables private structs, but doesn't implement any.
This change only affects the C api. Language bindings aren't affected,
but probably should be in the future.
Without this change all the tests in the haskell bindings are rebuilt
every time. The primary motivation for this change is to fix this.
The fix for the above also allows parallel builds to be re-enabled.
Use the macro like this to create temporary variables which are
automatically cleaned up when the scope is exited:
{
CLEANUP_FREE char *foo = safe_strdup (bar);
...
// no need to call free (foo)!
}
The following code is also valid. The initialization of foo as 'NULL'
prevents any chance of free being called on an uninitialized pointer.
It may not be required in all cases.
{
CLEANUP_FREE char *foo = NULL;
...
foo = safe_malloc (100);
...
// no need to call free (foo)!
}
This is also valid:
{
CLEANUP_FREE char *foo = ..., *bar = ...;
...
// no need to call free (foo) or free (bar)!
}
The CLEANUP_FREE_STRING_LIST macro calls guestfs___free_string_list
on its argument. The argument may be NULL.
The CLEANUP_HASH_FREE macro calls hash_free on its argument. The
argument may be NULL.
Important implementation note:
------------------------------
On GCC and LLVM, this is implemented using __attribute__((cleanup(...))).
There is no known way to implement this macro on other C compilers, so
this construct will cause a resource leak.
Important note about close/fclose:
----------------------------------
We did NOT implement 'CLEANUP_CLOSE' or 'CLEANUP_FCLOSE' macros. The
reason is that I am not convinced that these can be used safely. It
would be OK to use these to collect file handles along failure paths,
but you would still want a regular call to 'close'/'fclose' since you
must test for errors, and so you end up having to do:
if (close (fd) == -1) {
// failure case
// avoid double-close in cleanup handler:
fd = -1;
...
}
// avoid double-close in cleanup handler:
fd = -1;
...
This is just code motion, but it allows us to read this flag inside
the 'construct_libvirt_xml_seclabel' function in future (as a possible
way to fix RHBZ#890291).
When debugging is enabled, this produces output like below. This is
useful when diagnosing what URI libguestfs is using.
libguestfs: opening libvirt handle: URI = NULL, auth = virConnectAuthPtrDefault, flags = 0
libguestfs: successfully opened libvirt handle: conn = 0xb05580
guestfs_last_errno (g) == 0 doesn't mean "no error". It means the
errno was not captured. In this case we have to substitute some sort
of errno, so choose EINVAL arbitrarily.
The reasons to do this are twofold:
(a) It's probably a tiny bit faster.
(b) It lets us capture the real errno if the link(2) syscall fails.
The errno is also passed through guestmount, fixing RHBZ#895905:
+ guestmount -a test1.img -m /dev/sda1:/ -m /dev/sda2:/boot /tmp/mnt
+ touch /tmp/mnt/foo
+ cd /tmp/mnt
+ ln foo boot/foo
ln: failed to create hard link ‘boot/foo’ => ‘foo’: Invalid cross-device link
cp will fail if /etc/lvm is an empty directory. Copy the entire
directory and adjust environment variable.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
RWMJ:
- Fixed a couple of whitespace issues.
Using // coverity[...] or /* coverity[...] */ comments in the source
it is possible to suppress specific Coverity errors. The suppressed
error should occur in the line following the comment.
In this case I have suppressed two false positives from Coverity:
(a) We deliberately assign to a NULL pointer in order to cause a
segfault, for testing how the library reacts when this happens.
Coverity flags this, but it is not an error in this case.
(b) Coverity does not model global variables (a known shortcoming).
Therefore the code 'errno = posix_memalign (...)' cannot be modelled
by Coverity, even though the code is correct. Coverity raises a false
positive about this.
(Thanks Kamil Dudka, Coverity)
If there was no quoting character, the options could run off the end
of the string and into whatever string which happened to be following
in memory.
This bug was revealed when libguestfs was compiled on arm.
For the rationale behind this renaming, see:
b8cbe191d0
You can still, for the time being, use
./configure --with-febootstrap-yum-config
but it's preferred to switch to the new option. The new option is
identical, just renamed.
Qe still pass the deprecated --yum-config option to febootstrap
itself, since both old and new febootstrap support this.
For some reason the fs->device field was not set for swap devices.
Apparently this did not matter before (this bug has existed since at
least libguestfs 1.16), but in commit 6ee80b8dac we started to depend
on the fs->device field being non-NULL. This resulted in segfaults
when inspecting any guest which had a swap device, eg:
#0 __strcmp_sse42 () at ../sysdeps/x86_64/multiarch/strcmp-sse42.S:164
#1 0x00007ffff7d740cb in parent_device_already_probed (g=0x703a70,
partition=0x6f79b0 "/dev/vda4") at inspect.c:156
#2 0x00007ffff7d73e00 in guestfs__inspect_os (g=0x703a70) at inspect.c:83
#3 0x00007ffff7cee11d in guestfs_inspect_os (g=0x703a70) at actions-1.c:397
This commit sets fs->device to the device name.
(Thanks Matthew Booth for first noticing this problem.)
If ntfs-3g is not installed, then tests/guests creates a 'windows.img'
file which is zero-sized.
This is problematical because other tests which use this image will
fail at best, or hit RHBZ#847549 (qemu/virtio-scsi bug) at worst.
Change the tests which use windows.img so they skip if 'windows.img'
has zero size.
If you want to selectively run tests, or if the test suite fails half
way through, use:
make print-subdirs
to print the subdirectories, select the ones you want to run / the
remaining ones, and do:
make check SUBDIRS="..."
User Phill Bandelow noted that virt-resize fails with an e2fsck error
on a host where the system clock had been accidentally set in the
past.
Unfortunately this was hard to diagnose because guestfsd 'ate' the
stdout of the e2fsck program. I have verified by code inspection that
e2fsck prints messages on stdout.
Thus this changes the daemon to fold stdout and stderr together so we
get to see all error messages from e2fsck when it fails.
Rename guestfs_safe_malloc et al to guestfs___safe_malloc etc.
To use the private functions, code now has to define
-DGUESTFS_PRIVATE_FUNCTIONS=1. This will make it easier for us in
future to work out which programs are using these functions and to
minimize both the number of programs and the functions they are
calling.
Note that the Perl, Python, OCaml, Ruby and Java bindings use
guestfs_safe_* calls. None of the other bindings do. This is a bug
(in the bindings using those functions): these functions will call the
out of memory callback on failure. This function defaults to abort(),
and since this happens from a language binding, there is no way to
change this default.
I guess no one uses GCC 3 any more, because if they did, including
<guestfs.h> would not work since GUESTFS_DLL_PUBLIC would not have
been defined.
At the same time, ensure that __attribute__((noreturn)) is only used
when the compiler is GCC >= 3.1.
Currently the code is doing a redundant fork when passed the
COMMAND_FLAG_CHROOT_COPY_FILE_TO_STDIN flag. The additional
sub-process does a chroot() which has no effect because all file
handles are already open at that point, then simply copies its input
to its output.
This change simply replaces the above with a dup2 of the passed file
handle to STDIN of the command process.
RWMJ:
Don't initialize 'pid' variable.
Improve readability of commandrvf() by replacing bare int values for
file descriptors with their symbolic names STD{IN,OUT,ERR}_FILENO.
Also add PIPE_READ and PIPE_WRITE for referencing relevant ends of a pipe.
When executing a command, we temporarily chroot, fork and exec the
command, then chroot back. We intentionally don't chdir in the parent
process so that we can 'jailbreak' the chroot later. However, this has
the effect that commands are executed with a current working directory
which is outside the current root. This unusual state can cause
errors in executed commands which don't anticipate it.
This change does a chdir("/") before executing and command. This
happens inside the fork, so the jailbreak isn't affected in the
parent.
Commit ed7fda161e changed the g->drives
linked list into an array, and also made a subtle change to the
(undocumented) semantics of drives in the rare case when qemu is
relaunched.
Every time we go back to the CONFIG state, we should free the drives
array. However there was one place where this was not happening.
This was caught by the tests of the protocol, but only when using the
default (appliance) method, not the libvirt method which we were more
interested in testing in Fedora.
When testing on Debian using the appliance method, we therefore hit
this bug for the first time.
This relates to or fixes the following commits:
commit 9df159b34a
commit 33f49d85c2
commit ed7fda161e
Since we as developers rarely test the case where some library is
statically not available, that side of the code was hardly tested,
except by unfortunate users in the field who often hit cases where
functions were missing or misdeclared. In fact, when making this
change I noticed several bugs like that.
Change it so that this code is autogenerated, and therefore always
correct and up to date.
Previous code which looked like this:
int
optgroup_acl_available (void)
{
return 0;
}
char * __attribute__((noreturn))
do_acl_get_file (const char *path, const char *acltype)
{
abort ();
}
/* etc */
is replaced by a single line:
OPTGROUP_ACL_NOT_AVAILABLE
Since I made lots of changes to the Erlang bindings (prompted by
Coverity), I wasn't confident that something didn't break because
there were no real tests before.
Comment from Coverity:
libguestfs-1.19.66/erlang/erl-guestfs-proto.c:115: sign_extension:
Suspicious implicit sign extension: "buf[0]" with type "unsigned char"
(8 bits, unsigned) is promoted in "(buf[0] << 24) | (buf[1] << 16) |
(buf[2] << 8) | buf[3]" to type "int" (32 bits, signed), then
sign-extended to type "unsigned long" (64 bits, unsigned). If
"(buf[0] << 24) | (buf[1] << 16) | (buf[2] << 8) | buf[3]" is greater
than 0x7FFFFFFF, the upper bits of the result will all be 1.
In the previous code, appliance was unconditionally set to NULL and so
we never added the dummy appliance slot to g->drives. This probably
doesn't matter, since hotplugging doesn't work for the 'appliance'
attach-method. Despite that, it's better to get this right.
This fixes commit ed7fda161e.
For some guestfs_set_* calls, add checks for error, when error might
possibly occur. eg. It's plausible that guestfs_set_network might
fail if the attach-method being used doesn't support it (although this
doesn't happen at the moment).
In other cases, don't check for errors, eg. if the error doesn't
matter or there's nothing we could plausibly do about it.
These configure flags enable code profiling (with gprof) and code
coverage (with gcov) respectively.
Although this is a nice idea, it's not currently very useful.
Libtool mangles filenames in such a way that gcov cannot locate its
datafiles.
Profiling is of dubious utility with libguestfs which is not CPU-bound
and relies extensively on running external programs (oprofile-like
system profiling that took into account libguestfs + qemu or
libguestfs + qemu + the appliance + filesystem tools *would* be
useful).
Also neither flag will help in capturing data from the appliance.
latest archlinux has some small changes:
btrfs-progs-unstable --> btrfs-progs
genisoimage --> cdrkit
iproute --> iproute2
module-init-tools --> kmod
procps --> procps-ng
udev --> systemd
util-linux-ng --> util-linux
and could you plz add the below packages
lrzip
less
sed
RWMJ: I took this opportunity to make some general cleanups to the file:
- add 'less' and 'sed' for everyone
- 'module-init-tools' & 'kmod'[1]
- 'procps' & 'procps-ng'[1]
- 'util-linux' & 'util-linux-ng'[1]
- sort the ArchLinux packages into order
[1] Note that febootstrap is supposed to ignore packages it doesn't
know about.
The way the existing configure script worked, @VG@ would be empty, and
it would run all the way through the check-valgrind rule successfully
without actually running anything under valgrind.
See the comment from the code:
NB about Windows 8: No location we've found so far contains a
suitable icon for Win8. In particular, explorer.exe definitely
does *not* contain any Windows logo as a resource (I checked).
Therefore the "basket icon" that this produces is just a stand-in
until we have a better idea for solving this problem.
and:
https://bugzilla.redhat.com/show_bug.cgi?id=801117
This is very well explained in the bugzilla comments:
https://bugzilla.redhat.com/show_bug.cgi?id=880801#c1
With this change, the disk name has a one or more plus signs ('+')
appended to show that there are additional disks:
$ virt-df -a Win7x32TwoDisks-a -a Win7x32TwoDisks-b
Filesystem 1K-blocks Used Available Use%
Win7x32TwoDisks-a+:/dev/sda1 102396 24712 77684 25%
Win7x32TwoDisks-a+:/dev/sda2 12478460 7403416 5075044 60%
Win7x32TwoDisks-a+:/dev/sdb1 521212 55728 465484 11%
The read_all function is used as a callback for
guestfs___cmd_set_stdout_callback (cmd, read_all, [str],
CMD_STDOUT_FLAG_WHOLE_BUFFER);
As noted in the documentation for CMD_STDOUT_FLAG_WHOLE_BUFFER, the
buffer returned is not \0-terminated, and so using memdup will create
an unterminated string, and therefore potentially a memory overrun
when reading or searching the string.
Use strndup instead so the final string is \0-terminated.
Try to get the host's loops_per_jiffy value and pass this on the
command line to the guest. In theory this should avoid the appliance
having to recalculate this value in a VM with TCG (which is generally
error-prone). This should avoid timing issues.
We only do this when we are certain that the guest will be TCG.
Currently we only have enough information to do this from the libvirt
attach-method. So mostly this will only affect people using Fedora in
a VM.
The host loops_per_jiffy value is not exported by the kernel. It is
only printed by the kernel early during boot, so if boot messages have
"scrolled off" the kernel ring buffer, it won't be available. Some
operating systems save early kernel messages in /var/log/dmesg but (a)
Fedora 18+ seem to have abandoned this file and (b) on Ubuntu this
file is unreadable for spurious "security" reasons.
I have submitted a patch to make lpj available through /proc/cpuinfo.
Fix the following memory leak found by valgrind:
==13629== 498 (112 direct, 386 indirect) bytes in 1 blocks are definitely lost in loss record 99 of 110
==13629== at 0x4A06B2F: calloc (vg_replace_malloc.c:593)
==13629== by 0x4CA564E: guestfs_safe_calloc (alloc.c:71)
==13629== by 0x4CA9B02: guestfs___new_command (command.c:143)
==13629== by 0x4CA66E9: guestfs___build_appliance (appliance.c:690)
==13629== by 0x4CBD1B9: launch_libvirt (launch-libvirt.c:188)
==13629== by 0x402E7E: main (virt-filesystems.c:349)
Also adjust the command code in several places to make it simpler. We
can almost always call guestfs___cmd_close right after
guestfs___cmd_run, avoiding any need to close the handle along error
paths.
Tested by running the test suite under valgrind.
Because the shell script was missing a space, the final output test
was failing, although for some reason this didn't cause the whole test
to fail.
Fixing the missing space also revealed a bug in the test output, which
is also fixed.
See http://article.gmane.org/gmane.comp.lang.lua.general/95065
Note that this is incompatible with existing code. You
have to do:
local G = require "guestfs"
local g = G.create ()
ie. give the module your own name ("G" in that example).
This is faster, but more importantly it avoids the strange error
'lua: attempt to index a string value' which appears with (some)
single element lists.
- add support for events (with test)
- test progress messages
- update documentation to describe events
- refactor handle closing code
- refactor error code
- use 'assert' in test code instead of 'if ... then error end'
This file was present in git, but missing from the tarball.
Don't actually run the test: it depends on details of how the
appliance is constructed which would fail for old-style appliances.
'make extra-tests' was a monolithic set of tests that did all sorts of
things: valgrind, tests over local guests, tests with upstream qemu,
tests with upstream libvirt, tests with the appliance attach method.
This made it hard to perform individual tests, eg. just valgrind
testing. It was also hard to maintain because the tests were not
located in the same directories as the programs and sometimes
duplicated tests that were run elsewhere.
This commit splits up 'make extra-tests' into 5 separate targets:
make check-valgrind # run a subset of tests under valgrind
make check-valgrind-local-guests # test under valgrind with local guests
make check-with-appliance # test with attach-method == appliance
make check-with-upstream-qemu # test with an alternate/upstream qemu
make check-with-upstream-libvirt # test with an alternate/upstream libvirt
(You can also still run 'make extra-tests' which is now simply
a rule that runs the above 5 targets in order).
This replaces everything that was in the tests/extra directory,
so that has now gone.
This is mostly just code motion, but it also changes the default
location where valgrind.log is written to be the top build directory
(instead of tests/extra/valgrind.log).
This file had grown by accretion to include:
- code related to handles (now in src/handle.c)
- safe allocation (src/alloc.c)
- debug, errors, warnings (src/errors.c)
- private data (src/private-data.c)
- miscellaneous functions (src/canonical-name.c, src/utils.c)
This commit also removes about a dozen #include files which were
probably not really used.
This is just code motion.
If you have a function which is in the X optgroup, and also has
IfAvailable X, then two tests for the optgroup are added to the
output.
Check for this case and give an error.
This also fixes existing APIs.
guestfs_pop_error_handler.
The addition of these calls makes it much simpler to temporarily
disable or ignore errors:
guestfs_push_error_handler (g, NULL, NULL);
guestfs_mkdir (g, "/foo"); /* We don't care if this fails. */
guestfs_pop_error_handler (g);
Note these calls are used by the C API only. They are not necessary
in other languages, because those map errors to exceptions.
The subsequent commits change existing code to use push/pop instead of
laboriously saving and restoring the error handler.
New APIs: set-tmpdir, get-tmpdir, set-cachedir, get-cachedir.
The current code has evolved over time and has a number of problems:
(a) A single environment variable ($TMPDIR) controls the
location of several directories.
(b) It's hard for the library user to predict which directory
libguestfs will use, unless the user simulates the same internal steps
that libguestfs performs.
This commit fixes these issues.
(a) Now three environment variables control the location of all small
temporary files, and the appliance cache:
For temporary files: $LIBGUESTFS_TMPDIR or $TMPDIR or /tmp.
For the appliance cache: $LIBGUESTFS_CACHEDIR or $TMPDIR or /var/tmp.
The user can also set these directories explicitly through API calls
(guestfs_set_tmpdir and guestfs_set_cachedir).
(b) The user can also retrieve the actual directories that libguestfs
will use, by calling guestfs_get_tmpdir and guestfs_get_cachedir.
These functions are also used internally.
This commit also:
- reworks the internal tmpdir code
- removes the internal (undocumented) guestfs_tmpdir call (replacing
it with calls to the documented guestfs_get_tmpdir API instead)
- changes the ./run script to set LIBGUESTFS_TMPDIR and
LIBGUESTFS_CACHEDIR
- adds a test
- fixes a few places like libguestfs-make-fixed-appliance which
depended on $TMPDIR
Previously wipefs (the utility) allowed this, but it's obviously a bad
thing to do and the new wipefs gives an error if you try it:
wipefs: error: /dev/VG/LV: probing initialization failed: Device or resource busy
Split the functions according to a simple hash across
C.nr_actions_files files (this number can be increased later if
necessary).
This is just code motion.
This is currently done implicitly because of the linker script.
However in order to do symbol versioning, we will have to do
this explicitly at each definition instead.
Use <cpu mode="host-model"> (not model=... which was a typo).
Also set fallback=allow so that libvirt knows it can use another
CPU model if it wants.
This fixes commit 890a4fbc87.
Commit 2e90f43129 added a debug message
which isn't filtered out by the set of grep -v's in this test.
Therefore this test would fail when run with LIBGUESTFS_DEBUG=1.
Commit a0722c7ad8 introduced a
dependency on the Unix module. This was not listed in the list of
'-package's, but as long as you had ocaml-gettext installed it would
still work because that pulled in Unix implicitly.
Thanks Olaf Hering.
This is a wrapper or mini-library for running external command,
loosely based on libvirt's virCommand interface.
Amongst the advantages are:
- Can redirect errors into the error log (RHBZ#713678).
- Can redirect output into a callback function.
- Handles shell quoting properly.
- Safely resets signal handlers, closes file descriptors, etc.
- Single place where we can implement other improvements in future.
Used to create temporary directory or file with an optional suffix.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Split this out into a new file (daemon/mktemp.c).
- I don't see a reason to deprecate the mkdtemp function which
works fine. Instead remove complex dir-making code from the
new function.
- Test and fix the patch (missing close(fd)).
The cleanup of package managers cache is already implemented
by sprep_operation_package_manager_cache.ml, so remove this
TODO item.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
In languages like Python where we release a global lock around
long-running libguestfs functions, it is also useful to *not* release
this lock for small, non-blocking functions.
Therefore mark all functions with a 'blocking' boolean flag. It
defaults to true, and is true by definition for all daemon functions.
For non-daemon functions, I have classified them manually.
Only when the blocking flag is set do we generate the code to release
and reacquire the lock around libguestfs calls.
guestfs_parse_environment_list.
Add a new function for creating a handle:
guestfs_h *guestfs_create_flags (unsigned flags [, ...]);
This variant lets you supply flags and extra arguments, although extra
arguments are not used at the moment.
Of particular interest is the ability to separate the creation of the
handle from the parsing of environment variables like
LIBGUESTFS_DEBUG. guestfs_create does both together, which prevents
us from propagating errors from parsing environment variables back to
the caller (guestfs_create has always printed any errors on stderr and
then just ignored them).
If you are interested in these errors, you can now write:
g = guestfs_create_flags (GUESTFS_CREATE_NO_ENVIRONMENT);
if (!g)
exit (EXIT_FAILURE);
r = guestfs_parse_environment (g);
if (!r)
exit (EXIT_FAILURE);
Also you can *omit* the call to guestfs_parse_environment, which
creates a handle unaffected by the environment (which was not possible
before).
This commit also includes new (backwards compatible) changes to the
OCaml, Perl, Python, Ruby and Java constructors that let you use the
flags.
This commit models libvirt authentication events through the API,
adding one new event (GUESTFS_EVENT_LIBVIRT_AUTH) and several new
APIs:
guestfs_set_libvirt_supported_credentials
guestfs_get_libvirt_requested_credentials
guestfs_get_libvirt_requested_credential_prompt
guestfs_get_libvirt_requested_credential_challenge
guestfs_get_libvirt_requested_credential_defresult
guestfs_set_libvirt_requested_credential
See the documentation and example which shows how to use the new API.
This commit also changes existing calls to virConnectOpen* within the
library so that the new API is used.
Also included is an example (but not a test, because it's hard to see
how to automatically test the libvirt API).
Normally we check if libvirt is installed, and the default is still to
do this.
However having this option makes it simpler to check if a change to
the code has broken the case where libvirt is not installed.
In fact, this change revealed one such place, which is also fixed in
this commit.
This means that we can use virConnectPtr and virDomainPtr instead of
using void* types. In the case where libvirt support is disabled or
not present, effectively nothing changes.
If TMPDIR is not set, we must choose one, because otherwise libvirt
will use a random TMPDIR: http://bugzilla.redhat.com/865464
The convenient guestfs___persistent_tmpdir function does everything
needed in this case.
The code was identical -- just copied with s/resize/sparsify/.
Instead of duplicating identical code, cause the Makefile.am to use
the code from the ../resize/ directory.
Unfortunately because there are two Utils modules (which are
different), this means we had to rename those modules to Resize_utils
and Sparsify_utils respectively. So this is a rather larger change
than intended. However it's just code motion.
Revert "btrfs: Add an extended workaround for btrfs failures seen with kernel 3.7.0."
Reverted these workaround, since we may have found a fix for the btrfs
bug itself (for details see RHBZ#863978).
This reverts commit d9e5b514aa
and commit a03f536f0d.
Commit ed7fda161e changed the
way that the drives are handled across appliance shutdowns.
Previously during the following sequence of calls:
create the handle
add drive(s)
launch
kill subprocess
launch
the added drives are still in the handle at the second launch.
After the above commit, the added drives are removed from the handle,
which means the second launch happens with no drives (which could be
an error).
This was never actually defined either way, so in this case fix the
test to re-add the drive before the second launch.
Since hotplugging was added, it isn't really feasible to return to the
original semantics, since users might remove drives, in which case we
have lost information about those drives so we cannot restore them on
the second launch.
NOTE: PLEASE CALLERS DON'T DO THIS! Always use a new handle for each
launch of the appliance.
When libvirt is used, we can allow disks to be hotplugged.
guestfs_add_drive can be called after launch to hot-add a disk.
When a disk is hot-added, we first ask libvirt to add the disk to the
appliance, then we make an internal call into the appliance to get it
to wait for the disk to appear (ie. udev_settle ()).
Hot-added disks are tracked in the g->drives array.
This also adds a test.
New API: list-disk-labels
Allow the user to pass an optional disk label when adding a drive.
This is passed through to qemu / libvirt using the disk serial field,
and from there to the appliance which exposes it through udev,
creating a special alias of the device /dev/disk/guestfs/<label>.
Partitions are named /dev/disk/guestfs/<label><partnum>.
virtio-blk and virtio-scsi limit the serial field to 20 bytes. We
further limit the name to maximum 20 ASCII characters in [a-zA-Z].
list-devices and list-partitions are not changed: these calls still
return raw block device names. However a new call, list-disk-labels,
returns a hash table allowing callers to map between disk labels, and
block device and partition names.
This commit also includes a test.
Instead of adding the snapshot=on option via <qemu:arg>, create qcow2
overlays for any read-only drives and the appliance using 'qemu-img
create' + a temporary file.
This is a workaround for missing support for <transient/> in libvirt's
qemu driver. Also for the unpredictable way that libvirtd handles
$TMPDIR: we want to control where the temporary disk is created.
Currently it is also much slower, because qemu-img is slow. However
we hope to fix qemu upstream.
(RHBZ#858128).
If compare_device_names was given two devices with devices with
different interfaces (eg. /dev/sda and /dev/vda) then it would try to
compare the partition numbers, and fail when it could parse them.
It's arguable what we should be doing in this case (except for
strongly discouraging people from using the interface feature), but
let's at least not cause the daemon to assert-fail.
Found by Red Hat QA, thanks Mohua Li.
Don't assume that if guestfs_case_sensitive_path returns NULL, that it
means the file does not exist.
The (previously undefined) behaviour of case_sensitive_path was that a
NULL return meant "either the file doesn't exist or some other error".
However in commit 973581780d this was
changed so that if the last element of the path didn't exist, it was
assumed to be a new file and the (non-NULL) path of the new file is
returned.
This change breaks code (including in libguestfs) which tries to use
case_sensitive_path as a dual-purpose call to fix-up a path for
Windows and test if the file exists. Such code should be rewritten so
that it explicitly tests for file existence after calling
case_sensitive_path.
I examined all the calls to case_sensitive_path in libguestfs and
modified them where necessary.
The 'windows_path' function was blindly copied from virt-cat. In
virt-cat, errors are checked by the caller to 'windows_path'. But
virt-edit lacks this check. Change the function in virt-edit to add a
check and exit on error.
There's not really any point just documenting that people
should label the tmp directory, when we can just label it.
So change the ./run script so it calls chcon.
However it's not a failure if chcon doesn't work. If SELinux
is not installed or is disabled, then labelling will fail.
Everyone ignores the documentation. Everyone ignores me even
when I specifically tell them to include the complete output.
Place a prominent notice at the beginning of the output.
Let tests/bigdirs/test-big-dirs.pl uses mke2fs to create small
inode ratio files.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Fix invocation of mke2fs with optional arguments.
- Run the test to check it works.
New api mke2fs for full configuration of filesystem.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Update description.
- Run and fix the tests.
- Remove bogus filename from po/POTFILES.
Without this empty rule, automake doesn't generate the rest of the
testing infrastructure, so you can't enable the test yourself by doing
'make check TESTS=test-big-dirs.pl'.
Since generator source files were renamed, the comment at the
top of each generated file was wrong.
Unfortunately we cannot allow /* to appear within a comment,
so the space is necessary.
cron-spool claims to remove at jobs, but it has no code to actually do
that. Add patterns to remove files in known at spool locations.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Currently firstboot would only work on redhat-based images.
Handle redhat-based, suse-based and debian guests, error out in case of an
unknown distro.
Update firstboot.sh:
- make sure scripts exists and can be executed
- add LSB header to avoid insserv warnings later on
- run script only if called with "start"
Update functions, pass only required options.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
The second parameter to 'config' may be NULL.
In commit 52fa23d74f (refactoring of
guestfs_config) the code this got lost, and guestfs_config would
segfault if qemu_value was NULL.
Also this fixes the libvirt method to handle the same case.
I checked libguestfs-1.18 and -1.16 branches, and this problem does
NOT affect them.
Update SuSE Linux detection. Up to now everything with a
/etc/SuSE-release file was treated as openSuSE. With this change SLES
based distributions such as "SUSE Linux Enterprise Server", "SUSE
Linux Enterprise Desktop" and "Novell Linux Desktop" will show up as
"sles". The 'opensuse' detection was updated to handle older openSuSE
releases as well as the well known SuSE Linux releases, starting from
6.0. Everything else with a /etc/SuSE-release file will show up as
"suse-based".
Here is a collection of release files:
==> Dist/6.0/etc/SuSE-release <==
SuSE Linux 6.0 (i386)
VERSION = 6.0
==> Dist/6.1/etc/SuSE-release <==
SuSE Linux 6.1 (i386)
VERSION = 6.1
==> Dist/6.2/etc/SuSE-release <==
SuSE Linux 6.2 (i386)
VERSION = 6.2
==> Dist/6.3/etc/SuSE-release <==
SuSE Linux 6.3 (i386)
VERSION = 6.3
==> Dist/6.4/etc/SuSE-release <==
SuSE Linux 6.4 (i386)
VERSION = 6.4
==> Dist/7.0/etc/SuSE-release <==
SuSE Linux 7.0 (i386)
VERSION = 7.0
==> Dist/7.1/etc/SuSE-release <==
SuSE Linux 7.1 (i386)
VERSION = 7.1
==> Dist/7.2/etc/SuSE-release <==
SuSE Linux 7.2 (i386)
VERSION = 7.2
==> Dist/7.3/etc/SuSE-release <==
SuSE Linux 7.3 (i386)
VERSION = 7.3
==> Dist/8.0/etc/SuSE-release <==
SuSE Linux 8.0 (i386)
VERSION = 8.0
==> Dist/8.1/etc/SuSE-release <==
SuSE Linux 8.1 (i386)
VERSION = 8.1
==> Dist/8.2/etc/SuSE-release <==
SuSE Linux 8.2 (x86-64)
VERSION = 8.2
==> Dist/9.0/etc/SuSE-release <==
SuSE Linux 9.0 (x86-64)
VERSION = 9.0
==> Dist/9.1/etc/SuSE-release <==
SuSE Linux 9.1 (x86-64)
VERSION = 9.1
==> Dist/9.2/etc/SuSE-release <==
SuSE Linux 9.2 (x86-64)
VERSION = 9.2
==> Dist/9.3/etc/SuSE-release <==
SuSE Linux 9.3 (x86-64)
VERSION = 9.3
==> Dist/10.0/etc/SuSE-release <==
SUSE LINUX 10.0 (X86-64) OSS
VERSION = 10.0
==> Dist/10.1-remastered/etc/SuSE-release <==
SUSE LINUX 10.1 (X86-64)
VERSION = 10.1
==> Dist/10.1/etc/SuSE-release <==
SUSE Linux 10.1 (X86-64) Beta8
VERSION = 10.1
==> Dist/10.2/etc/SuSE-release <==
openSUSE 10.2 (X86-64)
VERSION = 10.2
==> Dist/10.3/etc/SuSE-release <==
openSUSE 10.3 (X86-64)
VERSION = 10.3
==> Dist/11.0/etc/SuSE-release <==
openSUSE 11.0 (X86-64)
VERSION = 11.0
==> Dist/11.1/etc/SuSE-release <==
openSUSE 11.1 (x86_64)
VERSION = 11.1
==> Dist/11.2/etc/SuSE-release <==
openSUSE 11.2 (x86_64)
VERSION = 11.2
==> Dist/11.3/etc/SuSE-release <==
openSUSE 11.3 (x86_64)
VERSION = 11.3
==> Dist/11.4/etc/SuSE-release <==
openSUSE 11.4 (x86_64)
VERSION = 11.4
CODENAME = Celadon
==> Dist/12.1/etc/SuSE-release <==
openSUSE 12.1 (x86_64)
VERSION = 12.1
CODENAME = Asparagus
==> Dist/12.2/etc/SuSE-release <==
openSUSE 12.2 (x86_64)
VERSION = 12.2
CODENAME = Mantis
==> Dist/sles8/etc/SuSE-release <==
SuSE SLES-8 (AMD64)
VERSION = 8.1
==> Dist/sles9/etc/SuSE-release <==
SUSE LINUX Enterprise Server 9 (x86_64)
VERSION = 9
==> Dist/sles9sp2/etc/SuSE-release <==
SUSE LINUX Enterprise Server 9 (x86_64)
VERSION = 9
PATCHLEVEL = 2
==> Dist/sles9sp3/etc/SuSE-release <==
SUSE LINUX Enterprise Server 9 (x86_64)
VERSION = 9
PATCHLEVEL = 3
==> Dist/sles9sp4/etc/SuSE-release <==
SUSE LINUX Enterprise Server 9 (x86_64)
VERSION = 9
PATCHLEVEL = 4
==> Dist/sled10/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 10 (x86_64)
VERSION = 10
==> Dist/sled10sp1/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 10 (x86_64)
VERSION = 10
PATCHLEVEL = 1
==> Dist/sled10sp2/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 10 (x86_64)
VERSION = 10
PATCHLEVEL = 2
==> Dist/sled10sp3/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 10 (x86_64)
VERSION = 10
PATCHLEVEL = 3
==> Dist/sled10sp4/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 10 (x86_64)
VERSION = 10
PATCHLEVEL = 4
==> Dist/sled11/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 11 (x86_64)
VERSION = 11
PATCHLEVEL = 0
==> Dist/sled11sp1/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 11 (x86_64)
VERSION = 11
PATCHLEVEL = 1
==> Dist/sled11sp2/etc/SuSE-release <==
SUSE Linux Enterprise Desktop 11 (x86_64)
VERSION = 11
PATCHLEVEL = 2
==> Dist/sles10/etc/SuSE-release <==
SUSE Linux Enterprise Server 10 (x86_64)
VERSION = 10
==> Dist/sles10sp1/etc/SuSE-release <==
SUSE Linux Enterprise Server 10 (x86_64)
VERSION = 10
PATCHLEVEL = 1
==> Dist/sles10sp2/etc/SuSE-release <==
SUSE Linux Enterprise Server 10 (x86_64)
VERSION = 10
PATCHLEVEL = 2
==> Dist/sles10sp3/etc/SuSE-release <==
SUSE Linux Enterprise Server 10 (x86_64)
VERSION = 10
PATCHLEVEL = 3
==> Dist/sles10sp4/etc/SuSE-release <==
SUSE Linux Enterprise Server 10 (x86_64)
VERSION = 10
PATCHLEVEL = 4
==> Dist/sles11/etc/SuSE-release <==
SUSE Linux Enterprise Server 11 (x86_64)
VERSION = 11
PATCHLEVEL = 0
==> Dist/sles11sp1/etc/SuSE-release <==
SUSE Linux Enterprise Server 11 (x86_64)
VERSION = 11
PATCHLEVEL = 1
==> Dist/sles11sp2/etc/SuSE-release <==
SUSE Linux Enterprise Server 11 (x86_64)
VERSION = 11
PATCHLEVEL = 2
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Since this is the most common error seen by people who have
installation problems, buggy qemu, etc, and since no one reads the
FAQ, describe in this error message what resources are available to
debug launch problems.
libzypp and zypper store their cache files in
/var/cache/zypp/{packages,raw,solv}/ and /var/cache/zypper/RPMS/
Add a single pattern to match both cases.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
SuSE based installations store the hostname in /etc/HOSTNAME.
Add code to handle both opensuse and sles. Code to properly detect
the latter will be added with another patch.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
<file>: error: jump skips variable initialization [-Werror=jump-misses-init]
This has only just appeared, possibly related to previous gnulib
update. In any case, this is just code motion / cleanup.
For some reason, argument declarations were handled
inconsistently for strings and non-string arguments.
Non-string arguments were all declared at the top of the
function stub, where as string arguments were declared
in place.
Be consistent.
This is just code motion.
* cfg.mk (sc_prohibit_trailing_blank_lines): Remove rule. It is now
subsumed by the much more efficient sc_prohibit_empty_lines_at_EOF
rule in gnulib's maint.mk.
Passing the $upstream_qemu environment variable is not reliable when
libvirt is involved: Environment variables are passed to the first
instance of the session libvirtd, but because libvirt reuses the
session libvirtd, subsequent environment variable settings are lost
(this is filed as RHBZ#856619).
Bypass all of this by writing a custom qemu-wrapper.sh which contains
the hard-coded values we want.
On Linux PATH_MAX is 4096, but on some platforms it can be much larger
or even not defined (ie. unlimited). Therefore using a PATH_MAX-sized
stack buffer is not a great idea for portable programs.
This change removes use of PATH_MAX-sized stack-allocated buffers.
This change only applies to the library and standalone programs.
Inside the daemon, memory allocation is much more complicated so I
have not changed those (yet).
Found by 'make syntax-check'.
If these are not defined in the header files, it's a problem on
the platform which we should fix with gnulib. Our replacement
definitions might be wrong here.
Having a separate directory means:
(1) It's easy to clean up orphaned temporary files, the appliance,
etc.
(2) You can put an SELinux label on this directory so that qemu can
write to it when you're using sVirt and SELinux is enforcing:
chcon --reference=/tmp tmp
This is a fix for multiprogramming: If two instances of libguestfs
share the same appliance disk, then libvirt would unlabel the disk
when one of the instances closes the handle, resulting in the other
qemu being unable to continue accessing the appliance.
Adding the flag makes libvirt understand that the disk is shared so it
doesn't do this, and it apparently handles locking correctly too if we
were using sanlock.
Since 'QEMU' is set (overwritten) by the autoconf-generated Makefile,
this test would always test installed qemu, not upstream qemu,
making it pretty useless. Use another name for this variable.
If you go back in time in git (eg. git reset, git bisect) then you can
end up in a situation where a file that was generated by a later
version is left around unused in the earlier version.
This isn't a problem for most things, but gobject documentation
generation picks up any file in a directory, even unreferenced ones,
and breaks. So the correct thing to do is to remove these files.
I: Program returns random data in a function
E: libguestfs no-return-in-nonvoid-function guestmount.c:75
The function fuse_opt_add_opt_escaped has only one caller and a return
code is not checked.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
This hint tells the backend whether anyone cares about errors when the
appliance is shut down.
Currently this only has any effect on the libvirt backend, where it
controls whether or not we use the VIR_DOMAIN_DESTROY_GRACEFUL flag.
This command was not tested at all. As a result we didn't notice that
it was broken for a long time (RHBZ#853159).
This adds a test that drives the command through a pty. It uses the
perl 'Expect' module, although this is not required.
https://bugzilla.redhat.com/show_bug.cgi?id=853159
git bisect pointed to the following commit:
commit ec8e3b6cad
Author: Richard W.M. Jones <rjones@redhat.com>
Date: Fri Jul 20 14:24:10 2012 +0100
launch: Abstract attach method operations.
g->attach_ops points to a structure which contains the
operations supported by each attach method backend
(ie. appliance, unix, etc.).
Since that commit was essentially just code motion, it wasn't clear
why virt-rescue should be affected by it.
In fact the reason is as follows:
(1) In direct mode, we don't need g->fd[] (which would normally be
connected to the stdin/stdout of qemu). So we opened them on
/dev/null so they had some value.
(2) accept_from_daemon / read_log_message_or_eof reads from g->fd[1].
Since this is connected to /dev/null, it always reads EOF.
(3) This would cause child_cleanup to be called. This is completely
unintentional: we don't want to cleanup the child at this point, even
in direct mode.
(4) Prior to the commit above, child_cleanup first waited for the
process to exit (ie. waitpid). This happened to work, since we are
effectively waiting for the user to exit virt-rescue.
(5) After the commit above, the order of operations was changed so
that we first killed qemu before waiting for it. This broke
virt-rescue.
The fix is to change direct mode so that it leaves g->fd[]'s as -1.
The rest of the protocol code can deal with this situation -- it
ignores the log fd instead of trying to read from it.
The libvirt backend currently doesn't support direct mode. We should
make that work in future. As a workaround, force the attach-method to
be 'appliance' in this case.
We set the error handler to NULL in order to mask "normal" error
messages that we expect to see because of the unusual way that
virt-rescue runs the appliance. However if the user selected -v /
enabled debugging, then it is reasonable to expect they want to see
every message, so do not mask anything.
guestfs_disk_image makes a symbolic link to the real filename in order
to sanitize the filename. However this fails if the filename is a
relative path. Call realpath(3) to make the filename canonical.
This fixes commit 20902e7ce0.
udevadm is included in all reasonable recent distributions.
This avoids 'command not found' errors in verbose mode.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
This is a simple renaming of the files/modules.
Note that in OCaml, module names are derived from filenames by
capitalizing the first letter. Thus the old module names had the form
"Generator_api_versions". The new modules names have the form
"Api_versions".
This is almost just code motion. While doing this I discovered two
prototypes for functions which no longer exist:
* guestfs___launch_appliance
* guestfs___launch_unix
so these prototypes have been removed. The rest are rearranged more
logically.
guestfsd calls many different tools. Keeping track of all of them is
error prone. This patch introduces a new helper macro to put the command
string into its own ELF section:
GUESTFSD_EXT_CMD(C_variable, command_name);
This syntax makes it still possible to grep for used command names.
The actual usage of the collected list could be like this:
objcopy -j .guestfsd_ext_cmds -O binary daemon/guestfsd /dev/stdout |
tr '\0' '\n' | sort -u
The resulting output will be used to tell mkinitrd which programs to
copy into the initrd.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
RWMJ:
- Move str_vgchange at request of author.
- Fix snprintf call in daemon/debug.c
Add a new api xfs_repair for repairing an XFS filesystem.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Fix non-error return path so it doesn't send two replies.
- Document return code.
I tested this by comparing the output of virt-inspector over Windows
guests before and after the change, which was identical:
$ md5sum `ls -1 /tmp/*.before /tmp/*.after`
c292d6629b5a761eccb4a279754399b4 /tmp/Win2003.after
c292d6629b5a761eccb4a279754399b4 /tmp/Win2003.before
eb1e1ff29208a9ee46e9c100dfec26b2 /tmp/Win2012.after
eb1e1ff29208a9ee46e9c100dfec26b2 /tmp/Win2012.before
d060a95d7ffe5dce6c4e66feb80c2837 /tmp/Win7x32.after
d060a95d7ffe5dce6c4e66feb80c2837 /tmp/Win7x32.before
8914eee70ac4f8a0317659e09e00dcdc /tmp/Win7x32Dynamic.after
8914eee70ac4f8a0317659e09e00dcdc /tmp/Win7x32Dynamic.before
a2dcdfc0f9d64054640875aa791889e0 /tmp/Win7x32TwoDisks.after
a2dcdfc0f9d64054640875aa791889e0 /tmp/Win7x32TwoDisks.before
5ed49568a5147dce7517c99de41ebf2e /tmp/Win8previewx64.after
5ed49568a5147dce7517c99de41ebf2e /tmp/Win8previewx64.before
fdfc7d272b79a665ae3313ae1ae30660 /tmp/WinXP.after
fdfc7d272b79a665ae3313ae1ae30660 /tmp/WinXP.before
3c705444be664f1316b21c5d8d3cb0be /tmp/WinXPRecConsole.after
3c705444be664f1316b21c5d8d3cb0be /tmp/WinXPRecConsole.before
Transscribe many hivex(3) APIs into the libguestfs API.
There is one hive handle per libguestfs handle, as with Augeas.
Note that hivex uses iconv_open for some APIs (eg. hivex_value_string).
But since we delete all the i18n files from the appliance, this
doesn't work -- iconv_open returns EINVAL. Therefore hivex APIs which
require iconv cannot be bound in the daemon.
The library doesn't actually use libaugeas, except indirectly via the
libguestfs API. The libguestfs API implements this in the daemon, so
there's no need for the library to link to augeas at all.
libguestfs fails to build with --enable-fuse on openSuSE 11.4 and
earlier because the included fuse version does not include
libulockmgr.so. configure already used pkgconfig to retrieve the correct
CFLAGS, so there is no need to hardcode -lulockmgr.
With this change the build succeeds again.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Without this patch, if JAVAC points to a valid executable, --with-java-home=no
is not respected but the build simply fails.
Furthermore, --with{,out}-java is a more common form and more in-line with all
other options.
The assert (conn != NULL) was being triggered with this
stack trace:
at launch-libvirt.c:1305
fd=<optimized out>, error_if_eof=error_if_eof@entry=0) at proto.c:222
size_rtn=size_rtn@entry=0x7fffffffdb34,
buf_rtn=buf_rtn@entry=0x7fffffffdb58) at proto.c:548
libvirt_uri=<optimized out>) at launch-libvirt.c:391
--disable-perl isn't currently respected because $PERL correctly points
to the perl executable and $missing_perl_modules doesn't prevent the
perl parts to be installed either. Thus, a check for $enable_perl is
needed.
These don't contain literal strings like "AUTHORS" (they have the
translated versions instead) and so checking for these English strings
is rather useless.
Also we don't usually have the '--license' parameter conveniently
around so don't enforce this section.
This adds standard LICENSE and BUGS sections to all of the man pages
that are processed by podwrapper.
Modify all the calls to $(PODWRAPPER) to add the right --license
parameter according to the content. Note that this relaxes the
license on some code example pages, making them effectively BSD-style
licensed.
section.
Ensure each man page contains consistent COPYRIGHT and AUTHOR
sections.
Remove the LICENSE section. We will add that back in podwrapper in a
later commit.
Add new api xfs_admin to change parameters of an XFS filesystem.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Remove printuuid, printlabel, since they don't change any settings.
- Adjusted the documentation.
- Fix the tests.
Since our minimum supported version is now 1.16 and mount was fixed in
1.13.16, it is now safe to replace mount-options + empty options with
mount wherever it occurs.
Note that we keep the old daemon calls, but rename them as
"internal_write" and "internal_write_append". This lets us implement
the new library-side calls more efficiently in the common case when
the uploaded content is smaller than the message buffer. In most
cases the new calls won't end up using a temporary file.
These internal (testing) APIs allow the bindtests output to
be sent to some other place than stdout.
This is necessary for Erlang, since stdout is used to communicate with
the Erlang interpreter.
Previously with Python it was impossible to set a boolean or integer
optarg to -1 because that was used as a special sentinel value to
indicate that the optarg was not set.
Instead, use None as the sentinel value, since that cannot be a
boolean or integer type.
The compress flag can be used to control compression, one of: (none),
"compress", "gzip", "bzip2", "xz", "lzop". Thus these calls can now
be used instead of tgz-in/tgz-out/txz-in/txz-out, and also support
more compression types.
Mark these APIs as once_had_no_optargs so that compatibility code is
generated.
Use the safer, higher level Store_field macro when constructing arrays
of structs to return.
I don't know if it is strictly necessary in this case, but it's safer.
We just remove the process accounting files previously without
touch a empty file, this will cause psacct runs error.
Restart the service can't help us create this file auto.
couldn't open file '/var/account/pacct': No such file or directory
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
In libguestfs 1.20, you will be able to use 'add_drive'
instead of 'add_drive_opts' (except in the C bindings).
However until libguestfs 1.20 is the minimum stable version
people will still be using old versions where you have to use
'add_drive_opts'. This makes the examples confusing.
Therefore continue to use 'add_drive_opts' in the examples
for now.
This commit makes grep into an optargs API, with flags for extended,
fixed, [case-]insensitive and compressed.
At the same time it deprecates: egrep, fgrep, grepi, egrepi, fgrepi,
zgrep, zegrep, zfgrep, zgrepi, zegrepi and zfgrepi.
NB: The patch to implement this feature in qemu is not upstream, and
may never make it upstream. However this is so useful for
virt-sparsify that I decided to add this to libguestfs while we see
what qemu decides to do.
Allow optargs functions to be called as either:
g.mkfs_opts(fs, device)
or:
g.mkfs_opts(fs, device, { optargs hash })
This also preserves backwards compatibility with once_has_no_optargs
functions such as mkfs.
qcow2 v3 is generally more efficient, although unfortunately it
doesn't support automatically sparsifying zero writes.
Note this *only* uses qcow2 v3 for the intermediate overlay file, not
for the final output (since we want the output to be broadly
compatible with old hypervisors).
These APIs will allow sysprep to change the UUIDs of all PVs and VGs
in the system.
LVs don't have UUIDs AFAICT, or at least there seems to be no way to
change them if they do have them.
Operations that need to work directly on guest block devices will fail
because the block devices are busy. Therefore add a phase with the
filesystems unmounted, and allow operations to specify that they need
to work in this phase.
remove the local machine ID, leave it empty, let it be generated
during next booting.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Whitespace change.
- Remove completed items from TODO.
Note this does not yet enable sVirt confinement of the appliance.
That requires at least that SELinux policy goes upstream, plus there
may still be bugs.
Use Dev_or_Path instead of String.
Remove the RESOLVE_DEVICE since Dev_or_Path will generate
REQUIRE_ROOT_OR_RESOLVE_DEVICE instead.
RWMJ:
Note a change in semantics: this now requires root. However this is
OK and still works with mkmountpoint and friends because
'is_root_mounted' works even if something is mounted below the root.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
If ./configure --with-default-attach-method is set to something other
than 'appliance', then this will legitimately return a different
string. Simply test that it runs, rather than testing the output.
This fixes commit 20a5b4de7d.
We were using format= which was bogus and libvirt was ignoring it and
forcing raw format instead.
Also in its default configuration libvirt won't do disk format
autodetection at all, so we must do it instead.
And in any case there is no point, since proto.c already sets the
SOCK_CLOEXEC flag (using accept4) on this socket.
The code now matches the code in launch-appliance.c.
You can now choose the default attach method in two ways:
(1) Set the LIBGUESTFS_ATTACH_METHOD environment variable.
(2) ./configure --with-default-attach-method=appliance|libvirt|...
Note that (1) overrides (2).
We need to select the <domain type=...> attribute depending on whether
KVM is detected in the libvirt capabilities.
This is a hack because it doesn't try to parse the libvirt
capabilities XML.
For compatibility with the appliance backend, devices added readonly
are in fact added writable with snapshots on top. This is necessary
because options like 'mount -o ro' usually require write access to the
disk (eg. to replay journals).
Complete the attach-method libvirt backend.
This backend uses libvirt to create a transient KVM domain to run the
appliance.
Note that this still will only work with local libvirt URIs since the
<kernel>, <initrd> and appliance links in the libvirt XML refer to
local files, and virtio serial only works locally (limitation of
libvirt). Remote support will be added later.
Since we will be calling guestfs___build_appliance from the libvirt
code in future, there's no point having two places where we have to
acquire the lock. Push the lock down into this function instead.
Because "glthread/lock.h" includes <errno.h> we have to add this
header to the file too.
Move and rewrite guestfs_config so it accumulates a list of qemu
parameters in the handle. These are added to the appliance at launch
time (with attach method == unix:... you'll now get an error).
launch-appliance.c contains the code associated with the 'appliance'
attach-method. Mostly. In fact there are a few APIs which don't fit
so nicely:
- config: deprecated API which fiddles with the qemu command
line directly
- max-disks: depends on the qemu implementation (virtio-scsi
or not)
- debug-drives: used for testing only
launch-unix.c contains the code associated with 'unix:<path>'.
launch.c is the common code for launching, along with a few other APIs
such as guestfs_add_drive_opts.
This commit also reduces the number of headers to just those
which are required.
Unfortunately the parallel test keeps hitting this bug:
https://bugzilla.redhat.com/show_bug.cgi?id=838081
which could be a bug in the OCaml runtime.
Just test simple mount-local. We will write a parallel test in C to
replace this.
When using tar-in or tools built around it such as virt-make-fs, if
the target filesystem is vfat then unpacking the tarball would fail
because tar tries to chown(2) files and fails.
You would see errors such as:
tar: <file>: Cannot change ownership to uid 500, gid 500: Operation not permitted
This change detects whether the target filesystem supports chown(2).
If not, it adds the --no-same-owner flag to tar to stop it from trying
to change the ownership of newly created files.
Add proper dependencies for these files, but also use 'cmp' to ensure
they only get overwritten if the new files have actually changed, so
we don't rebuild the appliance unnecessarily.
Add xfs_info to show the geometry of the xfs filesystem.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Updated po/POTFILES.
- Use xfs_ prefix for all struct fields.
- Return uninitialized fields as -1 / empty string.
- Copyedit the description.
case_sensitive_path is undefined when the final path element doesn't
exist. Currently it returns an error, but this means that creating a
new file doesn't work as expected:
$ guestfish --rw -i -d windows touch 'win:c:\blah'
libguestfs: error: case_sensitive_path: blah no file or directory found with this name
We should allow this case (provided there is no trailing slash) so
that new files or directories can be created.
This uses Pod::Simple so it properly parses the input POD and can
generate proper custom output as required specifically by libguestfs.
One immediate benefit is that links between and within manual pages
now work mostly correctly.
By using the once_had_no_optargs flag, this change is backwards
compatible for callers (except Haskell, PHP and GObject as discussed
in earlier commit).
In C, a function called 'func' which has once_had_no_optargs=true will
(because of the previous commit) generate 'func_opts' and a
backwards-compatibility function called 'func'.
This commit changes some of the non-C bindings so that they also
generate 'func_opts' which is merely a wrapper that calls 'func'.
This avoids incompatibility when we rename 'mkfs_opts' etc back to
plain 'mkfs', and it also makes it easier to translate between other
language bindings and C code.
NB: Some bindings do not include aliases:
PHP: There's no way to easily alias methods in PHP < 5.3, and we
can't assume everyone has this minimum version.
GObject: Very complex to add aliases, but we should probably do this
at some point.
Haskell: No support for optargs in these bindings. Unfortunately
this means that we can no longer bind 'Guestfs.add_drive'
(since it will be changed to add optional arguments) making
the Haskell bindings even less useful than they were already.
This commit adds a flag (once_had_no_optargs) which can be used to add
optargs to functions that currently don't have any.
The idea is that if 'func' currently has no optargs, we can safely add
optargs provided we are backwards compatible for existing callers.
In C that means we leave 'guestfs_func' alone and provide an extra
function 'guestfs_func_opts' that takes the optargs ('guestfs_func'
becomes a wrapper that calls 'guestfs_func_opts').
In the C generator this means there are two names for each function
(although the two names are normally identical). 'c_name' is the name
that we export publicly (eg. [guestfs_] 'func_opts'). 'name' is the
internal name of the function (eg. 'func') which is used for
everything apart from the public interface, and also to generate the
no-optargs compat function.
In other languages that can add optional arguments safely, we simply
add the arguments to the existing 'func', so for example in Perl:
$g->func (required_args)
$g->func (required_args, optional_args)
can be used.
Note that this commit does not cause any change to the output of the
generator. I verified this by diffing the output before and after.
For example the existing method:
public void mkfs_opts (String fstype, String device, Map<..> optargs);
is now accompanied by this overloaded method which is a simple wrapper:
public void mkfs_opts (String fstype, String device)
throws LibGuestFSException
{
mkfs_opts (fstype, device, null);
}
This field, which is generated internally by the generator, is the
name of the C function corresponding to each action.
For actions that have NO optional arguments, it's just "guestfs_<name>".
For actions that have any optional arguments, it is
"guestfs_<name>_argv" (since any binding has to construct the optional
argument struct explicitly).
In a future commit, this mapping may become more complex.
This commit also "fixes" the C# bindings which didn't handle optional
arguments properly at all. In fact, it doesn't fix this, it just
changes it enough that it probably now compiles. We should either
compile and test the bindings routinely with Mono, or drop them, since
they are starting to bit-rot.
In the GObject bindings, I have added a space between the C function
name and the first paren.
Apart from the C# and GObject changes, this is just code motion. It
was verified by diffing the output of the generator before and after.
Store the camel-case name directly in the struct instead
of generating it on the fly in only the GObject bindings.
This is just code motion. Tested by verifying that the generator
output is identical.
Each action changes from a tuple like this:
("cat", (RString "content", [Pathname "path"], []), 4,
[ProtocolLimitWarning],
[InitISOFS, Always, TestOutput (
[["cat"; "/known-2"]], "abcdef\n")],
"list the contents of a file",
"[...]");
to a slightly longer but more readable struct:
{ defaults with
name = "cat";
style = RString "content", [Pathname "path"], [];
proc_nr = Some 4;
protocol_limit_warning = true;
tests = [
InitISOFS, Always, TestOutput (
[["cat"; "/known-2"]], "abcdef\n")
];
shortdesc = "list the contents of a file";
longdesc = "[...]" };
["defaults" is a struct which contains the defaults for every field,
allowing us to use the "{ defaults with ... }" syntax to just update
the fields we want to be different from the defaults.]
This is a mechanical change and there is no change to the output of
the generator. I checked the output before and after with diff to
verify this. There are no changes in the output apart from UUIDs
which are expected to change with each run.
Earlier versions of qemu contained a bug in the qcow2 code which
causes qemu to segfault when shutting down and flushing its internal
cache, and this can result in data loss.
The new API splits orderly close into a two-step process:
if (guestfs_shutdown (g) == -1) {
/* handle the error, eg. qemu error */
}
guestfs_close (g);
Note that the explicit shutdown step is only necessary in the case
where you have made changes to the disk image and want to handle write
errors. Read the documentation for further information.
This change also:
- deprecates guestfs_kill_subprocess
- turns guestfs_kill_subprocess into the same as guestfs_shutdown
- changes guestfish and other tools to call shutdown + close
where necessary (not for read-only tools)
- updates documentation
- updates examples
The order is now:
- remove the handle from the list of handles
- send close trace message
- sync and shutdown qemu
- run user close callback
- free temporary directory
- free memory
This commit ought to be no functional change.
On Linux, sync(2) does not actually issue a write barrier, thus it
doesn't force a flush of the underlying hardware write cache (or
qemu's disk cache in the virtual case).
This can be a problem, because libguestfs relies on running sync in
the appliance, followed by killing qemu (using SIGTERM).
In most cases, this is fine, because killing qemu with SIGTERM should
cause it to flush out the disk cache before it exits. However we have
found various bugs in qemu which cause qemu to crash while doing the
flush, leaving the data unwritten (see RHBZ#836913).
The solution is to issue fsync(2) to the block devices. This has a
write barrier, so it ensures that qemu writes out its cache long
before we get around to killing qemu.
Replace:
cp tests/guests/fedora.img test.img
with the longer but possibly more space-efficient equivalent:
qemu-img create -F raw -b tests/guests/fedora.img -f qcow2 test.qcow2
This returns the number of whole block devices added. It is usually
simpler to call this than to list the devices and count them, which
is what we do in some places in the current codebase.
On Debian, the Ruby C extensions library isn't '-lruby', it's
something like '-lruby1.8' or '-lruby-1.9.1' and these can even be
parallel-installed.
Fix detection so we use Ruby's own rbconfig.rb file to find the right
library to use.
MALLOC_PERTURB_ is a glibc feature which causes malloc to wipe memory
before and after it is used, allowing both use-after-free and
uninitialized reads to be detected with relatively little performance
penalty:
http://udrepper.livejournal.com/11429.html?nojs=1
Modify the ./run script so that it always sets this.
We were already using MALLOC_PERTURB_ in most tests. Since ./run is
now setting this, we can remove it from individual Makefiles. Most
TESTS_ENVIRONMENT will now simply look like this:
TESTS_ENVIRONMENT = $(top_builddir)/run --test
On RHEL 5, because _GNU_SOURCE was not defined, open_memstream was not
being declared, resulting in miscompilation and a segfault in the
trace code whenever open_memstream returned a pointer >= 0x80000000
(which would be truncated to a 32 bit int and then sign-extended).
When libguestfs calls 'error (g, "error")', gjs in F17 throws
error.message == "Error invoking Guestfs.test0rinterr: error"
In F18, error.message is simply the string "error".
Fix the test so it works for both cases.
This option, when added via
TESTS_ENVIRONMENT = [...] $(top_builddir)/run --test
allows us to run the tests and only print the full output (including
debugging etc) when the test fails.
The stdin and stdout of the qemu process are aliased to g->fd:
g->fd[0] = wfd[1];
g->fd[1] = rfd[0];
However if the child exits early, then child_cleanup closes g->fd[0],
g->fd[1], AND the code at the cleanup1 label closes wfd[1], rfd[0],
resulting in a double-close.
Avoid this case by setting wfd[1], rfd[0] to -1. In the cleanup1
label, only close wfd[1], rfd[0] if they are not -1, and add the same
for g->fd[0], g->fd[1].
Old KVM can't add /dev/null readonly. Treat /dev/null as a special
case.
We also fix a few tests where /dev/null was being used with
format=qcow2. This was always incorrect behaviour, but qemu appears
to tolerate it.
Old <libvirt.h> had a conflicting definition of virDomainGetXMLDesc:
fake_libvirt_xml.c:36: error: conflicting types for 'virDomainGetXMLDesc'
/usr/include/libvirt/libvirt.h:715: error: previous declaration of 'virDomainGetXMLDesc' was here
The difference is not material ('int' vs. 'unsigned int').
Avoid the error by not including <libvirt.h>.
The original fix for this in
commit 511c82df46 was not complete, in
that it did not fix the case of the old (pre '-m' option) parted.
This doesn't matter for Fedora, but it matters for RHEL 5 which has
this ancient parted.
If -lruby was not available, this used to define HAVE_LIBRUBY=0.
However this meant that the later test -n "$HAVE_LIBRUBY" would
be successful, whereas it should fail in this case.
RHEL 5-era autoconf did not define these, so define them manually
when they are missing.
Define builddir as '.' The scripts require this. It won't work
in the srcdir != builddir case, but we don't care about that for
RHEL 5.
This commit also moves the builddir / abs_srcdir variable setting
above the include of subdir-rules.mk, in case that include uses
these variables.
Useful script:
for f in $(find -name Makefile.am | xargs fgrep '$(abs_srcdir)' -l) ; do
if ! grep -q '^abs_srcdir' $f; then
echo missing in $f
fi
done
This returns the index of the device, eg. /dev/sdb => 1.
Or you can think of it as the order that the device was
added, or the index of the device in guestfs_list_devices.
Apparently e2fsprogs only knows that "/dev/sda" is a whole device, but
doesn't think that "/dev/vda" is. On switching the default device
over to virtio-scsi, that causes mke2fs without -F option to complain
and ask for an interactive prompt. Adding -F forces it to go ahead
anyway.
This caused several less-used APIs to break with virtio-scsi.
QEMU 1.0 was released at the end of 2011.
Remove all the cruft about detecting broken -machine type which
was only required for QEMU 0.15.
This also reverts commit 30ecbf3ec2.
Even on ARM you can pass -machine accel=kvm:tcg and qemu does the
right thing, so I'm not sure why we wanted to disable that.
These were used to select the default drive and network interface.
They both default to 'virtio'.
These were added back in the day when virtio was buggy, so that
packagers could revert to using ide/ne2k_pci to work around distro
bugs. However virtio has been stable in qemu for a very long time, so
it seems unlikely that any packager would need to use these, and in
any case it would be better to do this detection at runtime (cf. for
virtio-scsi).
The test used the external 'od' command to compare the output of
guestfish with what it's supposed to be. Unfortunately by default
this outputs groups of 2-byte words, with the words' endianness
affected by the current hardware endianness. For example:
x86-64$ echo -n ab | od
0000000 061141
0000002
ppc64$ echo -n ab | od
0000000 060542
0000002
By using 'od -b' instead we can output bytes instead of words, and
there is no endianness issue, and the output is clearer:
x86-64$ echo -n ab | od -b
0000000 141 142
0000002
ppc64$ echo -n ab | od -b
0000000 141 142
0000002
This is closer to the real meaning of "availability of btrfs", since
just having the btrfs tool doesn't help much if it's not supported by
the kernel too.
Since RHBZ was updated, the old python-bugzilla program broke. When
it was fixed, I found the (undocumented) ordering of the results from
the old tool was no longer true.
This commit adds a small Perl script to sort the output predictably.
Now bugs are sorted by both bug state and bug number, so the output
should be more stable than before.
Re-enable creation of the BUGS file in 'make dist'
(this reverts commit 91f3456244).
Also the BUGS file has been updated.
Previously the code would drop the first character in any sequence of
non-printing characters that occurred within a string.
Fix for commit e6f18c59d5.
Change guestfs___first_line_of_file so that if the file is empty this
returns an empty string instead of an error. This is consistent with
the behaviour of this function in the case where the file starts with
a \n character, where it would previously have returned an empty
string.
Change all callers so that they handle this case.
Then change the hostname parsing code so that it doesn't give up when
/etc/HOSTNAME is empty, but falls through to the next alternative, and
similarly for /etc/hostname.
Thanks Todd Mummert for finding and diagnosing this bug.
parse_release_file should be used to set fs->product_name, instead of
calling guestfs___first_line_of_file directory, although currently the
two are equivalent.
This is code motion and fixes
commit b648b1813f.
Missing package caused this error in compress-device-out:
libguestfs: error: compress_device_out: compression type bzip2 is not supported
Thanks Mohua Li.
Add the new API btrfs-fsck to check the btrfs filesystem.
Btrfs is currently under heavy development, and not suitable for
any uses other than benchmarking and review. But it'll be useful
in the near future.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
I used scsi_debug to create a 4k sector virtual disk:
modprobe scsi_debug dev_size_mb=128 sector_size=4096
I then used 'gdisk' to create lots of partitions, and used 'hexdump'
to examine what was written to disk.
* Move MOUNT_OPTIONS to #define at top.
* Move SIZE_MB to #define at top.
* Set PS1 in subshell (doesn't work - why?).
* Hide libguestfs errors from guestfs_mount_local_run.
This fixes commit 46b6766156.
Presently KVM is only applicable to x86 and x86-64 (although that will
change in future, and there are rumoured to be implementations for
some current non-x86 architectures). In any case having these options
breaks ARM, so disable them for non-x86 architectures at the moment.
For ARM Versatile ("versatilepb") architecture, memsize must be set to
256 (megabytes) else the appliance is unable to boot. Previously
setting LIBGUESTFS_MEMSIZE=256 was not allowed.
Versatile Express has larger limits.
'./configure --disable-probes' will disable these probes. Otherwise
they are autodetected as before.
The <sys/sdt.h> produces lots of errors when you try to compile with
these probes enabled under clang, so it is necessary to provide a way
to disable them for clang users.
This commit adds a tests/xml directory, and an LD_PRELOAD module which
can fake arbitrary libvirt XML from an external file (and is therefore
a much more flexible test than using the libvirt test:// driver
alone).
Also added is one regression test for:
https://bugzilla.redhat.com/show_bug.cgi?id=701814
Loading the given libvirt XML using Sys::Guestfs::Lib::open_guest
used to fail with the error:
format parameter is empty or contains disallowed characters at /home/rjones/d/libguestfs/perl/blib/lib/Sys/Guestfs/Lib.pm line 256.
Thanks to Tom Horsley for supplying the test data.
Old versions of libvirt allowed you to define disks like this:
<disk type='file' device='disk'>
<driver name='qemu'/>
...
Since the <driver> element does not have a 'type' attribute (which
defines the format), we are supposed to do autodetection, so the
format should be undefined.
However what actually happened was that the code in
Sys::Guestfs::Lib::open_guest received format as an empty string from
the xpath query, causing libguestfs to give an error.
If the xpath query returns the format as an empty string, undefine it.
This commit tidies up the code for the 'glob' command. It also makes
the command return an error if malloc fails (previously it would just
print a message but not return any error indication).
For a test case, see:
https://bugzilla.redhat.com/show_bug.cgi?id=803533
In guestfish, we use the gnulib closeout module which sets up an
atexit handler to detect if there were any errors on stdout when the
program exits. This is so we can fail correctly in out of disk space
cases like:
guestfish [...] > output
However the atexit handler just checks that there was any error on
stdout (using ferror). If a pipe command such as:
><fs> command_with_lots_of_output | head
ran at any time during the session, the error flag would be set
(because the underlying writes failed with EPIPE errors).
So the commit first adds a test for ferror (stdout) after each command
that we issue. This brings error handling closer to the point of
failure, and so is generally a good thing.
Secondly we reset the error flag after detecting and dealing with the
error, so that avoids the redundant 'guestfish: write error' message
produced by gnulib closeout.
Thirdly we add a --pipe-error flag which causes guestfish commands to
fail on pipe commands line the one above. The default is off for
backwards compatibility reasons.
Note that this support is optional: To enable it, install the
ocaml-gettext library from
http://forge.ocamlcore.org/projects/ocaml-gettext . If this library
is not installed, then configure detects this and inserts dummy
gettext functions that do nothing.
gettextize provides a local file called "gettext.h". Remove this and
use <libintl.h> from glibc headers instead.
Most of this change is mechanical: #include <libintl.h> in every C
file which uses any gettext function. But also we remove the
gettext.h file, and adjust the "_" macros.
Note that this effectively removes the ./configure --disable-nls
option, although we don't know if that ever worked.
Remove the logfiles configured by /etc/logrotate.d/*.
Omit the logfile of "samba" and "sssd" which we removed
them separately .
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Originally this state was intended so that in some way you could find
out if the appliance was running a command. However there was never a
thread-safe way to access the state of the handle, so in effect you
could never do anything useful safely with this information.
This commit completely removes the BUSY state.
The only visible change is to the guestfs_is_busy API. Previously you
could never call this safely from another thread. If you called it
from the same thread it would always return false (since the current
thread can't be running a libguestfs command at that point by
definition). Now it always returns false.
The gobject bindings generate a large number of header files, which pollute
/usr/include when installed. This patch moves them all into a guestfs-gobject/
subdirectory. guestfs-gobject.h remains in the same place.
This change also moves generated source files into src/, because it makes the
gobject directory a bit tidier.
Implement libguestfs events as GObject signals. Callback arguments are passed in
a boxed object.
Note that this patch fixes the length of the uint64_t array in the callback
arguments at 16, whereas it is actually arbitrary length. This is to make it
introspectable. There is currently no way to pass an arbitrary length array to a
callback, and have its type introspected.
virt-rescue prints errors such as:
rm: cannot remove `/proc': Is a directory
mkdir: cannot create directory `/proc': File exists
rm: cannot remove `/sys': Is a directory
mkdir: cannot create directory `/sys': File exists
People have reported these errors (which are nothing to worry about)
as bugs in the past, so avoid them where possible.
The two tarballs:
8252 bytes tests/data/filesanddirs-10M.tar.xz
41152 bytes tests/data/filesanddirs-100M.tar.xz
contain, as their names suggest, highly compressed files and
directories, with unpacked size respectively around 10M and 100M (the
precise size after unpacking depends on the filesystem).
These can be used for testing purposes.
Bind the easy parts of the 'btrfs' program.
The new APIs are:
btrfs-device-add: add devices to a btrfs filesystem
btrfs-device-delete: remove devices from a btrfs filesystem
btrfs-filesystem-sync: sync a btrfs filesystem
btrfs-filesystem-balance: balance a btrfs filesystem
btrfs-subvolume-create: create a btrfs snapshot
btrfs-subvolume-delete: delete a btrfs snapshot
btrfs-subvolume-list: list btrfs snapshots and subvolumes
btrfs-subvolume-set-default: set default btrfs subvolume
btrfs-subvolume-snapshot: create a writable btrfs snapshot
Flag the system for reconfiguration by touching /.unconfigured .
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Updated so this is disabled by default, since it requires
user intervention at next boot.
Remove user accounts from /etc/passwd, /etc/group,
/etc/shadow, and the home directory of the user,
except the root user.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Updated to use Augeas to delete accounts.
- Disable this by default, since it is very invasive.
All current operations are enabled by default.
Also:
* The POD description is split into a heading and the description body.
* An 'enabled_by_default' flag is added to the structure.
'del g' is a trap for the unwary. If the handle has any other
references, it does nothing (in fact, it can be actively dangerous if
the user was expecting the appliance to go away). In non-CPython it
can be delayed arbitrarily long.
Using 'g.close()' on the other hand is always safe.
Colon characters are not allowed in qemu/qemu-img filenames. There's
no way to quote them.
Comma characters CAN be used. However they must be quoted (by
doubling) when used in the '-o' option.
Fix general quoting problems in the external command.
The output looks like this:
F16x64:/dev/sda1 1048576 1024K ok
F16x64:/dev/sda2 2097152 2048K ok
F16x64:/dev/sda3 526385152 2048K ok
If the --uuid option is used, then UUIDs are shown instead of names.
The new APIs are:
get-e2attrs: List ext2 file attributes of a file.
set-e2attrs: Set or clear ext2 file attributes of a file.
get-e2generation: Get ext2 file generation of a file.
set-e2generation: Set ext2 file generation of a file.
These are implemented using the lsattr and chattr programs from
e2fsprogs.
Remove the .ssh directory of user "root" and any other
users who have a .ssh directory in their home directory.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Remove the bogus check_path function and move the functionality into
the two places where it was being used.
qemu -cdrom ,
works fine, I tested it.
Colon cannot be used in a block device filename anywhere, since the
qemu block driver interprets it as a prefix. There is no known way to
work around this problem. I checked this is true with kwolf.
Comma is fine in -drive options, provided it is escaped by doubling it.
On the new faster computer, Thread.yield wasn't yielding, so the
second thread would block the main test from proceeding (only when run
under valgrind however).
The first time the directory is checked out from git, there are no
*.pod files. In a parallel build, the first thread notices this and
runs the 'make -C .. update-po' rule, but the other threads also try
to run at the same time and fail.
This means renaming various temporary files (eg. fstab.tmp) that
several of the scripts were using so they don't conflict
(eg. fstab.tmp.$$).
Also the rule:
fedora-md1.img fedora-md2.img: guest-aux/make-fedora-img.pl
causes the script to run twice in parallel when using 'make -jN'.
Replace this with a stamp file so it runs once:
fedora-md1.img fedora-md2.img: stamp-fedora-md.img
stamp-fedora-md.img: guest-aux/make-fedora-img.pl
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.
guestfs-gobject.h (in the source directory) was not found.
In file included from <stdin>:4:0:
/path/to/libguestfs/gobject/guestfs-gobject.h:23:37: fatal error: guestfs-gobject-session.h: No such file or directory
compilation terminated.
Error while processing the source.
make[2]: *** [Guestfs-1.0.gir] Error 1
This greatly improves the usability of the generated gtk-doc.
Although there is a lot of churn in generator_gobject.ml, this is almost
exclusively code motion.
javah from old GNU classpath won't overwrite the target *.h file,
instead leaving the old one which results in a predictable build
failure. Delete the target so this won't happen.
virt_df-domains.o: In function `add_domain':
/home/feeliwood/Downloads/libguestfs-1.17.21/df/domains.c:274: undefined reference to `guestfs___for_each_disk'
guestfs___for_each_disk is only available when libvirt _and_ libxml2
dependencies are available at compile time.
Thanks Tho Huynh.
This function allows you to pass an explicit errno back to the
library. reply_with_error is redefined as a macro that calls
reply_with_error_errno with errno == 0.
Previously there was no programmatic way to tell if inspection APIs
were unavailable because they are not compiled in (because hivex isn't
around). This contrasts with daemon APIs where the availability is
covered by the guestfs_available API.
Change the inspection APIs so that when they are not available, the
last errno is set to ENOTSUP. (Note that ENOTSUP must be defined on
all POSIX platforms).
This allows programs to detect if they are using a version of
libguestfs that was compiled without support for inspection, without
having to parse error messages.
We were not using an unsigned type in the XDR. This doesn't
affect data integrity or the protocol, but it makes it more
complicated for the daemon to set these fields.
Add an API for doing what virt-sparsify was doing: freeing up free
space in a filesystem.
The current implementation is simple-minded: we create a file, fill it
with zeroes until we run out of space, then delete the file. However
the description leaves it open to do a better implementation, eg.
using sparsification support that is currently being worked on in ext4
and qemu.
The implementation also sends progress notifications, which is an
advantage over the old 'dd' method.
When a Windows guest doesn't have a HKLM\SYSTEM\MountedDevices node,
inspection fails. However inspection should not completely fail just
because we cannot get the drive letter mapping from a guest.
The presumption is that all file descriptors should be created with
the close-on-exec flag set. The only exception are file descriptors
that we want passed through to exec'd subprocesses (mainly pipes and
stdin/stdout/stderr).
For open calls, we pass O_CLOEXEC as an extra flag, eg:
fd = open ("foo", O_RDONLY|O_CLOEXEC);
This is a Linux-ism, but using a macro we can easily make it portable.
For sockets, similarly:
sock = socket (..., SOCK_STREAM|SOCK_CLOEXEC, ...);
For accepted sockets, we use the Linux accept4 system call which
allows flags to be supplied, but we use the Gnulib 'accept4' module to
make this portable.
For dup, dup2, we use the Linux dup3 system call, and the Gnulib
modules 'dup3' and 'cloexec'.
Previously a lot of daemon code used three variables (a string list,
'int size' and 'int alloc') to track growable strings buffers. This
commit implements a simple struct containing the same variables, but
using size_t instead of int:
struct stringsbuf {
char **argv;
size_t size;
size_t alloc;
};
Use it like this:
DECLARE_STRINGSBUF (ret);
//...
if (add_string (&ret, str) == -1)
return NULL;
//...
if (end_stringsbuf (&ret) == -1)
return NULL;
return ret.argv;
In two places, we were counting things in an array using an 'int'. In
theory, the int could overflow, so gcc determines this to be undefined
behaviour.
The fix is to use size_t or ssize_t instead.
When these tests were originally in the old regressions/ directory,
they used to refer to guestfish via the path '../fish/guestfish'.
Some of the tests were also called 'test-guestfish-*'.
Now that the tests have been moved into the fish/ directory, neither
of these things make sense. So change the relative path to
'./guestfish' and rename all 'test-guestfish-*' as 'test-*'.
If the parent process uses a pipe (or any fd, but pipes are a
particular problem), then the recovery process would hold open the
file descriptor(s) of the pipe, meaning that it could not be fully
closed in the parent. Because the recovery process doesn't use
exec(2), this wasn't avoidable even using FD_CLOEXEC.
Avoid this by closing all file descriptors when starting the recovery
process.
After discussion with Dan Berrange, he points out that it's also a
good idea to set signal handlers to the default after forking, so that
any signal handlers set up in the parent don't affect the child.
Error: UNINIT:
/builddir/build/BUILD/libguestfs-1.16.5/daemon/glob.c:32: var_decl: Declaring variable "buf" without initializer.
/builddir/build/BUILD/libguestfs-1.16.5/daemon/glob.c:36: uninit_use_in_call: Using uninitialized value "buf.gl_offs" when calling "rpl_glob".
/builddir/build/BUILD/libguestfs-1.16.5/gnulib/lib/glob.c:557: read_parm_fld: Reading a parameter field.
/builddir/build/BUILD/libguestfs-1.16.5/daemon/glob.c:36: uninit_use_in_call: Using uninitialized value "buf.gl_pathc" when calling "rpl_glob".
/builddir/build/BUILD/libguestfs-1.16.5/gnulib/lib/glob.c:557: read_parm_fld: Reading a parameter field.
The Coverity error is this (which I think is wrong):
Error: TAINTED_SCALAR:
/builddir/build/BUILD/libguestfs-1.16.5/daemon/inotify.c:211: tainted_data_argument: Calling function "read" taints argument "inotify_buf".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/inotify.c:232: var_assign_var: Assigning: "event" = "(struct inotify_event *)&inotify_buf[n]". Both are now tainted.
/builddir/build/BUILD/libguestfs-1.16.5/daemon/inotify.c:258: lower_bounds: Checking lower bounds of unsigned scalar "event->len" by "event->len > 0U".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/inotify.c:272: var_assign_var: Compound assignment involving tainted variable "16UL + event->len" to variable "n" taints "n".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/inotify.c:228: lower_bounds: Checking lower bounds of unsigned scalar "n" by "n < inotify_posn".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/inotify.c:281: tainted_data: Using tainted variable "n" as an index into an array "inotify_buf".
Adding a sanity check of event->len is prudent.
Error: RESOURCE_LEAK:
/builddir/build/BUILD/libguestfs-1.16.5/daemon/debug.c:469: open_fn: Calling opening function "open".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/debug.c:469: var_assign: Assigning: "fd" = handle returned from "open("/proc/sys/kernel/core_pattern", 1)".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/debug.c:474: noescape: Variable "fd" is not closed or saved in function "write".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/debug.c:476: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
This also includes some tidying up of the generated code.
Error: RESOURCE_LEAK:
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13254: alloc_fn: Calling allocation function "parse_string_list".
/builddir/build/BUILD/libguestfs-1.16.5/fish/fish.c:1386: alloc_fn: Storage is returned from allocation function "realloc".
/builddir/build/BUILD/libguestfs-1.16.5/fish/fish.c:1386: var_assign: Assigning: "argv_new" = "realloc(argv, 8UL * argv_len)".
/builddir/build/BUILD/libguestfs-1.16.5/fish/fish.c:1392: var_assign: Assigning: "argv" = "argv_new".
/builddir/build/BUILD/libguestfs-1.16.5/fish/fish.c:1396: return_alloc: Returning allocated memory "argv".
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13254: var_assign: Assigning: "devices" = storage returned from "parse_string_list(argv[i++])".
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13271: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13288: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13293: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13311: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13316: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13334: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13349: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
/builddir/build/BUILD/libguestfs-1.16.5/fish/cmds.c:13355: leaked_storage: Variable "devices" going out of scope leaks the storage it points to.
Error: RESOURCE_LEAK:
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:894: open_fn: Calling opening function "open".
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:894: var_assign: Assigning: "fd" = handle returned from "open(filename, 0)".
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:903: noescape: Variable "fd" is not closed or saved in function "read".
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:911: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:918: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
Error: RESOURCE_LEAK:
/builddir/build/BUILD/libguestfs-1.16.5/daemon/upload.c:225: open_fn: Calling opening function "open".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/upload.c:225: var_assign: Assigning: "fd" = handle returned from "open(filename, 0)".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/upload.c:233: noescape: Variable "fd" is not closed or saved in function "lseek".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/upload.c:235: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
Error: RESOURCE_LEAK:
/builddir/build/BUILD/libguestfs-1.16.5/daemon/9p.c:130: open_fn: Calling opening function "open".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/9p.c:130: var_assign: Assigning: "fd" = handle returned from "open(filename, 0)".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/9p.c:142: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
/builddir/build/BUILD/libguestfs-1.16.5/daemon/9p.c:149: noescape: Variable "fd" is not closed or saved in function "read".
/builddir/build/BUILD/libguestfs-1.16.5/daemon/9p.c:153: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
Error: RESOURCE_LEAK:
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:1125: open_fn: Calling opening function "open".
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:1125: var_assign: Assigning: "fd" = handle returned from "open(filename, 833, 438)".
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:1133: noescape: Variable "fd" is not closed or saved in function "xwrite".
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:1146: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
/builddir/build/BUILD/libguestfs-1.16.5/src/proto.c:1173: leaked_handle: Handle variable "fd" going out of scope leaks the handle.
Error: NULL_RETURNS:
/builddir/build/BUILD/libguestfs-1.16.5/examples/virt-dhcp-address.c:129: var_assigned: Assigning: "guest_distro" = null return value from "guestfs_inspect_get_distro".
/builddir/build/BUILD/libguestfs-1.16.5/examples/virt-dhcp-address.c:131: dereference: Dereferencing a pointer that might be null "guest_distro" when calling "__coverity_strcmp".
[...]
/builddir/build/BUILD/libguestfs-1.16.5/examples/virt-dhcp-address.c:126: var_assigned: Assigning: "guest_type" = null return value from "guestfs_inspect_get_type".
/builddir/build/BUILD/libguestfs-1.16.5/examples/virt-dhcp-address.c:128: dereference: Dereferencing a pointer that might be null "guest_type" when calling "__coverity_strcmp".
Error: DEADCODE:
/builddir/build/BUILD/libguestfs-1.16.5/fish/event-names.c:65: dead_error_condition: On this path, the condition "comma" cannot be true.
/builddir/build/BUILD/libguestfs-1.16.5/fish/event-names.c:57: const: After this line, the value of "comma" is equal to 0.
/builddir/build/BUILD/libguestfs-1.16.5/fish/event-names.c:57: assignment: Assigning: "comma" = "0".
/builddir/build/BUILD/libguestfs-1.16.5/fish/event-names.c:65: dead_error_line: Execution cannot reach this statement "fputc(44, fp);".
Remove dead code; however only comment it out since if we change the
preceeding code we may need this line again.
Error: DEADCODE:
/builddir/build/BUILD/libguestfs-1.16.5/src/listfs.c:107: dead_error_condition: On this path, the condition "lvs" cannot be true.
/builddir/build/BUILD/libguestfs-1.16.5/src/listfs.c:54: const: After this line, the value of "lvs" is equal to 0.
/builddir/build/BUILD/libguestfs-1.16.5/src/listfs.c:91: const: After this line, the value of "lvs" is equal to 0.
/builddir/build/BUILD/libguestfs-1.16.5/src/listfs.c:54: assignment: Assigning: "lvs" = "NULL".
/builddir/build/BUILD/libguestfs-1.16.5/src/listfs.c:91: new_values: Noticing condition "lvs == NULL".
/builddir/build/BUILD/libguestfs-1.16.5/src/listfs.c:107: dead_error_line: Execution cannot reach this statement "guestfs___free_string_list(...".
Error: CHECKED_RETURN:
/builddir/build/BUILD/libguestfs-1.16.5/java/com_redhat_et_libguestfs_GuestFS.c:9552: example_assign: Assigning: "r" = return value from "guestfs_internal_autosync(g)".
/builddir/build/BUILD/libguestfs-1.16.5/java/com_redhat_et_libguestfs_GuestFS.c:9555: example_checked: "r" has its value checked in "r == -1".
/builddir/build/BUILD/libguestfs-1.16.5/ocaml/guestfs_c_actions.c:5584: example_assign: Assigning: "r" = return value from "guestfs_internal_autosync(g)".
/builddir/build/BUILD/libguestfs-1.16.5/ocaml/guestfs_c_actions.c:5586: example_checked: "r" has its value checked in "r == -1".
/builddir/build/BUILD/libguestfs-1.16.5/perl/Guestfs.xs:5990: example_assign: Assigning: "r" = return value from "guestfs_internal_autosync(g)".
/builddir/build/BUILD/libguestfs-1.16.5/perl/Guestfs.xs:5991: example_checked: "r" has its value checked in "r == -1".
/builddir/build/BUILD/libguestfs-1.16.5/python/guestfs-py.c:13702: example_assign: Assigning: "r" = return value from "guestfs_internal_autosync(g)".
/builddir/build/BUILD/libguestfs-1.16.5/python/guestfs-py.c:13707: example_checked: "r" has its value checked in "r == -1".
/builddir/build/BUILD/libguestfs-1.16.5/ruby/ext/guestfs/_guestfs.c:16000: example_assign: Assigning: "r" = return value from "guestfs_internal_autosync(g)".
/builddir/build/BUILD/libguestfs-1.16.5/ruby/ext/guestfs/_guestfs.c:16001: example_checked: "r" has its value checked in "r == -1".
/builddir/build/BUILD/libguestfs-1.16.5/src/guestfs.c:191: check_return: Calling function "guestfs_internal_autosync" without checking return value (as is done elsewhere 5 out of 6 times).
/builddir/build/BUILD/libguestfs-1.16.5/src/guestfs.c:191: unchecked_value: No check of the return value of "guestfs_internal_autosync(g)".
Netpbm and icoutils (wrestool) have always been dependencies. Since
they are not always present, make these into optional dependencies
(which they were, sort of, before).
Also document these dependencies in the README file.
guestfs_set_autosync sets a flag which affects guestfs_close, and so
this call can be made at any time before the handle is closed, not
just in the config state.
Function guestfs_mkdtemp uses c++ keyword "template" as a parameter
name. In result, attempt to use guestfs.h header in c++ program
results in compile error.
This is just the 'kernel', 'initrd' and 'root' files, copied from one
machine to another, along with a 'README.fixed' file which is also
used for identification.
This allows the appliance to be copied from one machine to another,
making it easier for us to distribute a starter appliance for people
who cannot get febootstrap or appliance-building working.
--20cf303ea4a84a7a7c04ba331375
Content-Type: text/plain; charset=ISO-8859-1
Hello Richard,
I found what the problem was and the udev symlinks where not created. The
init script in the appliance uses a full paths for udevadm (/sbin/udevadm)
which in my case was wrong. In Arch Linux udevadm is hosted under /usr/bin,
not /sbin. When I fixed this, libguestfs was able to communicate with the
appliance VM.
Since PATH variable is defined and exported in init, there is no need in
using full paths for external programs. As far as I've seen this affects
all the git branches in libguestfs's repository. Please find attached a
patch for the master branch.
Nikos Skalkotos,
Athens, Greece
On 24 February 2012 10:25, Richard W.M. Jones <rjones@redhat.com> wrote:
> On Thu, Feb 23, 2012 at 08:50:12PM +0200, Nikos Skalkotos wrote:
> [...]
>
> I don't know specifically why it fails with ArchLinux, but the problem
> is caused by the /dev/virtio-ports/* symlinks not getting created by
> udev.
>
> /dev/vport0p1 exists:
>
> > crw------- 1 root root 252, 1 Feb 23 18:17 vport0p1
>
> but udev doesn't make the corresponding /dev/virtio-ports symlink:
>
> > /dev/virtio-ports/org.libguestfs.channel.0: No such file or directory
>
> The symlink is supposed to be created by this udev rule:
>
> /lib/udev/rules.d/50-udev-default.rules:KERNEL=="vport*",
> ATTR{name}=="?*", SYMLINK+="virtio-ports/$attr{name}"
>
> So I'd start by looking to see if that rule exists in the udev rules
> that Arch is using. Secondly if it does exist, is the corresponding
> *.rules being copied into the appliance? (Check appliance/supermin.d/
> hostfiles)
>
> In an old Ubuntu that has udev that predates having this rule, we add
> the following patch:
>
>
> http://libguestfs.org/download/binaries/ubuntu1004-packages/0002-ubuntu-10.04-Use-dev-vport0p1.patch
>
> Rich.
>
> --
> Richard Jones, Virtualization Group, Red Hat
> http://people.redhat.com/~rjones
> virt-p2v converts physical machines to virtual machines. Boot with a
> live CD or over the network (PXE) and turn machines into Xen guests.
> http://et.redhat.com/~rjones/virt-p2v
>
Hello Richard,<br><br>I found what the problem was and the udev symlinks where not created. The init script in the appliance uses a full paths for udevadm (/sbin/udevadm) which in my case was wrong. In Arch Linux udevadm is hosted under /usr/bin, not /sbin. When I fixed this, libguestfs was able to communicate with the appliance VM.<br>
<br>Since PATH variable is defined and exported in init, there is no need in using full paths for external programs. As far as I've seen this affects all the git branches in libguestfs's repository. Please find attached a patch for the master branch.<br>
<br>Nikos Skalkotos,<br>Athens, Greece<br><br><div class="gmail_quote">On 24 February 2012 10:25, Richard W.M. Jones <span dir="ltr"><<a href="mailto:rjones@redhat.com">rjones@redhat.com</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
On Thu, Feb 23, 2012 at 08:50:12PM +0200, Nikos Skalkotos wrote:<br>
[...]<br>
<br>
I don't know specifically why it fails with ArchLinux, but the problem<br>
is caused by the /dev/virtio-ports/* symlinks not getting created by<br>
udev.<br>
<br>
/dev/vport0p1 exists:<br>
<div class="im"><br>
> crw------- 1 root root 252, 1 Feb 23 18:17 vport0p1<br>
<br>
</div>but udev doesn't make the corresponding /dev/virtio-ports symlink:<br>
<div class="im"><br>
> /dev/virtio-ports/org.libguestfs.channel.0: No such file or directory<br>
<br>
</div>The symlink is supposed to be created by this udev rule:<br>
<br>
/lib/udev/rules.d/50-udev-default.rules:KERNEL=="vport*", ATTR{name}=="?*", SYMLINK+="virtio-ports/$attr{name}"<br>
<br>
So I'd start by looking to see if that rule exists in the udev rules<br>
that Arch is using. Secondly if it does exist, is the corresponding<br>
*.rules being copied into the appliance? (Check appliance/supermin.d/<br>
hostfiles)<br>
<br>
In an old Ubuntu that has udev that predates having this rule, we add<br>
the following patch:<br>
<br>
<a href="http://libguestfs.org/download/binaries/ubuntu1004-packages/0002-ubuntu-10.04-Use-dev-vport0p1.patch" target="_blank">http://libguestfs.org/download/binaries/ubuntu1004-packages/0002-ubuntu-10.04-Use-dev-vport0p1.patch</a><br>
<br>
Rich.<br>
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Richard Jones, Virtualization Group, Red Hat <a href="http://people.redhat.com/%7Erjones%0Avirt-p2v" target="_blank">http://people.redhat.com/~rjones<br>
virt-p2v</a> converts physical machines to virtual machines. Boot with a<br>
live CD or over the network (PXE) and turn machines into Xen guests.<br>
<a href="http://et.redhat.com/%7Erjones/virt-p2v" target="_blank">http://et.redhat.com/~rjones/virt-p2v</a><br>
</font></span></blockquote></div><br>
Add a flag "ConfigOnly" to make sure that some non-daemon functions
should be called only at CONFIG state (RHBZ#796520).
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Appliance building can be called from multiple processes, but this is
only safe if each process holds a lock on the 'checksum' file.
However threads within a process are not excluded by a file lock, and
so this strategy completely failed for a multithreaded program calling
guestfs_launch in parallel.
Since it makes no sense for threads in a single program to race each
other to try to create the appliance, add a lock around appliance
building.
This serialises building the appliance, but the rest of guestfs_launch
(eg. starting up qemu) can run in parallel.
rhbz727178.sh exists, but historically was never added to the list of
tests, so it doesn't run. Create a special variable for this test and
ensure it is added to EXTRA_DIST.
Test was originally added in
commit d7356a2801.
In the case where the caller attempts to mount the "hidden"
appliance root device (eg. /dev/vdb if /dev/vda is the only
normal block device added), we were calling reply_with_error
but not actually returning immediately, resulting in protocol
desynchronization.
This commit fixes this obvious mistake.
For stubs of functions that had arguments, code did this:
static void
mount_stub (XDR *xdr_in)
{
int r;
struct guestfs_mount_args args;
if (optargs_bitmask != 0) {
//...
goto done;
}
// possibly other tests here
memset (&args, 0, sizeof args);
[...]
done:
xdr_free ((xdrproc_t) xdr_guestfs_mount_args, (char *) &args);
return;
}
This caused xdr_free to be called on uninitialized 'args' struct,
causing a segfault.
The fix is to add another label, so the code looks like:
static void
mount_stub (XDR *xdr_in)
{
int r;
struct guestfs_mount_args args;
if (optargs_bitmask != 0) {
//...
goto done_no_free;
}
// possibly other tests here
memset (&args, 0, sizeof args);
[...]
done:
xdr_free ((xdrproc_t) xdr_guestfs_mount_args, (char *) &args);
done_no_free:
return;
}
This fixes commit 330fbea5b2
and commit 0344248af5.
erase the filesystem signatures on each device, then erase
the partitions, avoid to list all the partitions here.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Check /sbin, /usr/sbin, /bin and /usr/bin to ensure that neither the
directory nor the parent (eg. /usr) is a symlink, and use this
directory for the virtual directory used to store guestfsd in the
appliance.
This file shouldn't be included in the tarball as it contains
hard-coded paths from the maintainer's machine which cannot be
relevant to the target machine.
Callers are supposed to use the availability API to check for
functions that may not be available in particular builds of
libguestfs. If they don't do this, currently they tend to get obscure
error messages, eg:
libguestfs: error: zerofree: /dev/vda1: zerofree: No such file or directory
This commit changes the error message to explain what callers ought to
be doing instead:
libguestfs: error: zerofree: feature 'zerofree' is not available in this
build of libguestfs. Read 'AVAILABILITY' in the guestfs(3) man page for
how to check for the availability of features.
This patch makes the stubs check for availability. The stub code
changes to:
static void
zerofree_stub (XDR *xdr_in)
{
[...]
/* The caller should have checked before calling this. */
if (! optgroup_zerofree_available ()) {
reply_with_error ("feature '%s' is not available in this\n"
"build of libguestfs. Read 'AVAILABILITY' in the guestfs(3) man page for\n"
"how to check for the availability of features.",
"zerofree");
goto done;
}
[...]
Set LD_LIBRARY_PATH so we link against the just-built library, not
the installed library.
Use OCAMLCFLAGS instead of OCAMLOPTFLAGS where appropriate.
Remove unnecessary -cclib option for bytecode linking (gcc is not used
in this case).
This fixes commit eb68a31413.
The error messages would be prefixed by the path to the
alternate binary, eg:
../fish/guestfish: invalid escape sequence in string (starting at offset 0)
Fix this by normalizing the messages further.
RConstOptString cannot return an error in the C api. This makes it a special
case for the GObject api, as all other return types have a corresponding GError
**err argument to return an error. This change removes this special case, and
includes the possibility of an error return in the API. An error is indicated by
setting *err to a non-NULL value.
This change is in preparation for adding a close api. An attempt to call any
api, even RConstOptString, on a closed handle must return an error.
Change 'ifconfig' and 'netstat' commands to use 'ip' instead.
'iproute' was already included in the appliance, so this reduces the
size of the appliance accordingly.
This commit tidies up the code by splitting out the path
element-searching code into a separate function.
Valgrind found that 'closedir' frees the 'struct dirent *', which
wasn't immediately obvious. So now we do the 'closedir' after all
operations which touch 'd->d_name'.
This commit allows you to run the daemon under valgrind. You have to
enable it at configure time:
./configure --enable-valgrind-daemon
This should *not* be done for production builds.
When this feature is enabled, valgrind is added to the appliance and
the daemon is run under valgrind. Log messages from valgrind are
passed back over a virtio-serial channel into a file called
'valgrind.log.$PID' in the top build directory.
Running 'make check', 'make extra-tests' etc causes many
valgrind.log.* files to be created which must be examined by hand.
I noticed some uses of ${srcdir} in shell scripts.
That is almost always better written as $srcdir.
The patch below converts most such variable references.
Here are the few remaining candidates:
$ git grep -i -E '\$\{[a-zA-Z_0-9]+\}'|grep -v Makefile.in.in
configure.ac: JAR_INSTALL_DIR=\${prefix}/share/java
configure.ac: JNI_INSTALL_DIR=\${libdir}
debian/rules: for TEST in ${DEBIAN_SKIP_TEST}; do \
debian/rules:# mv $${mod} $$(dirname $${mod})/libguestfsmod.so; \
java/Makefile.am:libguestfs_jar_DATA = libguestfs-${VERSION}.jar
java/Makefile.am:libguestfs-${VERSION}.jar: $(libguestfs_jar_class_files)
perl/lib/Sys/Guestfs/Lib.pm: "-f", '${Package} ${Version} ${Architecture} ${Status}\n',
perl/typemap: croak (\"${Package}::$func_name(): called on a closed handle\");
perl/typemap: croak (\"${Package}::$func_name(): $var is not a blessed HV reference\");
tests/data/Makefile.am: echo "$${i}abcdefghijklmnopqrstuvwxyz"; \
We could change all of those, too, except the ones in configure.ac
and Makefile.am, since they refer to Make variables. Even those
should be changed, but to use the preferred Makefile notation:
$(prefix), $(libdir), $(VERSION).
>From a86770ecd45666232a94d76c8725c8f9b1c76e3a Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering@redhat.com>
Date: Mon, 23 Jan 2012 11:15:12 +0100
Subject: [PATCH libguestfs] maint: use $var notation rather than ${var} when
possible
The only case to avoid in a shell script is when the byte after the
"}" is word-constituent, and concatenating it would thus change the
name of the variable.
These changes were induced by running this command:
git grep -l -i -E '\$\{(srcdir|md)' \
|xargs perl -pi -e 's/\$\{(srcdir|md)\}($|\w)/\$$1$2/gi'
The "g" was needed because there was one line with two instances.
The "i" is to handle ${SRCDIR}. The ($|\w) ensures that concatenating
whatever follows the "}" won't change semantics.
* gobject/run-bindtests: Use "$srcdir", not "${srcdir}".
* haskell/run-bindtests: Likewise.
* java/run-bindtests: Likewise.
* ocaml/run-bindtests: Likewise.
* perl/run-bindtests: Likewise.
* python/run-bindtests: Likewise.
* ruby/run-bindtests: Likewise.
* tests/guests/guest-aux/make-debian-img.sh: Likewise, but $SRCDIR.
* tests/guests/guest-aux/make-ubuntu-img.sh: Likewise.
* tests/guests/guest-aux/make-windows-img.sh: Likewise.
* tests/md/test-mdadm.sh: Likewise, but $md.
Hi Rich,
I realized a day or two late that my suggestion was not quite right.
Here's the fix:
>From 5294c21cf07c4ec2f094182ba9f32696f3de2751 Mon Sep 17 00:00:00 2001
From: Jim Meyering <meyering@redhat.com>
Date: Mon, 23 Jan 2012 10:01:40 +0100
Subject: [PATCH libguestfs] build: don't use automake-internal variable
* ocaml/Makefile.am (mlguestfs.cma): Use the documented variable form,
$(libguestfsocaml_a_OBJECTS), not the $(am_libguestfsocaml_a_OBJECTS)
that I suggested for commit 1.15.16-17-g8b9eaec.
Note that this change disables compiling and running the haskell bindtests. The
haskell bindings do not implement optargs, and adding optargs to test0 causes
that method not to be bound in the haskell bindings. This prevents the haskell
bindtests from compiling. These should be re-enabled when optargs are
implemented.
We can make a good guess at camel case names for most APIs. For example,
add_drive_opts can be automatically transformed to AddDriveOpts. However, other
apis don't produce a satisfactory name when transformed automatically. For
example, we would want md_create to produce MDCreate rather than MdCreate.
This change adds a CamelName flag which allows a camel case name to be specified
explicitly when the automatic transformation isn't satisfactory.
We were passing the name of the last argument directly to va_start. However, if
the last argument is a BufferIn it expands to 2 arguments, the latter of which
is <name>_size.
When you call close on any block device, udev kicks off a rule which
runs blkid to reexamine the device. We need to wait for this rule to
finish running since it holds the device open and can cause other
operations to fail, notably BLKRRPART.
However since the OCaml bindings use guestfs_safe_strdup and
guestfs_safe_memdup we need to export those two (in the private
functions section) from <guestfs.h>.
Previously, passing NULL optargs to an _argv call resulted in a segfault. This
change causes NULL optargs to be interpreted as no optargs, and to be correctly
handled accordingly.
Currently any api which takes a FileIn or FileOut parameter is implicitly
cancellable. This change make cancellable an explicit flag in anticipation of it
being added to other apis.
Note that a Cancellable function must be able to return an error, which means it
can't return RConstOptString.
We can make a good guess at camel case names for most APIs. For example,
add_drive_opts can be automatically transformed to AddDriveOpts. However, other
apis don't produce a satisfactory name when transformed automatically. For
example, we would want md_create to produce MDCreate rather than MdCreate.
This change adds a CamelName flag which allows a camel case name to be specified
explicitly when the automatic transformation isn't satisfactory.
Since we implement the new api e2fsck, just change the
internal of e2fsck_f to use e2fsck now.
v1->v2: use optargs_bitmask
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
m: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Add a new api e2fsck with two options:
correct: same as '-p' option of e2fsck
forceall: same as '-y' option of e2fsck
Thanks for Rich's idea.
v1->v2: use optargs_bitmask
v2->v3: change the optargs_bitmask check
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Tweak the error message "e2fsck -f" and "e2fsck -fy".
Indicate the user to use the correct and/or forceall options.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
If you have a lot of guests, running extra tests takes a lot of time,
but (arguably) you're not really doing any more detailed testing.
Thus pick up to 5 guests at random to test. A different random
selection is done on each run, ensuring that data-driven errors (eg in
inspection) will still be found eventually.
Change the test for duplicate drives so that you're allowed to
add /dev/null multiple times. This corresponds to traditional
usage.
This amends commit be47b66c30.
1. Change the g->path to restore a absolute path instead of the mixed.
2. Check that if the adding drive is duplicated with the added drive.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ:
- Make sure abs_path is NULL before it is assigned, so freeing it
will work along the error path.
- Fix the test which added /dev/null multiple times.
Previously, optional arguments had the same type as regular arguments, but were
constrained by various runtime tests to be only Bool, Int, Int64 or String. This
change makes the type of optional arguments stronger by giving them their own
type.
A convenience function, optargs_to_args is defined to convert optargs in the few
places where they are genuinely treated identically to mandatory arguments.
It also allows for future changes to optional arguments which do not affect
mandatory arguments.
RWMJ:
- removed redundant parens
- readded the check for > 64 optargs, but changed it to > 63
- changed the new function to args_of_optargs
This map was originally included just for the java bindings, but is generally
useful to any binding which uses camel case by requirement or convention.
This function does 'rm -rf <dir>' for temporary directories, safely
working if '<dir>' contains shell meta-characters.
Replace existing code for removing directories with this.
extra-tests is depending on tests-guests.
We need guests/*.img when doing extra-tests.
/work/git/libguestfs/run valgrind --log-file=/work/git/libguestfs/tests/extra/valgrind.log --leak-check=full --error-exitcode=119 --suppressions=/work/git/libguestfs/tests/extra/suppressions ../../cat/virt-cat -a ../guests/fedora.img /etc/fstab >/dev/null
../guests/fedora.img: No such file or directory
make[1]: *** [test-tools-internal] Error 1
make[1]: Leaving directory `/work/git/libguestfs/tests/extra'
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
v1->v2: fix a typo pointed by Matt
Optimizations by reducing the STREQ operations and do some
code cleanup.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
RWMJ: Whitespace changes.
CVE-2011-4127 is a serious qemu & kernel privilege escalation bug
found by Paolo Bonzini.
http://seclists.org/oss-sec/2011/q4/536
An untrusted guest kernel is able to issue special SG_IO ioctls on
virtio devices which qemu passes through to the host kernel without
filtering or sanitizing. These ioctls allow raw sectors from the
underlying host device to be read and written. Significantly, neither
qemu nor the host kernel checks that the range of sectors is within
the partition / LV assigned to the guest. For example, if the guest
is assigned host partition /dev/sda3, it would be able to read or
write any part of /dev/sda including other partitions and the boot
sector. Exploits through LVs passed to the guest are also possible,
with some limitations. File-backed virtual block devices are not
vulnerable. Non-virtio block devices are not vulnerable.
This patch mitigates the problem by disabling the SG_IO ioctl
passthrough in qemu. Thus if libguestfs is examining an untrusted
guest and the libguestfs appliance/daemon is compromised (eg. by
executing guest commands, or through some other compromise), then the
compromised appliance will not be able to issue the above SG_IO ioctls
and exploit the host.
Note that this is just mitigation for libguestfs. Users will still
want to fully update their host kernel, qemu/KVM and libvirt, in order
to prevent other (non-libguestfs) routes to compromise.
The following versions of libguestfs (will/have) this patch applied.
libguestfs >= 1.15.13
libguestfs >= 1.14.8
libguestfs >= 1.12.11
libguestfs >= 1.10.12
libguestfs >= 1.8.16
Earlier versions may be vulnerable unless a downstream packager has
applied this patch.
Cc: Hilko Bengen <bengen@hilluzination.de>
Add 'event', 'list-events' and 'delete-event' commands so that event
handlers can be registered, listed and deleted in guestfish. The
event handler is a shell script snippet or host command.
Cc: Pádraig Brady <P@draigBrady.com>
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)
In particular this stops a problem with guestmount where if the -i
option half-recognizes the guest OS, it would mount the disk (and
fail, giving only warnings), leaving the mountpoint unusable.
(Thanks Pádraig Brady)
This also adds comprehensive tests for utimens on regular files,
directories (RHBZ#761451), named pipes (RHBZ#761460), symbolic links,
block and char devices.
Note that there is a small change in the (previously undefined)
semantics of this call: It now sets the time on a symbolic link
itself, not on what the symbolic link points to.
You don't need to open the file O_WRONLY in order to call futimens on
the file descriptor. Opening it O_WRONLY fails for directories.
Therefore open O_RDONLY instead.
RHEL5 shoult not support '-p', '-i' and '-o export' options.
But we just split it according to the '-p' option.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Augeas 0.10 depends on libxml2, so this is now required in the
appliance (in fact, it was already present).
However this exposed two bugs:
(1) In libguestfs we use a home-brewed recipe for Augeas flags,
resulting in this error:
/usr/include/augeas.h:24:25: fatal error: libxml/tree.h: No such file or directory
(2) Augeas's own augeas.pc didn't include the libxml2 flags, so
it was broken. This requires a patch to Augeas 0.10, see:
https://www.redhat.com/archives/augeas-devel/2011-December/msg00008.html
Change to using pkg-config to detect Augeas. It is still an optional
library.
A NEW API blkid.
It can print the device attributes.
Use it after list-devices, we can list ower devices and the attributes
of each device.
Use it like:
blkid <device>
It's should be a usefull function.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
This patch fixes inspection when fstab contains devices md devices
specified as /dev/mdN. The appliance creates these devices without reference to
the guest's mdadm.conf so, for e.g. /dev/md0 in the guest will often be created
as /dev/md127 in the appliance. With this patch, we match the uuids of detected
md devices against uuids specified in mdadm.conf, and map them appropriately
when we encounter them in fstab.
Just add the -f option to mkfs.xfs to make sure we can
make a xfs filesystem when the device already has a
filesystem on it.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
Since some modules (`getopt', for example) may copy files
into the build directory, `top_builddir/lib' is needed as well as
`top_srcdir/lib'. -- GNU Gnulib manual, section 2.2 Initial import
This adds the virt-sparsify --debug-gc option which causes
virt-sparsify to call Gc.compact before exiting, allowing
GC and memory problems to be tested.
Add an extratest which runs virt-sparsify under valgrind.
This adds the virt-resize --debug-gc option which causes
virt-resize to call Gc.compact before exiting, allowing
GC and memory problems to be tested.
Add an extratest which runs virt-resize under valgrind.
Select non-comment labels using an augeas path to return the correct
nodes in the first instance, rather than applying a regular expression
to all results.
There is no functional change to the code.
RWMJ:
- Rebased to current HEAD.
- Move variable decls to top of function.
- Some whitespace changes.
- Remove check for errors from snprintf since it can never happen.
The error message now looks like this:
guestfish: '/dev/vda6' could not be mounted. Did you mean one of these?
/dev/sda1 (ext4)
/dev/vg_f15x32/lv_root (ext4)
/dev/vg_f15x32/lv_swap (swap)
Note that '/dev/sda1' has been canonicalized.
This API is used to stop a md device.
When we want to move a device to another md array, we should
stop the md device which contained this device first.
Signed-off-by: Wanlong Gao <gaowanlong@cn.fujitsu.com>
This change renames the following 2 apis:
* mdadm_create -> md_create
* mdadm_detail -> md_detail
This is more consistent with list_md_devices, and removes a reference to an
implementation detail from the api.
This change involves rewriting make-fedora-img.sh in perl. This allows the
flexibility to write mdadm.conf containing whichever uuids where randomly
generated when the md devices were created.
Previously we bound the 'tune2fs -l' command so that we could list out
the tunables of an ext2/3/4 filesystem. Also commands like
set_e2label and set_e2uuid used tune2fs.
This commit binds many of the tunables that can be set using tune2fs.
The coverage is not complete, but we can add more later because this
uses optional parameters so the call is extensible without breaking
ABI. The current change gives us enough for using libguestfs within
OpenStack.
The bitmask was being constructed backwards(!)
As a result, any test which tested optional arguments didn't work.
There are very few such tests and they happened not to be affected by
this.
In the libguestfs live case we need to be careful not to modify the
real /etc/lvm/lvm.conf file (when setting the filter rule).
When the daemon starts, make a complete copy of /etc/lvm in a
temporary directory, and adjust LVM_SYSTEM_DIR to point to the copy.
All changes are made in the temporary copy.
Although this doesn't matter for the ordinary (appliance) case, it
matters for the libguestfs live case. In that case it could cause the
guest to be exploited by a tmp/symlink attack.
xmlstarlet is good, but not available in Red Hat Enterprise Linux.
Build a simple but sane XPath query parser into virt-inspector
directly so that we don't need any external tools.
The major/minor fields only exist in Python >= 2.7. This works for at
least Python 2.5 and 2.6.
Thanks to Hilko Bengen.
This updates commit 646142f513.
These fixes allow libguestfs bindings to work with Python 3 (tested
with Python 3.2)
You can select which Python you compile against by doing:
PYTHON=python ./configure && make && make check
or:
PYTHON=python3 ./configure && make && make check
If the user set PYTHON when configuring, this variable is not passed
through to the tests, so it is possible the tests will fail because
they are testing the wrong version of python. By passing $PYTHON
through to the tests we ensure that we test against the same version
of python that we configured with.
The manual for sys.version warns that you should not parse the string,
which we were doing. It would have failed on python "2.10" or
similar.
Do it correctly using the sys.version_info struct instead.
This is a short (10-15 min) talk that I give to introduce the main
features of libguestfs. The "slides" are in the form of a complete
self-contained HTML page with a handful images that can be easily
distributed before the talk.
When given an invalid debug command, libguestfs responds with the
error message:
libguestfs: error: debug: use 'debug help' to list the supported commands
However this command does not work, as debug requires two
arguments. This change updates the message to prompt the user to use
'debug help 0'.
size_after on Debian was 10136K. The host filesystem was ext3 (vs
ext4 for Fedora where we did the original testing). Probably indirect
blocks (vs more efficient extents on ext4) causes the difference,
although I did not look at it in detail.
This optional argument controls how <readonly/> disks in
the libvirt XML are handled. It can be set to one of:
"write" (default) - add them R/W
"read" - add them R/O
"error" - throw an error if asked to add them R/W
"ignore" - ignore these disks
I have added limited regression tests for this feature. However
libvirt's test:/// XML does not allow you to specify that a domain
starts off shut down, so we cannot fully test this. Instead I tested
it by hand.
This changes the private function guestfs___for_each_disk so that the
<readonly/> flag on libvirt disks is detected and passed through to
the callback function.
Use a Gaussian distribution for random numbers so that smaller numbers
are chosen more frequently.
This also exposes a bug in this test which only happens when small
numbers are chosen:
If the test thread starts up quickly, it can get to
guestfs_user_cancel before the guestfs_upload command has happened in
the main thread. This causes the cancel to be ignored (correctly) and
we end up in the second loop writing data until the disk runs out of
space. Fix this by repeatedly asserting the cancel flag in the second
loop.
Commit 2910413850 caused Windows 7
resizes to break with the 0xc0000225 boot error.
Change the --align-first auto (default) option so that it is more
conservative about when it moves the first partition. In particular
it doesn't move it if it's already aligned (as it is for Win7), nor if
there is more than one partition (also Win7).
Tested with: Windows XP, 2003, 7, Ubuntu 10.10 and RHEL 5.
Also we only permit MBR (DOS) and GPT partition tables. In theory
previously we allowed other partition table types, but it is unlikely
that it would have worked in reality.
The four new APIs:
guestfs_copy_device_to_device,
guestfs_copy_device_to_file,
guestfs_copy_file_to_device, and
guestfs_copy_file_to_file
let you copy from a source to a destination, between files and
devices, optionally allowing source and destination offsets and size
to be specified.
Add a dependency so these tools are rebuilt from scratch if the
Guestfs API changes. This prevents the error:
"[...] make inconsistent assumptions over interface Guestfs".
This commit includes the generated changes to .depend files.
This adds an extra column containing this information, looking
like this:
Name Type VFS Label MBR Size Parent
/dev/sda1 filesystem ntfs - - 6.0G -
/dev/sda1 partition - - 07 6.0G /dev/sda
/dev/sda device - - - 6.0G -
In particular you can use this to tell if a partition is an extended
partition, because the field will contain '05' or '0f'.
This converts a partition device name (eg. /dev/sda1) to a partition
number (eg. 1). This is useful in conjunction with the parted APIs
that mostly take a disk device + partnum.
This reverts commit 025dba7f80.
If build and source directories are the same, you get this error:
make[3]: Leaving directory `/home/rjones/d/libguestfs/generator'
make[2]: Circular guestfs_protocol.c <- guestfs_protocol.c dependency dropped.
make[2]: Circular guestfs_protocol.h <- guestfs_protocol.h dependency dropped.
rm -f guestfs_protocol.h
ln guestfs_protocol.h
ln: accessing `guestfs_protocol.h': No such file or directory
The old code mixed the business of planning the layout of the target
partitions with the creation of the target partitions. The
replacement code separates these into two tasks: firstly we create a
new 'partitions' list with the target layout, secondly this directly
drives the creation of the partitions.
As part of this change I have *removed* the old code that was supposed
to handle extended/logical MBR partitions. It simply didn't work, and
didn't have any hope of working, and there is a separate bug open to
fix it.
Fixed one bug in the nl PO file:
msgid "security: cached appliance %s is not owned by UID %d"
msgstr "beveiliging: UID %d is geen eigenaar van toepassing %s in cache"
should be:
msgid "security: cached appliance %s is not owned by UID %d"
msgstr "beveiliging: UID %2$d is geen eigenaar van toepassing %1$s in cache"
It is possible for the pcre library to return a variable number of captures for
a single regular expression. e.g.:
^/dev/(cciss/c\d+d\d+)(?:p(\d+))?$
This will return either 1 or 2 captures depending on whether the device has a
partition suffix. The current match wrappers don't allow for this, and require
that a predictable number of matches are returned.
This change updates match, match1, match2, and match3 to ignore the specific
number of matches returned. Instead, any returned captures are assigned to the
given arguments, and any remaining arguments are set to NULL.
We currently use a heuristic to guess how drive names we find
referenced in the guest map to drive names in the appliance. If this
heuristic fails it can cause inspection to fail.
This change adds a new 'name' option to add_drive_opts, which allows
the user to explicitly pass the name of a drive to libguestfs if it is
known. This change also updates the fstab-parsing inspection code to
use this information if it is available.
This is a NFC on its own, but provides a place-holder for drive metadata which
can be used after launch.
Fixes by RWMJ:
- Fix the tests: this requires a new internal function 'debug-drives'
that dumps out the g->drives information so it can be checked in
two of the tests. Previously these tests used 'debug-cmdline'.
- Test file existence / use_cache_off in the add_drive_opts function,
not when launching qemu in the child process.
- Call free along error paths.
- Add comments.
The regular expression matching disk name assumed that there was only a single
letter suffix. This change handles a naming scheme for any number of disks.
For now, only detect the tool, but support from reading
installed package could be added later ( may require either
a package of pkgsrc, or a smaller tool to read the db ).
While testing on Fedora 16, I noticed that Linux will mount the disk
without trouble by using automatically ufstype=old, and yet do not
let us read the directories. So we should start directly with usftype=ufs2,
and if it fail, try 44bsd, as used for netbsd and openbsd ( as seen on
http://wiki-static.aydogan.net/How_to_mount_FFS_partition_under_Linux ).
Ths logo is present on Lucid Lynx, but may not always be installed, like
in case of a minimal installation ( or server ). While Edubuntu and Kubuntu
have different logo, I didn't check where it would be placed.
Without this, configure will always enable erlang, no matter what
argument are passed. Now, we can disable it, even if configure still
need the erlang compiler for some obscure reason.
This switches virt-sysprep to use guestmount instead of guestfish.
This makes the script a little bit easier to modify for sysadmins.
This commit also adds:
- dhcp-client-state
- dhcp-server-state
- logfiles
- random-seed
- smolt-uuid
- yum-uuid
SELinux relabelling, and a section on security in the manual page.
Killing it is always a bad idea, because the qemu subprocess will be
left hanging around. The best thing is to send the exit command.
This also makes the change to virt-sysprep.
These APIs let you copy compressed files or devices out from the disk
image.
Compression is useful for large images which are mostly zeroes. We
cannot currently do sparseness detection, and compression gives us a
form of zero detection for free.
Example usage:
$ guestfish --ro -a /dev/vg_pin/F16x64 -i \
compress-out gzip /etc/passwd /tmp/passwd.gz
$ file -z /tmp/passwd.gz
/tmp/passwd.gz: ASCII text (gzip compressed data, was "passwd", from
Unix, last modified: Sun Aug 28 14:40:46 2011)
Originally this was disabled to work around RHBZ#502058. That bug was
never officially fixed, but it may have fixed itself.
In the meantime, KVM has broken ordinary PIC support (RHBZ#723822).
Since APIC is the most common way that regular Linux and Windows
guests run it makes sense to remove this hack.
This means "the device that holds /etc/fstab", so map it correctly.
This fixes support for ttylinux and also some other guests that use
/dev/root instead of a real device name.
This also improves the documentation for this call, pointing out
several pitfalls in using it.
This unfortunately breaks existing callers that might use
guestfs_grub_install without checking for this new group.
We could sometimes hit the 120 second timeout, eg. if the appliance
needed to be rebuilt and the machine was very slow and/or under heavy
I/O load. 10 minutes should be enough for any reasonable situation.
The guestfs_ll command currently lists files in the appliance if you
prefix filenames with "/..". However this is a bug, not a feature,
and we should not be testing it.
We partitioned the disk, and then tried to create a PV on the whole
disk. LVM gave the error:
Device /dev/vda not found (or ignored by filtering).
It is unclear how this bug persisted for so long. It might be due to
a change in LVM.
errno is pretty much random on the error path back from guestfs_launch
so there is nothing useful to be gained by checking it.
We could do with a better way to detect if appliance launch failed.
There is another case where downloads of small files could fail if the
library side (writer) fails. In this case the library would send back
a cancellation, but it would be received after the daemon had finished
sending the whole file (because the file is small enough). The daemon
would reenter the main loop and immediately get an unexpected cancel
message, causing the daemon to die.
This commit also makes test-cancellation-download-librarycancels.sh
more robust. We use Monte-Carlo testing with a range of file sizes.
Small file sizes should trigger the error case.
The default event handler in libguestfs was simply writing all debug
output directly to stderr. However if the output contains
non-printable characters such as terminal control codes then these
would also be sent directly.
With newer SeaBIOS there is a lame attempt to implement a splash
screen using terminal control codes, thus when libguestfs tries to
display debugging output it would cause the screen to clear and debug
output to be lost.
This commit causes all non-printing characters to be escaped.
(\n and \r characters from the appliance are treated somewhat
specially).
Furthermore, instead of using write(2), use buffered stderr calls.
This function was used to print the qemu and
febootstrap-supermin-helper command lines.
Unfortunately in the qemu case it was used incorrectly: it called the
internal debug function (ie. event API callback) from the forked qemu
subprocess, which meant that higher level event callbacks might have
been invoked from the child process.
To fix this, convert the qemu case into a new function called
print_qemu_command line which just prints the command line directly to
stderr. This is called after stderr has been redirected into the pipe
to the main process. Thus the qemu command line will be marshalled
into the event API along with other qemu and appliance output.
After fixing this, only one use of guestfs___print_timestamped_argv
remained, for printing the febootstrap-supermin-helper command line.
This is converted to a local function print_febootstrap_command_line.
Also print_febootstrap_command_line is now called before we fork
febootstrap-supermin-helper, so that messages no longer overlap.
./run can now be run in a separate build directory. Since some files
needed in the image checks are found in the source tree (but not the
build tree), the source tree location is passed to make-*-img.sh via
an environment variable.
This code modifies zero, zero-device, is-zero, is-zero-device.
zero and zero-device are modified so that if the blocks of the device
already contain zeroes, then we don't write zeroes. The reason for
this is to avoid unnecessarily making the underlying storage
non-sparse or (in the qcow2 case) growing it.
is-zero and is-zero-device are modified so that zero detection is
faster. This is a nice side effect of making the first change.
Since avoiding unnecessary zeroing involves reading the blocks before
writing them, whereas before we just blindly wrote, this can be
slower. As you can see from the tests below, in the case where the
disk is sparse, it actually turns out to be faster, because we avoid
allocating the underlying blocks.
However in the case where the disk is non-sparse and full of existing
data, it is much slower. There might be a case for an API flag to
adjust whether or not we perform the zero check. I did not add this
flag because it is unlikely that the caller would have enough
information to be able to set the flag correctly.
(Elapsed time in seconds)
Format Test case Before After
Raw Sparse 16.4 5.3
Preallocated zero 17.0 18.8
Preallocated random 16.0 41.3
Qcow2 preallocation=off 18.7 5.6
preallocation=metadata 17.4 5.8
The current code uses a fixed block size of 4K for reading and
writing. I also tried the same tests with a block size of 64K but it
didn't make any significant difference.
(Thanks to Federico Simoncelli for suggesting this change)
Previously any disk that had /autoexec.bat or /boot.ini or /ntldr
would be picked up as a candidate for a Windows root disk. If further
checking could not find any systemroot (eg. /windows) then this would
result in complete failure of inspection.
In particular, this got confused by Hp_recovery partitions which have
/autoexec.bat, but don't have a systemroot in one of the usual places
(they have /MiniNT instead).
What we do now is to properly investigate all possible systemroot
places before deciding this is a Windows systemroot, so the subsequent
failure cannot occur.
(Thanks to lorimar for reporting this bug).
This avoids conflicts with the globally installed libguestfs
appliance, or lets us build in multiple local directories at the same
time without conflicts.
Combine the two Gnulib instances together.
Add checks from old daemon/configure.ac into configure.ac.
Fix daemon/Makefile.am so it is like a normal subdirectory
Makefile.am.
Because we are now using the replacement strerror_r function from
Gnulib (instead of the one from glibc directly), this requires a small
change to src/guestfs.c.
We weren't acquiring the GC lock around some allocations, resulting in
segfaults when an event callback ran at the same time as a main thread
allocation or garbage collection.
In particular this fixes a noticable crash in guestfs-browser.
A change to ExtUtils::CBuilder in Perl 5.14 causes CCFLAGS to
completely replace, rather than appending, the C flags.
The unfortunate consequence of this is that vital flags such as
-D_FILE_OFFSET_BITS=64 are missing. For 32 bit code, this means you
get binary-incompatible code that completely fails to load.
For further analysis see:
http://www.nntp.perl.org/group/perl.perl5.porters/2011/04/msg171535.html
This commit changes CCFLAGS so that it appends to the existing
$Config{ccflags} instead of replacing it. On earlier versions of Perl
this means we get two copies of the flags, which is unfortunate but
should be safe.
785603 NEW https://bugzilla.redhat.com/show_bug.cgi?id=785603
copy-out "No such file or directory"
790837 NEW https://bugzilla.redhat.com/show_bug.cgi?id=790837
Use of atexit to clean up handles is wrong in multithreaded programs
803650 NEW https://bugzilla.redhat.com/show_bug.cgi?id=803650
inspect-is-live return false when inspection results should be true
803657 NEW https://bugzilla.redhat.com/show_bug.cgi?id=803657
[RFE] inspect-is-netinst : support more distributions
805417 NEW https://bugzilla.redhat.com/show_bug.cgi?id=805417
RFE: support inspection of installation ISOs of WinVista, Win7, Win2008 & Win2008r2
806179 NEW https://bugzilla.redhat.com/show_bug.cgi?id=806179
RFE: support inspection of icon for more Windows guests
808193 NEW https://bugzilla.redhat.com/show_bug.cgi?id=808193
g.launch() crashes on RHEL 5
812970 NEW https://bugzilla.redhat.com/show_bug.cgi?id=812970
virt-rescue cannot set ><rescue> prompt, on Ubuntu 12.04
813290 NEW https://bugzilla.redhat.com/show_bug.cgi?id=813290
mdadm (md-detail) test occasionally fails with 'md device /dev/md125 does not appear to be active.'
815149 NEW https://bugzilla.redhat.com/show_bug.cgi?id=815149
virt-alignment-scan gives error "part_list: could not parse row from output of parted print command: /dev/vda:4194304B:virtblk:512:512:msdos:Virtio Block Device"
819086 NEW https://bugzilla.redhat.com/show_bug.cgi?id=819086
look for qemu-kvm on /usr/libexec
822538 NEW https://bugzilla.redhat.com/show_bug.cgi?id=822538
libguestfs tools hang on qcow2 encrypted disks
824021 NEW https://bugzilla.redhat.com/show_bug.cgi?id=824021
inspection cannot recognize guest which uses btrfs subvolumes for root
824782 NEW https://bugzilla.redhat.com/show_bug.cgi?id=824782
virt-resize cannot resize PowerPC guests
832602 NEW https://bugzilla.redhat.com/show_bug.cgi?id=832602
"error in chunked encoding" when trying to extract (tar-out) a truncated ISO image
833362 NEW https://bugzilla.redhat.com/show_bug.cgi?id=833362
virt-make-fs test fails on ppc64 because filesystem block size is 64k
835622 NEW https://bugzilla.redhat.com/show_bug.cgi?id=835622
RFE: virt-sparsify should be able to sparsify onto a thin-provisioned LV
837941 NEW https://bugzilla.redhat.com/show_bug.cgi?id=837941
Data loss when writing to qcow2-format disk files
838081 NEW https://bugzilla.redhat.com/show_bug.cgi?id=838081
ocaml/t/guestfs_500_parallel_mount_local crashes in caml_thread_reinitialize
845234 NEW https://bugzilla.redhat.com/show_bug.cgi?id=845234
RFE: virt-ls on Windows guest doesn't support drive letters
848464 NEW https://bugzilla.redhat.com/show_bug.cgi?id=848464
gobject javascript bindings cannot use 64 bit integers
857763 NEW https://bugzilla.redhat.com/show_bug.cgi?id=857763
libguestfs 'file-architecture' returns 'ARM' for arm binaries
863696 NEW https://bugzilla.redhat.com/show_bug.cgi?id=863696
[F18] libguestfs fails to mount a disk image file(in this case qcow2) as 'root'
864871 NEW https://bugzilla.redhat.com/show_bug.cgi?id=864871
libvirt error: could not create appliance through libvirt: unable to set security context 'unconfined_u:object_r:svirt_image_t:s0:cXXX,cYYY' on '.../console.sock'
866994 NEW https://bugzilla.redhat.com/show_bug.cgi?id=866994
tgz-out causes memory leak in guestfsd
869179 NEW https://bugzilla.redhat.com/show_bug.cgi?id=869179
libguestfs fail to startup on latest rhel7 tree
880806 NEW https://bugzilla.redhat.com/show_bug.cgi?id=880806
virt-df with two -a options displays incorrect disk image name
887826 NEW https://bugzilla.redhat.com/show_bug.cgi?id=887826
RFE: virt-sparsify cannot handle multi-disk virtual machines (eg. if they use LVM, MD, LDM, etc)
889089 NEW https://bugzilla.redhat.com/show_bug.cgi?id=889089
Block device /dev/sd* didn't translate to /dev/vd*
890027 NEW https://bugzilla.redhat.com/show_bug.cgi?id=890027
virt-sysprep: Setting the hostname on Fedora 18 is stupidly over-complicated
891909 NEW https://bugzilla.redhat.com/show_bug.cgi?id=891909
Can't make install with non-standard prefix as non-root
892271 NEW https://bugzilla.redhat.com/show_bug.cgi?id=892271
virt-format fail to format the same disk more than twice with lvm enable
892272 NEW https://bugzilla.redhat.com/show_bug.cgi?id=892272
Specify QEMU interface emulation will break libguestfs
892275 NEW https://bugzilla.redhat.com/show_bug.cgi?id=892275
checksums-out fail to compute the checksums of all regular files in directory
892291 NEW https://bugzilla.redhat.com/show_bug.cgi?id=892291
guestmount: link() incorrectly returns ENOENT, when it should be EXDEV
892834 NEW https://bugzilla.redhat.com/show_bug.cgi?id=892834
style="font-size:14.01373005px;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:condensed;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#000000;fill-opacity:1;stroke:none;font-family:Liberation Sans Narrow;-inkscape-font-specification:Sans"
x="371.08105"
y="372.44043"
id="text3054"
sodipodi:linespacing="125%"><tspan
sodipodi:role="line"
id="tspan3056"
x="371.08105"
y="372.44043">raw</tspan></text>
<text
xml:space="preserve"
style="font-size:14.01373005px;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:condensed;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#000000;fill-opacity:1;stroke:none;font-family:Liberation Sans Narrow;-inkscape-font-specification:Sans"
x="421.1301"
y="372.44043"
id="text3058"
sodipodi:linespacing="125%"><tspan
sodipodi:role="line"
id="tspan3060"
x="421.1301"
y="372.44043">qcow2</tspan></text>
<text
xml:space="preserve"
style="font-size:14.01373005px;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:condensed;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#000000;fill-opacity:1;stroke:none;font-family:Liberation Sans Narrow;-inkscape-font-specification:Sans"
style="font-size:14.01373005px;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:condensed;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#000000;fill-opacity:1;stroke:none;font-family:Liberation Sans Narrow;-inkscape-font-specification:Sans"
x="391.51471"
y="315.46762"
id="text3082"
sodipodi:linespacing="125%"><tspan
sodipodi:role="line"
id="tspan3084"
x="391.51471"
y="315.46762">ext4</tspan></text>
<text
xml:space="preserve"
style="font-size:14.01373005px;font-style:normal;font-variant:normal;font-weight:normal;font-stretch:condensed;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#000000;fill-opacity:1;stroke:none;font-family:Liberation Sans Narrow;-inkscape-font-specification:Sans"
style="font-size:16px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
style="font-size:16px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
style="font-size:16.01569176px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
xml:space="preserve"><tspan
y="93.218353"
x="126.08962"
id="tspan3912"
sodipodi:role="line">guestfish</tspan></text>
</g>
<g
id="g4006"
transform="translate(12.032604,19.818407)">
<rect
style="fill:#cc0000;fill-opacity:1;stroke:none"
id="rect3997"
width="145.80684"
height="31.851011"
x="523.87335"
y="229.11598"
ry="9.2014036"/>
<text
xml:space="preserve"
style="font-size:16.01569176px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
style="font-size:16px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
style="font-size:15.58510303px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
style="font-size:16.01569176px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
x="150.15482"
y="449.24185"
id="text4031"
sodipodi:linespacing="125%"><tspan
sodipodi:role="line"
id="tspan4033"
x="150.15482"
y="449.24185">virt-inspector</tspan></text>
</g>
<g
transform="translate(75.026826,46.714815)"
id="g4043">
<rect
ry="9.2014036"
y="428.71567"
x="134.58322"
height="31.851011"
width="145.80684"
id="rect4045"
style="fill:#cc0000;fill-opacity:1;stroke:none"/>
<text
sodipodi:linespacing="125%"
id="text4047"
y="449.24185"
x="150.15482"
style="font-size:16.01569176px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
style="font-size:16.01569176px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
x="281.09787"
y="529.22327"
id="text4055"
sodipodi:linespacing="125%"><tspan
sodipodi:role="line"
id="tspan4057"
x="281.09787"
y="529.22327">virt-df</tspan></text>
</g>
<g
id="g4108"
transform="translate(-7.0780024,24.773008)">
<rect
ry="9.2014036"
y="411.72845"
x="496.97693"
height="31.851011"
width="145.80684"
id="rect4061"
style="fill:#cc0000;fill-opacity:1;stroke:none"/>
<text
sodipodi:linespacing="125%"
id="text4063"
y="432.25464"
x="512.54852"
style="font-size:16.01569176px;font-style:normal;font-variant:normal;font-weight:bold;font-stretch:normal;text-align:start;line-height:125%;letter-spacing:0px;word-spacing:0px;writing-mode:lr-tb;text-anchor:start;fill:#ffffff;fill-opacity:1;stroke:none;font-family:Liberation Sans;-inkscape-font-specification:Liberation Sans Bold"
Some files were not shown because too many files have changed in this diff
Show More
Reference in New Issue
Block a user
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
