January 2017 - Libguestfs - Libguestfs List Archives

by Richard W.M. Jones

With the release of RHEL 7.3 happening today, fixing 88 reported bugs in libguestfs and virt-v2v, and countless other ones, it's time to look ahead to RHEL 7.4. Again we intend to rebase libguestfs, virt-v2v and virt-p2v: https://bugzilla.redhat.com/show_bug.cgi?id=1359086 1359086 NEW Rebase libguestfs in RHEL 7.4 To help with testing I have set up a preview repository here: http://people.redhat.com/~rjones/libguestfs-RHEL-7.4-preview/ As always: (1) THE PACKAGES ARE NOT SUPPORTED! Do not contact Red Hat Support about these packages. However you are welcome to file bugs or send comments to <libguestfs(a)redhat.com>. The link for filing bugs is: https://bugzilla.redhat.com/enter_bug.cgi?product=Red%20Hat%20Enterprise%... (2) These packages should work on top of RHEL 7.3 or CentOS 7.3. (3) To install the packages you should create a yum repo file like the one below, and then use yum commands as normal: $ cat /etc/yum.repos.d/libguestfs-RHEL-7.4-preview.repo [libguestfs-RHEL-7.4-preview] name=libguestfs RHEL 7.4 preview baseurl=http://people.redhat.com/~rjones/libguestfs-RHEL-7.4-preview/ enabled=1 gpgcheck=0 Notable changes in packaging: - The 'virt-v2v' package now only includes virt-v2v, not virt-p2v. - To build your own virt-p2v bootable USB key, install 'virt-p2v-maker'. - virt-v2v installs Windows virtio drivers using a new technique which tries to make the minimal possible changes to the Windows registry. This may create exciting new bugs, so I'm particularly interested in further testing in this area. - There is currently a bogus dependency on 'kernel-rt' which shouldn't be there ... under investigation. Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com virt-builder quickly builds VMs from scratch http://libguestfs.org/virt-builder.1.html

7 years, 9 months

2
2
0 / 0

ntfs-3g data deduplication support in guestmount for creating file based back-ups on visualization platform

by Jelle de Jong

Hello everybody, I am using libguestfs-tools version 1.28.1 in a bunch of bash scripts to make back-ups of the files on a visualisation platform. I been trying to get read access to ntfs volumes with data deduplication working. The ntfs-3g guys developed some support and I have this working when using the commands our back-up servers. However when using the guestmount tool the read suport foor dedub is gone. See this email for all the commands I use (working mount -t ntfs and non-working guestmount commands). https://sourceforge.net/p/ntfs-3g/mailman/message/35595392/ How can I see what version of ntfs-3g is used in guestmount and if it is using the /usr/lib/x86_64-linux-gnu/ntfs-3g/ntfs-plugin-80000013.so plugin that is on my back-up server? Thank you in advance! Kind regards, Jelle de Jong

7 years, 10 months

2
5
0 / 0

[PATCH v6 0/3] Import directly from OVA tar archive if possible

by Tomáš Golembiovský

v6: - just rebase v5: - rebase, patches 1,3,5 were merged - 1/3: we still need to discuss whether to detect compressed discs - 2/3: - renamed argument noempty to keep_empty - tests were not run - 3/3: - using JSON module to generate JSON (as suggested by Pino) - all the other comments raised by Pino v4: - rebase to more recent master - 1/6: using just "quote" instead of "Filename.quote" - 2/6: reformated block of code according to Richards suggestion - 4/6: added tests for nsplit v3: Addressed Pino's comments, namely: - input_ova.ml - untar takes list of paths - renamed untar_partial to untar_metadata - replaced uggly regex with nsplit - tests - test changes are part of the main commit - renamed test-data/guestfs-hashsums.sh to test-data/test-utils.sh - renamed qemu_version to qemu_is_version and moved it to test-data/test-utils.sh - normalize paths in expect files v2: - rewritten the tar invocations, the output processing is now done in OcaML rather than with a shell code; it turned out to be easier and more readable than I have feared. - added QEMU version check - addressed Pino's comments - changed tests; the expected result is now based on the QEMU used during testing This series is related to the problem of inefficient import of OVA files. The needed enhancements of QEMU were merged into the codebase and should be available in QEMU 2.8. From there we can use 'size' and 'offset' options in raw driver to tell QEMU to use only subset of a file as an image. The patch set is more or less complete. The only outstanding issue is the missing detection of sparse files inside tar. But this can be done in a separate patch. As pointed out before I didn't find a way how to detect that by using the tar tool only and would probably require use of some external library. Tomáš Golembiovský (3): v2v: ova: don't detect compressed disks, read the OVF instead mllib: modify nsplit to take optional noempty and count arguments v2v: ova: don't extract files from OVA if it's not needed mllib/common_utils.ml | 13 ++- mllib/common_utils.mli | 12 +- mllib/common_utils_tests.ml | 30 +++++ test-data/test-utils.sh | 20 ++++ v2v/Makefile.am | 2 + v2v/input_ova.ml | 198 +++++++++++++++++++++++++++++--- v2v/test-v2v-i-ova-formats.sh | 5 +- v2v/test-v2v-i-ova-gz.ovf | 2 +- v2v/test-v2v-i-ova-subfolders.expected2 | 18 +++ v2v/test-v2v-i-ova-subfolders.sh | 13 ++- v2v/test-v2v-i-ova-tar.expected | 18 +++ v2v/test-v2v-i-ova-tar.expected2 | 18 +++ v2v/test-v2v-i-ova-tar.ovf | 138 ++++++++++++++++++++++ v2v/test-v2v-i-ova-tar.sh | 72 ++++++++++++ v2v/test-v2v-i-ova-two-disks.expected2 | 19 +++ v2v/test-v2v-i-ova-two-disks.sh | 13 ++- 16 files changed, 561 insertions(+), 30 deletions(-) create mode 100644 v2v/test-v2v-i-ova-subfolders.expected2 create mode 100644 v2v/test-v2v-i-ova-tar.expected create mode 100644 v2v/test-v2v-i-ova-tar.expected2 create mode 100644 v2v/test-v2v-i-ova-tar.ovf create mode 100755 v2v/test-v2v-i-ova-tar.sh create mode 100644 v2v/test-v2v-i-ova-two-disks.expected2 -- 2.11.0

7 years, 10 months

3
11
0 / 0

[nbdkit PATCH] tls: Fix build on RHEL 5

by Eric Blake

Don't assume that errno will be implicitly available. Signed-off-by: Eric Blake <eblake(a)redhat.com> --- src/tls.c | 1 + 1 file changed, 1 insertion(+) diff --git a/src/tls.c b/src/tls.c index 5d380e3..0e9ea11 100644 --- a/src/tls.c +++ b/src/tls.c @@ -38,6 +38,7 @@ #include <string.h> #include <unistd.h> #include <assert.h> +#include <errno.h> #include <pthread.h> -- 2.9.3

7 years, 11 months

2
1
0 / 0

[nbdkit PATCH 0/2] fix pod and other errors in recent patches

by Eric Blake

7 years, 11 months

2
3
0 / 0

[nbdkit PATCH v3 0/3] bind .zero to Perl

by Eric Blake

Requires patch 1 and 2 of the python series: https://www.redhat.com/archives/libguestfs/2017-January/msg00126.html This is the perl implementation along the same lines. We still haven't decided if patch 1 of the python series should change OCaml to report errno as reliable or not, but perhaps we can commit that patch as-is now and then touch things up further when we actually get set_error and zero working in OCaml. Eric Blake (3): perl: Allow use of modules perl: Expose nbdkit_set_error to perl script perl: Support zero callback plugins/perl/example.pl | 16 +++++++++ plugins/perl/nbdkit-perl-plugin.pod | 56 +++++++++++++++++++++++++++++--- plugins/perl/perl.c | 65 ++++++++++++++++++++++++++++++++++++- 3 files changed, 131 insertions(+), 6 deletions(-) -- 2.9.3

7 years, 11 months

2
4
0 / 0

[PATCH nbdkit] Add support for socket activation.

by Richard W.M. Jones

Socket activation (aka systemd socket activation) is a simple protocol that lets you pass in an opened, listening socket to a server. Supporting socket activation allows you to use a modern superserver to serve infrequent NBD requests without needing nbdkit to be running the whole time. Although the protocol was invented by systemd, it has been implemented in a few other places, and the protocol is almost trivially simple. This implementation is based on the one in libvirt. Thanks: Dan Berrange, libvirt team --- .gitignore | 1 + docs/nbdkit.pod | 67 +++++++++++- src/main.c | 111 ++++++++++++++++++++ tests/Makefile.am | 10 +- tests/test-socket-activation.c | 226 +++++++++++++++++++++++++++++++++++++++++ 5 files changed, 411 insertions(+), 4 deletions(-) create mode 100644 tests/test-socket-activation.c diff --git a/.gitignore b/.gitignore index 4a51dea..f8918a7 100644 --- a/.gitignore +++ b/.gitignore @@ -50,6 +50,7 @@ Makefile.in /tests/test-perl /tests/test-python /tests/test-ruby +/tests/test-socket-activation /tests/test-streaming /tests/test-xz /test-driver diff --git a/docs/nbdkit.pod b/docs/nbdkit.pod index 2b11eed..ad02ec3 100644 --- a/docs/nbdkit.pod +++ b/docs/nbdkit.pod @@ -192,9 +192,11 @@ This option implies I<--foreground>. Don't fork. Handle a single NBD connection on stdin/stdout. After stdin closes, the server exits. -You can use this option to run nbdkit from inetd, systemd or similar +You can use this option to run nbdkit from inetd or similar superservers; or just for testing; or if you want to run nbdkit in a -non-conventional way. +non-conventional way. Note that if you want to run nbdkit from +systemd, then it may be better to use L</SOCKET ACTIVATION> instead of +this option. This option implies I<--foreground>. @@ -263,6 +265,48 @@ To list all the options supported by a plugin, do: nbdkit --help file +=head1 SOCKET ACTIVATION + +nbdkit supports socket activation (sometimes called systemd socket +activation). This is a simple protocol where instead of nbdkit itself +opening the listening socket(s), the parent process (typically +systemd) passes in pre-opened file descriptors. Socket activation +lets you serve infrequent NBD requests using a superserver without +needing nbdkit to be running the whole time. + +Socket activation is triggered when both the C<LISTEN_FDS> and +C<LISTEN_PID> environment variables are set. In this mode using +I<-i>, I<-p>, I<--run>, I<-s> or I<-U> flags on the command line is +illegal and will cause an error. Also in this mode nbdkit does not +fork into the background (ie. I<-f> is implied). + +=head2 Using socket activation with systemd + +To use nbdkit with socket activation from systemd, create a unit file +ending in C<.socket> (eg. C</etc/systemd/system/nbdkit.socket>) +containing: + + [Unit] + Description=NBDKit Network Block Device server + + [Socket] + ListenStream=0.0.0.0:10809 + + [Install] + WantedBy=sockets.target + +There are various formats for the C<ListenStream> key. See +L<systemd.socket(5)> for more information. + +Also create a service unit (eg. C</etc/systemd/system/nbdkit.service>) +containing: + + [Service] + ExecStart=/usr/sbin/nbdkit file file=/path/to/serve + +For more information on systemd and socket activation, see +L<http://0pointer.de/blog/projects/socket-activation.html> + =head1 CAPTIVE NBDKIT You can run nbdkit as a "captive process", using the I<--run> option. @@ -382,8 +426,23 @@ This signal is ignored. =back +=head1 ENVIRONMENT VARIABLES + +=over 4 + +=item C<LISTEN_FDS> + +=item C<LISTEN_PID> + +If present in the environment when nbdkit starts up, these trigger +L</SOCKET ACTIVATION>. + +=back + =head1 SEE ALSO +Other nbdkit manual pages: + L<nbdkit-plugin(3)>, L<nbdkit-curl-plugin(1)>, L<nbdkit-example1-plugin(1)>, @@ -398,6 +457,10 @@ L<nbdkit-python-plugin(3)>, L<nbdkit-vddk-plugin(1)>. L<nbdkit-xz-plugin(1)>. +Other manual pages of interest: + +L<systemd.socket(5)>. + =head1 AUTHORS Richard W.M. Jones diff --git a/src/main.c b/src/main.c index 9453cce..e935da2 100644 --- a/src/main.c +++ b/src/main.c @@ -54,6 +54,8 @@ #include "nbdkit-plugin.h" #include "internal.h" +#define FIRST_SOCKET_ACTIVATION_FD 3 /* defined by systemd ABI */ + static char *make_random_fifo (void); static void open_plugin_so (const char *filename); static void start_serving (void); @@ -64,6 +66,7 @@ static void write_pidfile (void); static void fork_into_background (void); static uid_t parseuser (const char *); static gid_t parsegroup (const char *); +static unsigned int get_socket_activation (void); const char *exportname; /* -e */ int foreground; /* -f */ @@ -77,6 +80,7 @@ int listen_stdin; /* -s */ char *unixsocket; /* -U */ const char *user, *group; /* -u & -g */ int verbose; /* -v */ +unsigned int socket_activation /* $LISTEN_FDS and $LISTEN_PID set */; volatile int quit; @@ -157,6 +161,9 @@ main (int argc, char *argv[]) tls_init (); + /* Returns 0 if no socket activation, or the number of FDs. */ + socket_activation = get_socket_activation (); + for (;;) { c = getopt_long (argc, argv, short_options, long_options, &option_index); if (c == -1) @@ -172,6 +179,11 @@ main (int argc, char *argv[]) dump_plugin = 1; } else if (strcmp (long_options[option_index].name, "run") == 0) { + if (socket_activation) { + fprintf (stderr, "%s: cannot use socket activation with --run flag\n", + program_name); + exit (EXIT_FAILURE); + } run = optarg; foreground = 1; } @@ -196,6 +208,11 @@ main (int argc, char *argv[]) break; case 'i': + if (socket_activation) { + fprintf (stderr, "%s: cannot use socket activation with -i flag\n", + program_name); + exit (EXIT_FAILURE); + } ipaddr = optarg; break; @@ -214,6 +231,11 @@ main (int argc, char *argv[]) break; case 'p': + if (socket_activation) { + fprintf (stderr, "%s: cannot use socket activation with -p flag\n", + program_name); + exit (EXIT_FAILURE); + } port = optarg; break; @@ -222,10 +244,20 @@ main (int argc, char *argv[]) break; case 's': + if (socket_activation) { + fprintf (stderr, "%s: cannot use socket activation with -s flag\n", + program_name); + exit (EXIT_FAILURE); + } listen_stdin = 1; break; case 'U': + if (socket_activation) { + fprintf (stderr, "%s: cannot use socket activation with -U flag\n", + program_name); + exit (EXIT_FAILURE); + } if (strcmp (optarg, "-") == 0) unixsocket = make_random_fifo (); else @@ -454,6 +486,7 @@ start_serving (void) { int *socks; size_t nr_socks; + size_t i; /* If the user has mixed up -p/-U/-s options, then give an error. * @@ -470,6 +503,27 @@ start_serving (void) set_up_signals (); + /* Socket activation -- we are handling connections on pre-opened + * file descriptors [FIRST_SOCKET_ACTIVATION_FD .. + * FIRST_SOCKET_ACTIVATION_FD+nr_socks-1]. + */ + if (socket_activation) { + nr_socks = socket_activation; + debug ("using socket activation, nr_socks = %zu", nr_socks); + socks = malloc (sizeof (int) * nr_socks); + if (socks == NULL) { + perror ("malloc"); + exit (EXIT_FAILURE); + } + for (i = 0; i < nr_socks; ++i) + socks[i] = FIRST_SOCKET_ACTIVATION_FD + i; + change_user (); + write_pidfile (); + accept_incoming_connections (socks, nr_socks); + free_listening_sockets (socks, nr_socks); /* also closes them */ + return; + } + /* Handling a single connection on stdin/stdout. */ if (listen_stdin) { change_user (); @@ -752,3 +806,60 @@ parsegroup (const char *id) return grp->gr_gid; } + +/* Returns 0 if no socket activation, or the number of FDs. + * See also virGetListenFDs in libvirt.org:src/util/virutil.c + */ +static unsigned int +get_socket_activation (void) +{ + const char *s; + unsigned int pid; + unsigned int nr_fds; + unsigned int i; + int fd; + + s = getenv ("LISTEN_PID"); + if (s == NULL) + return 0; + if (sscanf (s, "%u", &pid) != 1) { + fprintf (stderr, "%s: malformed %s environment variable (ignored)\n", + program_name, "LISTEN_PID"); + return 0; + } + if (pid != getpid ()) { + fprintf (stderr, "%s: %s was not for us (ignored)\n", + program_name, "LISTEN_PID"); + return 0; + } + + s = getenv ("LISTEN_FDS"); + if (s == NULL) + return 0; + if (sscanf (s, "%u", &nr_fds) != 1) { + fprintf (stderr, "%s: malformed %s environment variable (ignored)\n", + program_name, "LISTEN_FDS"); + return 0; + } + + /* So these are not passed to any child processes we might start. */ + unsetenv ("LISTEN_FDS"); + unsetenv ("LISTEN_PID"); + + /* So the file descriptors don't leak into child processes. */ + for (i = 0; i < nr_fds; ++i) { + fd = FIRST_SOCKET_ACTIVATION_FD + i; + if (fcntl (fd, F_SETFD, FD_CLOEXEC) == -1) { + /* If we cannot set FD_CLOEXEC then it probably means the file + * descriptor is invalid, so socket activation has gone wrong + * and we should exit. + */ + fprintf (stderr, "%s: socket activation: " + "invalid file descriptor fd = %d: %m\n", + program_name, fd); + exit (EXIT_FAILURE); + } + } + + return nr_fds; +} diff --git a/tests/Makefile.am b/tests/Makefile.am index afb9975..389bcfa 100644 --- a/tests/Makefile.am +++ b/tests/Makefile.am @@ -83,7 +83,14 @@ TESTS = \ test-single.sh \ test-captive.sh \ test-random-sock.sh \ - test-ipv4.sh + test-ipv4.sh \ + test-socket-activation + +check_PROGRAMS = \ + test-socket-activation + +test_socket_activation_SOURCES = test-socket-activation.c +test_socket_activation_CFLAGS = $(WARNINGS_CFLAGS) # In-depth tests need libguestfs, since that is a convenient way to # drive qemu. @@ -103,7 +110,6 @@ check_LTLIBRARIES = libtest.la libtest_la_SOURCES = test.c test.h libtest_la_CFLAGS = $(WARNINGS_CFLAGS) -check_PROGRAMS = check_DATA = check_SCRIPTS = diff --git a/tests/test-socket-activation.c b/tests/test-socket-activation.c new file mode 100644 index 0000000..5ecdde1 --- /dev/null +++ b/tests/test-socket-activation.c @@ -0,0 +1,226 @@ +/* nbdkit + * Copyright (C) 2017 Red Hat Inc. + * All rights reserved. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions are + * met: + * + * * Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * + * * Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * + * * Neither the name of Red Hat nor the names of its contributors may be + * used to endorse or promote products derived from this software without + * specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY RED HAT AND CONTRIBUTORS ''AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, + * THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A + * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL RED HAT OR + * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, + * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT + * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF + * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND + * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, + * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT + * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +/* Test socket activation. + * + * We cannot use the test framework for this since the framework + * always uses the -U flag which is incompatible with socket + * activation. Unfortunately this does mean we duplicate some code + * from the test framework. + * + * It's *almost* possible to test this from a shell script + * (cf. test-ipv4.sh) but as far as I can tell setting LISTEN_PID + * correctly is impossible from shell. + */ + +#include <config.h> + +#include <stdio.h> +#include <stdlib.h> +#include <stdint.h> +#include <inttypes.h> +#include <string.h> +#include <fcntl.h> +#include <unistd.h> +#include <signal.h> +#include <errno.h> +#include <sys/types.h> +#include <sys/socket.h> +#include <sys/un.h> +#include <sys/wait.h> + +#define FIRST_SOCKET_ACTIVATION_FD 3 + +#define NBDKIT_START_TIMEOUT 30 /* seconds */ + +#define NBDKIT_PLUGIN(name) \ + "../plugins/" name "/.libs/nbdkit-" name "-plugin.so" + +/* Declare program_name. */ +#if HAVE_DECL_PROGRAM_INVOCATION_SHORT_NAME == 1 +#include <errno.h> +#define program_name program_invocation_short_name +#else +#define program_name "nbdkit" +#endif + +static char tmpdir[] = "/tmp/nbdkitXXXXXX"; +static char sockpath[] = "/tmp/nbdkitXXXXXX/sock"; +static char pidpath[] = "/tmp/nbdkitXXXXXX/pid"; + +static pid_t pid = 0; + +static void +cleanup (void) +{ + if (pid > 0) + kill (pid, SIGTERM); + + unlink (pidpath); + unlink (sockpath); + rmdir (tmpdir); +} + +int +main (int argc, char *argv[]) +{ + int sock; + struct sockaddr_un addr; + char pid_str[16]; + size_t i, len; + char magic[8]; + + if (mkdtemp (tmpdir) == NULL) { + perror ("mkdtemp"); + exit (EXIT_FAILURE); + } + len = strlen (tmpdir); + memcpy (sockpath, tmpdir, len); + memcpy (pidpath, tmpdir, len); + + atexit (cleanup); + + /* Open the listening socket which will be passed into nbdkit. */ + sock = socket (AF_UNIX, SOCK_STREAM /* NB do not use SOCK_CLOEXEC */, 0); + if (sock == -1) { + perror ("socket"); + exit (EXIT_FAILURE); + } + + addr.sun_family = AF_UNIX; + len = strlen (sockpath); + memcpy (addr.sun_path, sockpath, len+1 /* trailing \0 */); + + if (bind (sock, (struct sockaddr *) &addr, sizeof addr) == -1) { + perror (sockpath); + exit (EXIT_FAILURE); + } + + if (listen (sock, 1) == -1) { + perror ("listen"); + exit (EXIT_FAILURE); + } + + if (sock != FIRST_SOCKET_ACTIVATION_FD) { + if (dup2 (sock, FIRST_SOCKET_ACTIVATION_FD) == -1) { + perror ("dup2"); + exit (EXIT_FAILURE); + } + close (sock); + } + + /* Run nbdkit. */ + pid = fork (); + if (pid == -1) { + perror ("fork"); + exit (EXIT_FAILURE); + } + if (pid == 0) { + /* Run nbdkit in the child. */ + setenv ("LISTEN_FDS", "1", 1); + snprintf (pid_str, sizeof pid_str, "%d", (int) getpid ()); + setenv ("LISTEN_PID", pid_str, 1); + + execlp ("../src/nbdkit", + "nbdkit", + "-P", pidpath, + "-o", + "-v", + NBDKIT_PLUGIN ("example1"), NULL); + perror ("exec: nbdkit"); + _exit (EXIT_FAILURE); + } + + /* We don't need the listening socket now. */ + close (sock); + + /* Wait for the pidfile to turn up, which indicates that nbdkit has + * started up successfully and is ready to serve requests. However + * if 'pid' exits in this time it indicates a failure to start up. + * Also there is a timeout in case nbdkit hangs. + */ + for (i = 0; i < NBDKIT_START_TIMEOUT; ++i) { + if (waitpid (pid, NULL, WNOHANG) == pid) + goto early_exit; + + if (kill (pid, 0) == -1) { + if (errno == ESRCH) { + early_exit: + fprintf (stderr, + "%s FAILED: nbdkit exited before starting to serve files\n", + program_name); + pid = 0; + exit (EXIT_FAILURE); + } + perror ("kill"); + } + + if (access (pidpath, F_OK) == 0) + break; + + sleep (1); + } + + /* Now nbdkit is supposed to be listening on the Unix domain socket + * (which it got via the listening socket that we passed down to it, + * not from the path), so we should be able to connect to the Unix + * domain socket by its path and receive an NBD magic string. + */ + sock = socket (AF_UNIX, SOCK_STREAM|SOCK_CLOEXEC, 0); + if (sock == -1) { + perror ("socket"); + exit (EXIT_FAILURE); + } + + /* Reuse addr which was set up above. */ + if (connect (sock, (struct sockaddr *) &addr, sizeof addr) == -1) { + perror (sockpath); + exit (EXIT_FAILURE); + } + + if (read (sock, magic, 8) != 8) { + perror ("read"); + exit (EXIT_FAILURE); + } + + if (memcmp (magic, "NBDMAGIC", 8) != 0) { + fprintf (stderr, "%s FAILED: did not read magic string from server\n", + program_name); + exit (EXIT_FAILURE); + } + + close (sock); + + /* Test succeeded. */ + exit (EXIT_SUCCESS); +} -- 2.10.2

7 years, 11 months

1
0
0 / 0

Re: [Libguestfs] virt-p2v-make-disk not making disk

by Richard W.M. Jones

On Sat, Jan 28, 2017 at 04:26:08PM -0800, Martin Lanner wrote: > I've used the following commands: > > ``` > # virt-p2v-make-disk -o /dev/sdb --inject-ssh-identity id_rsa_p2v > # virt-p2v-make-disk -o /dev/sdb centos-7.3 --inject-ssh-identity id_rsa_p2v > # virt-p2v-make-disk -o p2v/p2v.img centos-7.3 --inject-ssh-identity id_rsa_p2v > ``` > > ... on both CentOS 7.3 and Ubuntu 16.04, including from a bare metal machine to a USB drive, from a VM to a .img, as well as from a container to .img, and none of them seem to work. It downloads the base image, uncompresses it, and finally tries to open it. But that's where it stops: > > ``` > root@p2v:~# virt-p2v-make-disk -o p2v/p2v.img centos-7.3 --inject-ssh-identity id_rsa_p2v > [ 1.6] Downloading: http://libguestfs.org/download/builder/centos-7.3.xz > ################################################################## 100.0% > [ 223.9] Planning how to build this image > [ 223.9] Uncompressing > [ 235.7] Opening the new disk > ... > ``` Run libguestfs-test-tool and post the complete, unedited output on the mailing list. See also: http://libguestfs.org/guestfs-faq.1.html#debugging-libguestfs Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com libguestfs lets you edit virtual machines. Supports shell scripting, bindings from many languages. http://libguestfs.org

7 years, 11 months

1
0
0 / 0

[nbdkit PATCH v3 0/4] bind .zero to Python

by Eric Blake

This cleans up the existing code base with regards to implicit use of errno from language bindings, then rebases the previous work in python on top of that. I'm still playing with the perl bindings, but got further after reading 'perldoc perlembed'. Eric Blake (4): plugins: Don't use bogus errno from non-C plugins plugins: Add new nbdkit_set_error() utility function python: Expose nbdkit_set_error to python script python: Support zero callback docs/nbdkit-plugin.pod | 64 +++++++++++++++++++++++---- include/nbdkit-plugin.h | 14 +++++- plugins/ocaml/nbdkit-ocaml-plugin.pod | 11 +++++ plugins/ocaml/ocaml.c | 12 ++++++ plugins/perl/nbdkit-perl-plugin.pod | 5 +++ plugins/perl/perl.c | 12 ++++++ plugins/python/example.py | 11 +++++ plugins/python/nbdkit-python-plugin.pod | 53 +++++++++++++++++++++-- plugins/python/python.c | 76 +++++++++++++++++++++++++++++++++ plugins/ruby/nbdkit-ruby-plugin.pod | 5 +++ plugins/ruby/ruby.c | 12 ++++++ src/connections.c | 36 ++++++++++++---- src/errors.c | 6 +++ src/internal.h | 4 ++ src/plugins.c | 25 ++++++++++- src/tls.c | 30 ++++++++++++- 16 files changed, 349 insertions(+), 27 deletions(-) -- 2.9.3

7 years, 11 months

2
8
0 / 0

[PATCH v2 0/2] Fix srandom issues.

by Richard W.M. Jones

A better way to solve this I think. Rich.

7 years, 11 months

1
2
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

Libguestfs January 2017