search for: udev_settle_file

..., 3 insertions(+) > > diff --git a/daemon/guestfsd.c b/daemon/guestfsd.c > index dccfa15bc..9aa69301f 100644 > --- a/daemon/guestfsd.c > +++ b/daemon/guestfsd.c > @@ -402,6 +402,9 @@ int > is_root_device (const char *device) > { > struct stat statbuf; > + > + udev_settle_file(device); Add a space after the function name, but with that change ACK. > if (stat (device, &statbuf) == -1) { > perror (device); > return 0; Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization...

..., 3 insertions(+) > > diff --git a/daemon/guestfsd.c b/daemon/guestfsd.c > index b82c98acd..e6a4c65c4 100644 > --- a/daemon/guestfsd.c > +++ b/daemon/guestfsd.c > @@ -404,6 +404,9 @@ int > is_root_device (const char *device) > { > struct stat statbuf; > + > + udev_settle_file (device); > + > if (stat (device, &statbuf) == -1) { > perror (device); > return 0; This one is fine now, so ACK. Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpres...

On Wed, Apr 12, 2017 at 05:01:10PM +0300, Pavel Butsykin wrote: > Add udev_settle_file() to run 'udevadm settle' with --exit-if-exists option. It > will slightly reduce the waiting-time for pending events if we need to wait > for events related to a particular device/file. > > Signed-off-by: Pavel Butsykin <pbutsykin@virtuozzo.com> > --- > daemon/dae...

On Thursday, 13 April 2017 16:55:26 CEST Pavel Butsykin wrote: > Add udev_settle_file() to run 'udevadm settle' with --exit-if-exists option. It > will slightly reduce the waiting-time for pending events if we need to wait > for events related to a particular device/file. > > Signed-off-by: Pavel Butsykin <pbutsykin@virtuozzo.com> > --- > daemon/dae...

https://bugzilla.redhat.com/show_bug.cgi?id=1477623 The first two patches are cleanups. The third patch changes the way that we handle Device and Dev_or_Path parameters so that a parameter marked as such can really only contain a block device name (and not, for instance, a chardev). Using a chardev here caused hangs in the API. The next two patches fix API usage to conform to this new stricter

...ction tests if the parameter is a block + * device name. + * + * It can also be used in daemon code to test if the string passed + * as a C<Dev_or_Path> parameter is a device or path. + */ +int +is_device_parameter (const char *device) +{ + struct stat statbuf; + int fd; + uint64_t n; + + udev_settle_file (device); + + if (!STRPREFIX (device, "/dev/")) + return 0; + + /* Allow any /dev/sd device, so device name translation works. */ + if (STRPREFIX (device, "/dev/sd")) + return 1; + + /* Is it a block device in the appliance? */ + if (stat (device, &statbuf) == -1)...

Replace GUESTFSD_EXT_CMD with a command line option ‘./guestfsd --print-external-commands’

...dex b3f40628b..bcded501a 100644 --- a/daemon/guestfsd.c +++ b/daemon/guestfsd.c @@ -63,9 +63,6 @@ #include "daemon.h" -GUESTFSD_EXT_CMD(str_udevadm, udevadm); -GUESTFSD_EXT_CMD(str_uuidgen, uuidgen); - #ifndef MAX # define MAX(a,b) ((a)>(b)?(a):(b)) #endif @@ -1095,7 +1092,7 @@ udev_settle_file (const char *file) size_t i = 0; int r; - ADD_ARG (argv, i, str_udevadm); + ADD_ARG (argv, i, "udevadm"); if (verbose) ADD_ARG (argv, i, "--debug"); @@ -1124,7 +1121,7 @@ get_random_uuid (void) char *out; CLEANUP_FREE char *err = NULL; - r = command...

...exit (EXIT_FAILURE); } + /* Handle the --print-external-commands option. */ + if (do_print_external_commands) { + print_external_commands (); + exit (EXIT_SUCCESS); + } + #ifndef WIN32 /* Make sure SIGPIPE doesn't kill us. */ struct sigaction sa; @@ -1095,7 +1112,7 @@ udev_settle_file (const char *file) size_t i = 0; int r; - ADD_ARG (argv, i, str_udevadm); + ADD_ARG (argv, i, "udevadm"); if (verbose) ADD_ARG (argv, i, "--debug"); @@ -1124,7 +1141,7 @@ get_random_uuid (void) char *out; CLEANUP_FREE char *err = NULL; - r = command...

This is a simpler patch that removes GUESTFSD_EXT_CMD completely.