Eric Blake <ebl...@redhat.com> writes: > Management apps like libvirt don't know to pay attention to > stderr unless there is a non-zero exit status. > > * migration.c (process_incoming_migration_co): Exit with non-zero > status on failure. > > Signed-off-by: Eric Blake <ebl...@redhat.com>
It looks like the migration coroutine cleans itself so there's no obvious way to propagate the error. A nicer cleanup would be to switch the fprintf to error_report() too but this is certainly better than what we have now. Maybe we need an error_report_fatal()... Reviewed-by: Anthony Liguori <aligu...@us.ibm.com> Regards, Anthony Liguori > --- > > Noticed while reviewing: > https://lists.gnu.org/archive/html/qemu-devel/2013-04/msg03293.html > and it seems blatant enough to fix now, rather than waiting for > Pavel's series to stabilize. > > Side note: libvirt explicitly forbids all use of exit({0,1}), and > instead encourages exit(EXIT_{SUCCESS,FAILURE}), precisely because > it makes it harder to slip in unintentional successful exits. > > migration.c | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/migration.c b/migration.c > index 3b4b467..3eb0fad 100644 > --- a/migration.c > +++ b/migration.c > @@ -99,7 +99,7 @@ static void process_incoming_migration_co(void *opaque) > qemu_fclose(f); > if (ret < 0) { > fprintf(stderr, "load of migration failed\n"); > - exit(0); > + exit(EXIT_FAILURE); > } > qemu_announce_self(); > DPRINTF("successfully loaded vm state\n"); > -- > 1.8.1.4
