search for: nbd_is_dead

...)' probing for process existence. But on the command line, kill(1) has the behavior of sending SIGTERM by default when you omit a signal number (and not serving as a process existence probe). I guess our decision on what to do with the client requesting signal=0 boils down to whether we think nbd_is_dead() (for checking that the socket has gone away, but not necessarily if the subprocess still exists) and the knowledge that nbd_close() does wait() on the subprocess is sufficient for avoiding the need to do any direct probing. Thus, ACK with your wording fix. -- Eric Blake, Principal Software Eng...

Reverts commit 387cbe67c3db27e8a61117fedb6e7fad76e409ef. --- generator/generator | 18 +++++++++++++++++- lib/handle.c | 28 +++++++++++++++++++++++++++- tests/closure-lifetimes.c | 4 +++- 3 files changed, 47 insertions(+), 3 deletions(-) diff --git a/generator/generator b/generator/generator index 2cd83f1..25e4aa5 100755 --- a/generator/generator +++ b/generator/generator