[lnst] Machine: handle timeout exception

commit 98d43fde2764126910c23ac7cb7fe4c14e15c4a4 Author: Ondrej Lichtner <olichtne(a)redhat.com&gt; Date: Wed Jul 17 14:09:45 2013 +0200 Machine: handle timeout exception Previously when a command timed out, the exception was handled too late resulting in a message "Recieved an unknown exception...". This commit adds handling of this exception. The timed out command is then killed so that it doesn't leave a process running. Signed-off-by: Ondrej Lichtner <olichtne(a)redhat.com&gt; Signed-off-by: Jiri Pirko <jiri(a)resnulli.us&gt; lnst/Controller/Machine.py | 9 ++++++++- 1 files changed, 8 insertions(+), 1 deletions(-) --- diff --git a/lnst/Controller/Machine.py b/lnst/Controller/Machine.py index d4348e2..1847adf 100644 --- a/lnst/Controller/Machine.py +++ b/lnst/Controller/Machine.py @@ -174,7 +174,14 @@ class Machine(object): logging.debug("Setting timeout to \"%d\"", timeout) signal.alarm(timeout) - cmd_res = self._rpc_call("run_command", command) + try: + cmd_res = self._rpc_call("run_command", command) + except MachineError as exc: + if "bg_id" in command: + self._rpc_call("kill_command", command["bg_id"]) + else: + self._rpc_call("kill_command", None) + cmd_res = {"passed": False, "err_msg": str(exc)} signal.alarm(0) signal.signal(signal.SIGALRM, prev_handler)