@@ -229,29 +229,41 @@ def SSHCall(command, logger, timeout=None, **opts):
if timeout:
endtime = starttime + timeout
eof = False
+ os.set_blocking(process.stdout.fileno(), False)
while time.time() < endtime and not eof:
- logger.debug('time: %s, endtime: %s' % (time.time(), endtime))
try:
+ logger.debug('Waiting for process output: time: %s, endtime: %s' % (time.time(), endtime))
if select.select([process.stdout], [], [], 5)[0] != []:
- reader = codecs.getreader('utf-8')(process.stdout, 'ignore')
- data = reader.read(1024, 4096)
+ # wait a bit for more data, tries to avoid reading single characters
+ time.sleep(0.2)
+ logger.debug('Reading from process stdout')
+ data = process.stdout.read()
if not data:
- process.stdout.close()
+ logger.debug('End of file from process stdout')
eof = True
else:
- output += data
- logger.debug('Partial data from SSH call:\n%s' % data)
+ # ignore errors to capture as much as possible
+ string = data.decode('utf-8', errors='ignore')
+ output += string
+ logger.debug('Partial data from SSH call:\n%s' % string)
endtime = time.time() + timeout
except InterruptedError:
+ logger.debug('InterruptedError')
continue
+ logger.debug('Closing process stdout')
+ process.stdout.close()
+
# process hasn't returned yet
if not eof:
+ logger.debug('Terminating process')
process.terminate()
time.sleep(5)
try:
+ logger.debug('Killing process')
process.kill()
except OSError:
+ logger.debug('OSError when killing process')
pass
endtime = time.time() - starttime
lastline = ("\nProcess killed - no output for %d seconds. Total"
@@ -263,6 +275,19 @@ def SSHCall(command, logger, timeout=None, **opts):
output = process.communicate()[0].decode('utf-8', errors='ignore')
logger.debug('Data from SSH call:\n%s' % output.rstrip())
+ # timout or not, make sure process exits and is not hanging
+ if process.returncode == None:
+ try:
+ logger.debug('Process still running, waiting for it to end')
+ process.wait(timeout=5)
+ except TimeoutExpired:
+ try:
+ logger.debug('Timeout, killing process')
+ process.kill()
+ except OSError:
+ logger.debug('OSError')
+ pass
+
options = {
"stdout": subprocess.PIPE,
"stderr": subprocess.STDOUT,
@@ -282,12 +307,15 @@ def SSHCall(command, logger, timeout=None, **opts):
options['env'] = env
try:
+ logger.debug('run')
run()
except:
+ logger.debug('except from run()')
# Need to guard against a SystemExit or other exception ocurring
# whilst running and ensure we don't leave a process behind.
if process.poll() is None:
process.kill()
logger.debug('Something went wrong, killing SSH process')
raise
- return (process.wait(), output.rstrip())
+
+ return (process.returncode, output.rstrip())
When qemu machine hangs, the ssh commands done by tests are not timing out. do_testimage() task has last logs like this: DEBUG: time: 1673531086.3155053, endtime: 1673531686.315502 The test process is stuck for hours, or for ever if the executing command or test case did not set a timeout correctly. The default 300 second timeout is not working when target hangs. Note that timeout is really a "inactive timeout" since data returned by the process will reset the timeout. Added debug prints help to identify which calls are hanging and not timing out. Make the process stdout non-blocking so read() will always return right away using os.set_blocking() available in python 3.5 and later. Then change from python codec reader to plain read() and make the ssh subprocess stdout non-blocking. Even with select() making sure the file had input to be read, the codec reader was trying to find more stuff and blocking for ever when process hangs. While at it, add a small timeout to read data in larger chunks if possible. This avoids reading data one or few characters at a time and makes the debug logs more readable. close() the stdout file in all cases after read loop is complete. Then make sure to wait or kill the ssh subprocess in all cases. Just reading the output stream and receiving EOF there does not mean that the process exited, and wait() needs a timeout if the process is hanging. In the end kill the process and return the return value and captured output utf-8 encoded, just like before these changes. This fixes ssh run() related deadlocks when a qemu target hangs completely. Signed-off-by: Mikko Rapeli <mikko.rapeli@linaro.org> --- meta/lib/oeqa/core/target/ssh.py | 42 ++++++++++++++++++++++++++------ 1 file changed, 35 insertions(+), 7 deletions(-)