From 4cd4e6d9bad9750f083be7dd97d37b0c83a8482e Mon Sep 17 00:00:00 2001 From: Mikko Rapeli Date: Wed, 15 Feb 2023 16:50:40 +0200 Subject: oeqa ssh.py: fix hangs in run() 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. 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. (From OE-Core rev: 04f080802b4a28709a105e4f0ead56a7a2da42b4) Signed-off-by: Mikko Rapeli Signed-off-by: Alexandre Belloni (cherry picked from commit 9c63970fce3a3d6029745252a6ec2bf9b9da862d) Signed-off-by: Steve Sakoman Signed-off-by: Richard Purdie --- meta/lib/oeqa/core/target/ssh.py | 39 ++++++++++++++++++++++++++++++--------- 1 file changed, 30 insertions(+), 9 deletions(-) (limited to 'meta/lib') diff --git a/meta/lib/oeqa/core/target/ssh.py b/meta/lib/oeqa/core/target/ssh.py index 48a463861d..4ab0cddb43 100644 --- a/meta/lib/oeqa/core/target/ssh.py +++ b/meta/lib/oeqa/core/target/ssh.py @@ -226,27 +226,33 @@ def SSHCall(command, logger, timeout=None, **opts): def run(): nonlocal output nonlocal process + output_raw = b'' starttime = time.time() process = subprocess.Popen(command, **options) 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) + data = process.stdout.read() if not data: - process.stdout.close() eof = True else: - output += data - logger.debug('Partial data from SSH call:\n%s' % data) + output_raw += data + # ignore errors to capture as much as possible + logger.debug('Partial data from SSH call:\n%s' % data.decode('utf-8', errors='ignore')) endtime = time.time() + timeout except InterruptedError: + logger.debug('InterruptedError') continue + process.stdout.close() + # process hasn't returned yet if not eof: process.terminate() @@ -254,6 +260,7 @@ def SSHCall(command, logger, timeout=None, **opts): try: process.kill() except OSError: + logger.debug('OSError when killing process') pass endtime = time.time() - starttime lastline = ("\nProcess killed - no output for %d seconds. Total" @@ -262,8 +269,21 @@ def SSHCall(command, logger, timeout=None, **opts): output += lastline else: - output = process.communicate()[0].decode('utf-8', errors='ignore') - logger.debug('Data from SSH call:\n%s' % output.rstrip()) + output_raw = process.communicate()[0] + + output = output_raw.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: + process.wait(timeout=5) + except TimeoutExpired: + try: + process.kill() + except OSError: + logger.debug('OSError') + pass options = { "stdout": subprocess.PIPE, @@ -292,4 +312,5 @@ def SSHCall(command, logger, timeout=None, **opts): process.kill() logger.debug('Something went wrong, killing SSH process') raise - return (process.wait(), output.rstrip()) + + return (process.returncode, output.rstrip()) -- cgit v1.2.3-54-g00ecf