Don't raise AnsibleConnectionFailure if the ssh process has already died. (#53534)

* Don't raise AnsibleConnectionFailure if the ssh_process has already died. Fixes #53487

* Better support for file not found messages

* Add changelog fragment
pull/53684/head
Matt Martz 6 years ago committed by GitHub
parent 15303b05ef
commit e9f9bcae6a
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

@ -0,0 +1,5 @@
bugfixes:
- ssh - Check the return code of the ssh process before raising AnsibleConnectionFailure, as the error message
for the ssh process will likely contain more useful information. This will improve the missing interpreter messaging
when using modules such as setup which have a larger payload to transfer when combined with pipelining.
(https://github.com/ansible/ansible/issues/53487)

@ -975,8 +975,8 @@ class ActionBase(with_metaclass(ABCMeta, object)):
# try to figure out if we are missing interpreter
if self._used_interpreter is not None:
match = '%s: No such file or directory' % self._used_interpreter.lstrip('!#')
if match in data['module_stderr'] or match in data['module_stdout']:
match = re.compile('%s: (?:No such file or directory|not found)' % self._used_interpreter.lstrip('!#'))
if match.search(data['module_stderr']) or match.search(data['module_stdout']):
data['msg'] = "The module failed to execute correctly, you probably need to set the interpreter."
# always append hint

@ -674,7 +674,7 @@ class Connection(ConnectionBase):
return b_command
def _send_initial_data(self, fh, in_data):
def _send_initial_data(self, fh, in_data, ssh_process):
'''
Writes initial data to the stdin filehandle of the subprocess and closes
it. (The handle must be closed; otherwise, for example, "sftp -b -" will
@ -687,7 +687,15 @@ class Connection(ConnectionBase):
fh.write(to_bytes(in_data))
fh.close()
except (OSError, IOError):
raise AnsibleConnectionFailure('SSH Error: data could not be sent to remote host "%s". Make sure this host can be reached over ssh' % self.host)
# The ssh connection may have already terminated at this point, with a more useful error
# Only raise AnsibleConnectionFailure if the ssh process is still alive
time.sleep(0.001)
ssh_process.poll()
if getattr(ssh_process, 'returncode', None) is None:
raise AnsibleConnectionFailure(
'SSH Error: data could not be sent to remote host "%s". Make sure this host can be reached '
'over ssh' % self.host
)
display.debug('Sent initial data (%d bytes)' % len(in_data))
@ -865,7 +873,7 @@ class Connection(ConnectionBase):
# If we can send initial data without waiting for anything, we do so
# before we start polling
if states[state] == 'ready_to_send' and in_data:
self._send_initial_data(stdin, in_data)
self._send_initial_data(stdin, in_data, p)
state += 1
try:
@ -979,7 +987,7 @@ class Connection(ConnectionBase):
if states[state] == 'ready_to_send':
if in_data:
self._send_initial_data(stdin, in_data)
self._send_initial_data(stdin, in_data, p)
state += 1
# Now we're awaiting_exit: has the child process exited? If it has,

Loading…
Cancel
Save