Commit graph

5178 commits

Author SHA1 Message Date
Toshio Kuratomi
a1428d6bed Remove tmp as a parameter to the connection plugins
There doesn't appear to be anything that actually uses tmp_path in the
connection plugins so we don't need to pass that in to exec_command.
That change also means that we don't need to pass tmp_path around in
many places in the action plugins any more.  there may be more cleanup
that can be done there as well (the action plugin's public run() method
takes tmp as a keyword arg but that may not be necessary).

As a sideeffect of this patch, some potential problems with chmod and
the patch, assemble, copy, and template modules has been fixed (those
modules called _remote_chmod() with the wrong order for their
parameters.  Removing the tmp parameter fixed them.)
2015-09-24 13:33:57 -07:00
James Cammarata
95b371dd60 Use AnsibleFileNotFound instead of AnsibleParsingError when YAML files are not found
And update portions of code to expect the proper error.

Fixes #12512
2015-09-24 16:27:25 -04:00
Toshio Kuratomi
0250beb68a Remove compress option from paramiko connection for now
It's not available on older versions of paramiko such as shipped in RHEL6
2015-09-24 13:18:00 -07:00
James Cammarata
12df9f2e31 Make hostvars more dynamic again to improve performance with large inventories
Fixes #12477
2015-09-24 13:53:44 -04:00
Abhijit Menon-Sen
82b33c381f We don't need even a token timeout here; just poll once
The process is already gone, so there's not going to be any new data
showing up on its stderr; we only want to make sure that we haven't
missed something that was already written. So polling once is enough.
2015-09-24 12:10:16 -04:00
Abhijit Menon-Sen
6e82df451a Clarify select() handling for ssh connections
This change is motivated by an ssh oddity: when ControlPersist is
enabled, the first (i.e. master) connection goes into the background; we
see EOF on its stdout and the process exits, but we never see EOF on its
stderr. So if we ran a command like this:

    ANSIBLE_SSH_PIPELINING=1 ansible -T 30 -vvv somehost -u someuser -m command -a whoami

We would first do select([stdout,stderr], timeout) and read the command
module output, then select([stdout,stderr], timeout) again and read EOF
on stdout, then select([stderr], timeout) AGAIN (though the process has
exited), and select() would wait for the full timeout before returning
rfd=[], and then we would exit. The use of a very short timeout in the
code masked the underlying problem (that we don't see EOF on stderr).

It's always preferable to call select() with a long timeout so that the
process doesn't use any CPU until one of the events it's interested in
happens (and then select will return independent of elapsed time).

(A long timeout value means "if nothing happens, sleep for up to <x>";
omitting the timeout value means "if nothing happens, sleep forever";
specifying a zero timeout means "don't sleep at all", i.e. poll for
events and return immediately.)

This commit uses a long timeout, but explicitly detects the condition
where we've seen EOF on stdout and the process has exited, but we have
not seen EOF on stderr. If and only if that happens, it reruns select()
with a short timeout (in practice it could just exit at that point, but
I chose to be extra cautious). As a result, we end up calling select()
far less often, and use less CPU while waiting, but don't sleep for a
long time waiting for something that will never happen.

Note that we don't omit the timeout to select() altogether because if
we're waiting for an escalation prompt, we DO want to give up with an
error after some time. We also don't set exceptfds, because we're not
actually acting on any notifications of exceptional conditions.
2015-09-24 12:10:16 -04:00
Toshio Kuratomi
03127dcfae remove the stdin return value from connection plugin exec_command() methods
The value was useless -- unused by the callers and always hardcoded to
the empty string.
2015-09-24 08:57:19 -07:00
James Cammarata
9d47eabfa4 Merge pull request #12506 from hyperized/devel
Add Weekday (0-6) as a number and add weeknumber (00-52)
2015-09-24 11:44:39 -04:00
Toshio Kuratomi
24b9e2e6d1 Update extras submodule ref 2015-09-24 07:18:23 -07:00
Gerben Geijteman
4c20964475 Add Weekday (0-6) as a number and add weeknumber (00-52) 2015-09-24 15:05:44 +02:00
Toshio Kuratomi
5d3d9cfe0d Convert to byte strings to avoid UnicodeErrors
Fixes #12488
2015-09-23 15:24:17 -07:00
Brian Coca
de18bcb95f correct typo on error reporting
fixes #12495
2015-09-23 10:11:52 -04:00
Abhijit Menon-Sen
40f608a377 A bit more debugging output
We used to display input chunks earlier anyway, so this isn't making
things more verbose.
2015-09-23 22:35:14 +05:30
Abhijit Menon-Sen
9700d9c04f Fix typo in checking select results
It's possible for more than one fd to be set, so 'elif' is obviously not
the right thing to use.
2015-09-23 22:32:15 +05:30
James Cammarata
1164e83477 Remove unnecessary calls to save inventory restrictions since 81bf88b 2015-09-23 12:18:09 -04:00
Toshio Kuratomi
89a78ba16e Update submodule refs 2015-09-23 08:40:59 -07:00
James Cammarata
9e734df0ec Conditionally poll longer if we're still waiting for an auth prompt 2015-09-23 11:20:11 -04:00
Toshio Kuratomi
5f0f5363b6 Merge pull request #12487 from mgedmin/py3k
Fix one more failing test on Python 3
2015-09-23 08:18:17 -07:00
James Cammarata
2898e000a0 Don't use the connection timeout for the select poll timeout 2015-09-23 11:13:12 -04:00
Abhijit Menon-Sen
587054db2a Send initial data before calling select whenever possible
Without this, we could execute «ssh -q ...» and call select(), which
would timeout after the default 10s, and only then send initial data.
(This is a relic of the earlier change where we always ran ssh with
-vvv, so the situation where it would sit quietly never happened in
practice; but this would have been the right thing to do even then.)
2015-09-23 20:09:50 +05:30
James Cammarata
c9a004227e Improve error catching from malformed playbook data
Fixes #12478
2015-09-23 08:56:36 -04:00
James Cammarata
e8e1d9f6fb Apply --limit to inventory in adhoc commands
Fixes #12473
2015-09-23 08:28:38 -04:00
Marius Gedminas
95e655eb67 Python 3: there's no basestring
Fixes one failing test.

The long series of module_utils/basic.py fixes were all because
module_utils/basic is imported in ansible/inventory/script.py.
2015-09-23 10:04:26 +03:00
Marius Gedminas
2c4982b58d Python 3: there's no itertools.imap
Because the builtin map() acts like an iterator already.
2015-09-23 10:04:26 +03:00
Marius Gedminas
6708d56a21 Python 3: avoid long integer literals
Even Python 2.4 automatically promotes int to long.
2015-09-23 10:04:25 +03:00
Marius Gedminas
f5d4935197 Python 3: treat python as a function in module_utils/basic.py
NB: we can't use 'from __future__ import print_function', but luckily
print(one_thing) works fine on both Python 2 and Python 3 without that.
2015-09-23 10:04:25 +03:00
Marius Gedminas
e71a986e16 Python 3: avoid octal constants in module_utils/basic.py 2015-09-23 10:04:25 +03:00
Marius Gedminas
d2bec7f81f Python 3: avoid "except ..., e:" in module_utils/basic.py
Make the code compatible with Pythons 2.4 through 3.5 by using
sys.exc_info()[1] instead.

This is necessary but not sufficient for Python 3 compatibility.
2015-09-23 10:04:25 +03:00
James Cammarata
65630d2ce1 Fixing one more bug related to staticmethods in LookupBase 2015-09-23 02:33:32 -04:00
James Cammarata
cbbb270761 Cleanup bug from moving base lookup methods to staticmethods 2015-09-23 02:26:19 -04:00
Abhijit Menon-Sen
ac98fe9e89 Implement ssh connection handling as a state machine
The event loop (even after it was brought into one place in _run in the
previous commit) was hard to follow. The states and transitions weren't
clear or documented, and the privilege escalation code was non-blocking
while the rest was blocking.

Now we have a state machine with four states: awaiting_prompt,
awaiting_escalation, ready_to_send (initial data), and awaiting_exit.
The actions in each state and the transitions between then are clearly
documented.

The check_incorrect_password() method no longer checks for empty strings
(since they will always match), and check_become_success() uses equality
rather than a substring match to avoid thinking an echoed command is an
indication of successful escalation. Also adds a check_missing_password
connection method to detect the error from sudo -n/doas -n.
2015-09-23 01:55:00 -04:00
Abhijit Menon-Sen
840a32bc08 Reorganise ssh.py to cleanly separate responsibilities
The main exec_command/put_file/fetch_file methods now _build_command and
call _run to handle input from/output to the ssh process. The purpose is
to bring connection handling together in one place so that the locking
doesn't have to be split across functions.

Note that this doesn't change the privilege escalation and connection IO
code at all—just puts it all into one function.

Most of the changes are just moving code from one place to another (e.g.
from _connect to _build_command, from _exec_command and _communicate to
_run), but there are some other notable changes:

1. We test for the existence of sshpass the first time we need to use
   password authentication, and remember the result.
2. We set _persistent in _build_command if we're using ControlPersist,
   for later use in close(). (The detection could be smarter.)
3. Some apparently inadvertent inconsistencies between put_file and
   fetch_file (e.g. argument quoting, sftp -b use) have been removed.

Also reorders functions into a logical sequence, removes unused imports
and functions, etc.

Aside: the high-level EXEC/PUT/FETCH description should really be logged
from ConnectionBase, while individual subclasses log transport-specific
details.
2015-09-23 01:55:00 -04:00
James Cammarata
95c6fe88e4 Fix handling of conditional vars_files which contain variables
Fixes #12484
2015-09-23 01:26:24 -04:00
Toshio Kuratomi
c83f51b7f2 Some LookupBase cleanups:
* Make LookupBase an abc with required methods (run()) marked as an
  abstractmethod
* Mark methods that don't use self as @staticmethod
* Document how to implement the run method of a lookup plugin.
2015-09-22 16:19:36 -07:00
Toshio Kuratomi
049952fa50 Update submodule refs. 2015-09-22 14:01:53 -07:00
James Cammarata
1e7fd2196d Fixing synchronize + delegate_to user bug
Fixes #12464
2015-09-22 16:06:52 -04:00
James Cammarata
3ffc2783c4 Don't bomb out on handlers with undefined variables in their names 2015-09-22 12:42:02 -04:00
Toshio Kuratomi
4b0d52d2cb Merge pull request #12420 from ansible/win_prefix_modules
Fix for user defined modules not overriding modules from core.
2015-09-22 09:16:38 -07:00
James Cammarata
1076155d8d When failing because of vars_files templating, try and bubble up the file/line info 2015-09-22 12:13:55 -04:00
Toshio Kuratomi
18e2ee16ef Fix for user defined modules not overriding modules from core.
This fix takes into account that powershell modules are somewhat
different than regular modules and have to be kept separate.
2015-09-22 09:07:37 -07:00
Toshio Kuratomi
f61fb9787d Update submodule refs 2015-09-22 08:59:10 -07:00
James Cammarata
c30e464388 Additional tweaks to callback output for delegate_to 2015-09-22 09:25:38 -04:00
James Cammarata
513619867a Show delegated-to host in callback message
Fixes #12465
2015-09-22 08:45:04 -04:00
James Cammarata
f563b22446 Merge pull request #12461 from mgedmin/py3k
Python 3: there's no basestring
2015-09-22 03:11:43 -04:00
James Cammarata
d2949f5449 Merge pull request #12463 from mgedmin/fix-ansible-doc
Fix option descriptions in ansible-doc output
2015-09-22 03:10:59 -04:00
James Cammarata
0fb4a6a67b Tweak to the way new host variables are created for delegated hosts 2015-09-22 03:03:21 -04:00
Marius Gedminas
339790adc4 Fix option descriptions in ansible-doc output
Fixes #12462.
2015-09-22 10:00:33 +03:00
James Cammarata
18adfc6d1a Set some default vars on hosts created for delegate_to connections 2015-09-22 02:35:01 -04:00
James Cammarata
a22f7b883d Restrict role param vars to tasks within that role
Fixes #12460
2015-09-22 02:14:18 -04:00
Marius Gedminas
fc0dcc3947 Python 3: there's no basestring
Fixes one failing test.

Now technically a filename can be a bytestring, even on Python 3.  I
hope this is unlikely for Ansible.
2015-09-22 08:42:33 +03:00