395b03a77d
issue #549 : fix setrlimit() crash and hard-wire OS X default
...
OS X advertised unlimited, but really it means kern.maxfilesperproc.
2019-08-02 22:30:58 +01:00
33bceb6eb4
issue #602 : recover task_vars for synchronize and meta: reset_connection
2019-08-02 04:05:34 +01:00
6b4bcf4fe0
ansible: remove cutpasted docstring
2019-08-02 04:05:28 +01:00
619f4dee07
[linear2] merge fallout: restore optimization from #491 / 7b129e857
2019-08-02 04:05:19 +01:00
e4321f81a0
issue #600 : /etc/environment may be non-ASCII in an unknown encoding
2019-08-01 12:12:18 +01:00
75d179e4b9
remove unused imports flagged by lgtm
2019-07-31 11:46:23 +01:00
c80fddd487
[linear2]: merge fallout flaggged by LGTM
2019-07-31 11:41:29 +01:00
eeb7150f24
issue #549 : increase open file limit automatically if possible
...
While catching every possible case where "open file limit exceeded" is
not possible, we can at least increase the soft limit to the available
hard limit without any user effort.
Do this in Ansible top-level process, even though we probably only need
it in the MuxProcess. It seems there is no reason this could hurt
2019-07-31 04:20:04 +01:00
acab26d796
ansible: improve process.py docs
2019-07-31 04:09:48 +01:00
4dfbe82e76
tests: hide ugly error during Ansible tests
2019-07-31 01:50:37 +01:00
108015aa22
ansible: gracefully handle failure to connect to MuxProcess
...
It's possible to hit an ugly exception during early CTRL+C
2019-07-31 01:50:37 +01:00
bf1f3682aa
ansible: pin per-CPU muxes to their corresponding CPU
...
This slightly breaks the old scheme, in that CPU 1 may now end up with a
mux and the top-level process pinned to it.
2019-07-31 01:50:37 +01:00
dc9f4e89e6
ansible: reap mux processes on shut down
...
Previously we exitted without calling waitpid(), which meant the
top-level process struct rusage did not reflect the resource usage
consumed by the multiplexer processes.
Existing benchmarks are made using perf so this never created a problem,
but it could be confusing to others using the "time" command, and also
allows logging the final exit status of the process.
2019-07-31 01:50:37 +01:00
136dee1fb4
[linear2] more merge fallout, fix Connection._mitogen_reset(mode=)
2019-07-29 17:52:44 +01:00
a9755d4ad0
[linear2] update mitogen_get_stack for new _build_stack() return value
2019-07-29 16:30:01 +01:00
1fca0b7a94
[linear2] fix MuxProcess test fixture and some merge fallout
2019-07-29 16:10:36 +01:00
0f63ca4c68
Make setting affinity optional.
2019-07-29 13:52:30 +01:00
9035884c77
ansible: abstract worker process model.
...
Move all details of broker/router setup out of connection.py, instead
deferring it to a WorkerModel class exported by process.py via
get_worker_model(). The running strategy can override the configured
worker model via _get_worker_model().
ClassicWorkerModel is installed by default, which implements the
extension's existing process model.
Add optional support for the third party setproctitle module, so
children have pretty names in ps output.
Add optional support for per-CPU multiplexers to classic runs.
2019-07-29 13:52:30 +01:00
402dba4197
module_finder: pass raw file to compile()
...
Newer Ansibles have e.g. UTF-8 present in apt.py.
2019-07-23 16:04:44 +01:00
1aceacf89e
[stream-refactor] replace old detach_popen() reference
2019-07-23 14:07:00 +01:00
300f8b2ff9
ansible: fixturize creation of MuxProcess
...
This relies on the previous commit resetting global variables.
Update clean_shutdown() to handle duplicate calls, due to tests
repeatedly installing it.
2019-07-23 14:04:22 +01:00
26b6333787
[stream-refactor] fix unix.Listener construction
2019-07-22 18:30:51 +01:00
1a02a86331
Add buildah transport
2019-06-08 18:15:58 -05:00
7ae926b325
ansible: prevent tempfile.mkstemp() leaks.
...
This avoids a leak present in Ansible 2.7.0..current HEAD, and all
similar leaks.
See ansible/ansible#57327 .
2019-06-03 23:43:15 +01:00
3620fce071
issue #593 : expose configurables for SSH keepalive and increase the default
2019-06-03 17:40:58 +01:00
0b7fd3f290
issue #591 : ansible: restore CWD prior to AnsibleModule initialization.
2019-06-03 02:34:40 +01:00
4f23f0bec1
issue #590 : update comment to indicate the hack is permanent
2019-06-02 17:52:58 +01:00
1a92995a24
issue #590 : include nasty workaround for sys.modules junk
2019-05-30 06:00:54 +01:00
92b4724010
issue #587 : consistent become_exe() behaviour for older Ansibles.
2019-05-27 14:59:57 +01:00
f35194fe0f
issue #587 : mitogen_doas should not become_exe for doas_path
...
Looks like this has always been wrong - when used as a connection
method, PlayContext.become_method/become_exe may hold totally unrelated
data.
2019-05-27 14:59:57 +01:00
c1c8d5c31e
issue #587 : 2.8 PlayContext lacks sudo_flags attribute.
...
This is a huge bodge.
2019-05-26 16:46:38 +01:00
e11b251c75
issue #587 : 2.8 PluginLoader.get() introduced new collection_list kwarg
2019-05-26 12:07:55 +01:00
46dde95962
issue #587 : 2.8 PlayContext.connection no longer contains connection name
...
Not clear what the intention is here. Either need to ferret it out of
some other location, or just stop preloading the connection class in the
top-level process.
2019-05-26 12:06:53 +01:00
4a614c3950
issue #587 : bump max Ansible version
2019-05-26 12:06:14 +01:00
f105a81e20
ansible: descriptive version check during startup.
2019-05-19 13:13:24 +01:00
f30a4c05c8
issue #581 : expose mitogen_mask_remote_name variable.
2019-04-16 22:46:12 +01:00
65deb3feac
issue #575 : fix exception text rendering
2019-04-02 14:06:41 +01:00
34fb9da1be
issue #570 : add firewalld to always-fork list for now.
2019-03-22 19:46:08 +00:00
3ff6123483
issue #557 : support correct cpu_set_t size
2019-03-06 17:55:31 +00:00
2bd0bbd4df
issue #555 : ansible: workaround ancient reload(sys) hack.
...
This is the most minimal change for what might be relatively minimal
edge case. Alternative is replacing reload(), but let's not do that yet.
Closes #555
2019-03-06 14:05:29 +00:00
6309774be2
issue #554 : fix Ansible 2.4 compatibility
2019-03-06 12:06:55 +00:00
7743e57ff3
issue #554 : track and remove multiple make_tmp_path() calls.
2019-03-06 10:55:07 +00:00
7dacb68eeb
issue #552 : include process identity in log messages.
2019-02-25 17:25:09 +00:00
26e6194d0a
issue #548 : always treat transport=smart as 'ssh' for mitogen_via=.
...
The idea behind transport=smart is to select between paramiko and
OpenSSH given the availability of connection multiplexing and/or OSX
kernel bugs. We need to make no such choice.
2019-02-19 16:26:47 +00:00
458a4faa97
ansible: create stub __init__.py for sdist.
...
This went into 0.2.5 sdist tarball but it's not checked in.
2019-02-14 01:28:14 +00:00
8f9c67daf1
ansible: refactor affinity class and add abstract tests.
2019-02-14 00:35:16 +00:00
0f30808234
ansible: quiesce boto logger; closes #541 .
2019-02-13 17:28:01 +00:00
7fd0d34910
tests/ansible: Spec.port() test & mitogen_via= fix.
...
ansible_ssh_port was not respected.
2019-02-13 16:32:11 +00:00
1f77d24bec
Update copyright year everywhere.
2019-02-13 16:16:49 +00:00
b5b23e8f3d
tests/ansible: Spec.become_pass() test.
2019-02-13 16:14:15 +00:00
David Wilson