valitydev/salt
14
0
Fork 0
mirror of https://github.com/valitydev/salt.git synced 2024-11-08 01:18:58 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
36,165 Commits 24 Branches 182 Tags 341 MiB
e421aff68d
Commit Graph

36165 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Mathieu Le Marec - Pasquet
9987bb7a97 fix absent key dir 2014-10-13 15:26:46 -06:00
Erik Johnson
7048f9c760 Only skip yamlification if there's actually an underscore 
This fixes test breakage.
 2014-10-13 15:38:40 -05:00
Thomas S Hatch
a44bb73a2f Merge pull request #16596 from rallytime/bp-16501 
Backport #16501 to 2014.7
 2014-10-13 14:32:03 -06:00
Thomas S Hatch
ae3c2529cb Merge pull request #16595 from rallytime/bp-16420 
Backport #16420 to 2014.7
 2014-10-13 14:31:55 -06:00
Thomas S Hatch
bdff493542 Merge pull request #16586 from thatch45/new_args 
Change ext_module passing and make shim interactive to caller
 2014-10-13 14:27:25 -06:00
Thomas S Hatch
399a43d2ac Merge pull request #16580 from s0undt3ch/hotfix/vt-preexec-fn 
[READ CAREFULLY] User switching and umask setting does not belong in VT
 2014-10-13 14:24:51 -06:00
Thomas S Hatch
ef83ad23c3 Merge pull request #16579 from s0undt3ch/hotfix/vt-fixes 
More VT fixes
 2014-10-13 14:23:11 -06:00
Thomas S Hatch
718f778f8e Merge pull request #16594 from thatch45/cachedout-issue_16370 
merge #16561
 2014-10-13 14:20:21 -06:00
Gareth J. Greenaway
1baf90e8f3 Removing log.debug statement 2014-10-13 11:43:03 -06:00
Gareth J. Greenaway
645af2174a Consolidating 4 calls to salt.utils.network.interfaces() into one. Not completely happy with the approach of using a global. 2014-10-13 11:43:03 -06:00
Tomas Neme
7fa1142ff7 Make result=False if retcode is not 0 in module.run/wait 
This fixes commit 57423e031a which
introduced this code, but seems to check the wrong part of `ret`
 2014-10-13 11:29:17 -06:00
Pedro Algarvio
e683619f9f Having data left to read does not mean the process is alive 2014-10-13 17:19:42 +01:00
Thomas S Hatch
ca03400c1c Merge pull request #16568 from s0undt3ch/hotfix/thin-python-eggs 
Fix thin'ing python eggs
 2014-10-13 09:48:27 -06:00
Thomas S Hatch
95f81262ec lint 2014-10-13 09:25:14 -06:00
Thomas S Hatch
4a7db67fe1 Merge branch 'issue_16370' of https://github.com/cachedout/salt into cachedout-issue_16370 2014-10-13 09:21:46 -06:00
Thomas S Hatch
d482247540 Merge pull request #16550 from garethgreenaway/16536_forced_umount_with_test 
Fixes to mount state.
 2014-10-13 09:19:50 -06:00
Thomas S Hatch
dcdd88fcd5 Merge pull request #16548 from cachedout/issue_15133 
Remove expect fork from master upstart script
 2014-10-13 09:18:45 -06:00
Thomas S Hatch
b9db72fc9a Merge pull request #16529 from terminalmage/fix-virtual-grains-aix 
Suppress irrelevant log msg in virtual hardware detection
 2014-10-13 09:16:42 -06:00
Thomas S Hatch
c0333ed7f3 Merge pull request #16500 from UtahDave/2014.7local 
properly fix windows paths for storage on master when using cp.push_dir and cp.push
 2014-10-13 09:08:01 -06:00
Thomas S Hatch
9e879a1d91 lint 2014-10-13 08:44:40 -06:00
Erik Johnson
26deb6aaf5 Don't yamlify args containing only digits and underscores 
Fixes #16583.
 2014-10-13 02:22:27 -05:00
Thomas S Hatch
985ec7c05e IT WORKS!!! 
The shim is no longer passed in as a heredoc because it blocks
stdin. Now the shim python script can be passed into the ssh shell
command and can block! This means that it can prompt the calling
salt-ssh proceess to request extra data. this allows for the
download and update of the ext_mods, but as we move forward it
should allow for the thin deployment to all be in a single command
and not need any scp of sftp, this will also allow us to do bi-derectional
state transactions, maybe even distributing the compile process
which would make salt-ssh even faster
 2014-10-12 23:28:35 -06:00
Joseph Hall
1fdf232c5c Merge pull request #16565 from s0undt3ch/2014.7 
2014.7 fix opts loading in unittests
 2014-10-12 06:30:33 -06:00
Thomas S Hatch
f7e64cbe2f change shim to not take the options from the stdin 
this makes us modify the shim code inline with the options, while
less clean code wish it is cleaner execution wise.
this still doe snot fix the problem though, as the stdin readline still
does not block
 2014-10-12 00:59:34 -06:00
Thomas S Hatch
fdbb24eca6 change EOF string, set readline to stdin 
This still does not work, sys.stdin will not wait to accept anything
because of the way the pipe reads
 2014-10-11 23:44:45 -06:00
Thomas S Hatch
c5857d13a9 don't double send kwargs 2014-10-11 22:36:56 -06:00
Thomas S Hatch
6e79d28ec0 finish initial changes for passing inline ext_mods 
This also introduces using the shim as a flow control interface,
if this works then we may be able to continue to use the shim this
way to send mroe inline data down
 2014-10-11 22:24:02 -06:00
Pedro Algarvio
4470c0dc68 White-space lint fix 2014-10-12 04:29:59 +01:00
Thomas S Hatch
9c5550380c re-send the ext_mods line 2014-10-11 21:22:35 -06:00
Pedro Algarvio
00d7cf2db5 User switching and umask setting does not belong in VT 2014-10-12 04:21:37 +01:00
Pedro Algarvio
8aa113a94b We're sending a signal, not data. 2014-10-12 03:45:35 +01:00
Pedro Algarvio
51479e4592 Include a test case in order to check for full output from VT 2014-10-12 03:37:08 +01:00
Pedro Algarvio
9fdb27e4c0 Split flag_eof into flag_eof_stdout and flag_eof_stderr 2014-10-12 03:37:02 +01:00
Pedro Algarvio
743c26e562 Instead of checking for a live process, check for unread data. 
Add a unit test which tests proper behaviour.

Fixes #15941
 2014-10-12 02:28:35 +01:00
Pedro Algarvio
41870ef3e0 Fix ensure in syspath call 2014-10-12 02:16:23 +01:00
Pedro Algarvio
4fca1c96da Set the cachedir to a relative path 2014-10-12 02:16:23 +01:00
Pedro Algarvio
52d0d01d09 Set the cachedir to a relative path 2014-10-12 02:16:23 +01:00
Pedro Algarvio
e9fbd9910d Set the cachedir to a relative path 2014-10-12 02:16:23 +01:00
Pedro Algarvio
8d9570ef43 Set the cachedir to a relative path 2014-10-12 02:16:23 +01:00
Joseph Hall
130e0842c8 Merge pull request #16566 from terminalmage/docs 
Minor wording fix in highstate docs
 2014-10-11 06:34:39 -06:00
Joseph Hall
6706b64e5a Merge pull request #16569 from s0undt3ch/hotfix/vt-freezes-on-osx 
No need to check if process is alive. Fixes #16487. Refs #16317.
 2014-10-11 06:33:02 -06:00
Pedro Algarvio
79d89c7353 No need to check if process is alive. Fixes #16487. Refs #16317. 
There's no need to check for a live process because since we're changing flags on the file descriptor, if it's not none, we
reset the flags.
 2014-10-11 07:39:57 +01:00
Pedro Algarvio
ba834069e7 We need to decompress the eggs and include plain files in the thin tarball 2014-10-11 07:28:15 +01:00
Pedro Algarvio
09a524f696 Properly think python .egg's. Fixes #16567 2014-10-11 05:36:57 +01:00
Seth House
bbbfca8a87 Merge remote-tracking branch 'upstream/pr/16557' into 2014.7 2014-10-10 21:26:33 -06:00
Erik Johnson
af64f17fe4 Minor wording fix in highstate docs 
This fixes some grammar weirdness.
 2014-10-10 22:25:55 -05:00
Thomas Jackson
aa98ce3e28 Move to use lambda, to keep pylint happy 2014-10-10 17:19:13 -07:00
Thomas Jackson
aaf0069dd6 Wrap the split modules in objects, so we can avoid the attr/value ordering problem 2014-10-10 16:40:24 -07:00
Nicole Thomas
bb7dd9c8b7 Merge pull request #16562 from rallytime/bp-16559 
Backport #16559 to 2014.7
 2014-10-10 17:35:20 -06:00
Mike Place
cca7b80786 Pillar overwrite merge stategy 2014-10-10 16:38:19 -06:00