Rietveld Code Review Tool
Help | Bug tracker | Discussion group | Source code | Sign in
(3231)

Issue 6460118: Properly destroy local provider environment

Can't Edit
Can't Publish+Mail
Start Review
Created:
11 years, 8 months ago by bcsaller
Modified:
11 years, 8 months ago
Reviewers:
hazmat, mp+120497
Visibility:
Public.

Description

Properly destroy local provider environment Stop using temp dir to hold pid Properly pass juju-directory on destory-environment https://code.launchpad.net/~bcsaller/juju/lxc-killpid/+merge/120497 (do not edit description out of merge proposal)

Patch Set 1 #

Total comments: 2
Unified diffs Side-by-side diffs Delta from patch set Stats (+19 lines, -28 lines) Patch
A [revision details] View 1 chunk +2 lines, -0 lines 0 comments Download
M juju/lib/service.py View 4 chunks +10 lines, -17 lines 2 comments Download
M juju/providers/local/__init__.py View 1 chunk +2 lines, -1 line 0 comments Download
M juju/providers/local/agent.py View 2 chunks +5 lines, -10 lines 0 comments Download

Messages

Total messages: 3
bcsaller
Please take a look.
11 years, 8 months ago (2012-08-21 05:00:11 UTC) #1
hazmat
On 2012/08/21 05:00:11, bcsaller wrote: > Please take a look. this was one of the ...
11 years, 8 months ago (2012-08-28 13:37:12 UTC) #2
hazmat
11 years, 8 months ago (2012-08-28 14:08:22 UTC) #3
https://codereview.appspot.com/6460118/diff/1/juju/lib/service.py
File juju/lib/service.py (right):

https://codereview.appspot.com/6460118/diff/1/juju/lib/service.py#newcode112
juju/lib/service.py:112: yield self._call("kill", self.get_pid())
having seen wrap-around pid behavior where this refers to an entirely different
process, this makes me wary.

https://codereview.appspot.com/6460118/diff/1/juju/lib/service.py#newcode118
juju/lib/service.py:118: 
what happens to the stop/destroy case if the pidfile doesn't exist?
Sign in to reply to this message.

Powered by Google App Engine
RSS Feeds Recent Issues | This issue
This is Rietveld f62528b