David Shrewsbury afa66c7f58 Only DeletedNodeWorker should delete nodes ... The proper/expected way to delete a node is to set its state to DELETING and then allow the DeletedNodeWorker to actually delete the instance and zookeeper node. Not doing it this way introduces the race seen in commit 59d636740a. This change fixes all the places (other than the nodepool command to delete --now) where we were not doing it properly. Change-Id: I1962ad672d3d5e646ed29beb07b720fc53d604ed		2018-02-20 08:27:09 -05:00
..
cmd	Store build logs automatically	2018-02-09 07:50:20 -08:00
driver	Merge "Provider wedge fix"	2018-02-03 20:24:55 +00:00
tests	Hack for test_delete_now rare failures	2018-02-19 16:20:46 -05:00
__init__.py	Initial commit	2013-08-15 09:47:23 -07:00
builder.py	Store build logs automatically	2018-02-09 07:50:20 -08:00
config.py	Store build logs automatically	2018-02-09 07:50:20 -08:00
exceptions.py	Do pep8 housekeeping according to zuul rules	2018-01-17 02:17:45 +00:00
launcher.py	Only DeletedNodeWorker should delete nodes	2018-02-20 08:27:09 -05:00
logconfig.py	Add default logging configuration	2018-02-07 15:49:19 -08:00
nodeutils.py	Implement a static driver for Nodepool	2018-01-31 03:55:56 +00:00
provider_manager.py	Add a plugin interface for drivers	2018-01-19 00:45:56 +00:00
stats.py	Do pep8 housekeeping according to zuul rules	2018-01-17 02:17:45 +00:00
status.py	Add /node-list to the webapp	2018-02-06 13:21:08 +01:00
task_manager.py	Do pep8 housekeeping according to zuul rules	2018-01-17 02:17:45 +00:00
version.py	Fix nodepool --version	2015-02-06 12:09:52 -08:00
webapp.py	Add /node-list to the webapp	2018-02-06 13:21:08 +01:00
zk.py	Do not delete unused but allocated nodes	2018-02-06 12:01:54 -05:00