Clay Gerrard c70abba529 Adjust replica count before pulling parts from failed devices ... When your device count falls below your replica count you can either add devices or reduce the replica count. Trying to reduce your replica count fails about half the time because removing parts from from failed devices temporarily invalidates your _replica2part2dev table with NONE_DEV which can result in an IndexError in _adjust_replica2part2dev_size. If you adjust the replica count first you won't have to worry about tracking unassigned parts from failed devices. Closes-Bug: #1558751 Change-Id: I99dc776fd260a2ba68ca77d7b5ed5120d10b06de		2016-03-17 19:27:24 +00:00
..
account	Move HeaderKeyDict to avoid an inline import	2016-03-07 12:26:48 -08:00
cli	Merge "Adds storage policy option to recon"	2016-03-07 20:31:11 +00:00
common	Adjust replica count before pulling parts from failed devices	2016-03-17 19:27:24 +00:00
container	Move HeaderKeyDict to avoid an inline import	2016-03-07 12:26:48 -08:00
obj	Fix ssync related object-server docs	2016-03-16 08:58:31 +00:00
proxy	Put correct Etag and Accept-Ranges in EC 304 and 416 responses	2016-03-17 11:47:33 +00:00
test_locale	monkeypatch thread for keystoneclient	2015-11-03 16:36:19 +01:00
__init__.py	Move HeaderKeyDict to avoid an inline import	2016-03-07 12:26:48 -08:00