Files
nodepool/nodepool
James E. Blair 36da762a0d Use a persistent recursive watch for caches
This replaces the three treecache derived caches with a new
cache based on a persistent recursive watcher.

This should greatly reduce the number of watches on the zk server,
radically simplify the caching logic, and allows us to ignore
events for znodes we are uninterested in (for example, we can avoid
performing "get" calls on lock nodes) which should reduce zk
network traffic.

Change-Id: Ie8b578f3b8dd39fdf43025a31d55ce4e5801b563
2023-04-10 15:57:01 -07:00
..
2013-08-15 09:47:23 -07:00
2023-01-31 16:43:43 -08:00
2023-02-13 06:56:03 -08:00
2023-03-09 13:33:51 -08:00
2021-03-05 13:34:19 -08:00
2022-05-23 13:30:50 -07:00
2023-01-31 16:43:43 -08:00