2016-10-22 15:55:11 |
Yuriy Zveryanskyy |
description |
We ignore exception.NoFreeConductorWorker in the agent driver and waiting for the next heartbeat, but 'resume' event already processed in the continue_node_clean() conductor's method before spawning a thread. As result node stuck on CLEANING state even if next heartbeat can execute thread via conductor. |
We ignore exception.NoFreeConductorWorker in the agent driver and waiting for the next heartbeat, but 'resume' event already processed in the continue_node_clean() conductor's method before spawning a thread. As result cleaning fails even if next heartbeat can execute thread via conductor. |
|