-
Type: Task
-
Resolution: Done
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: Server
-
Labels:None
Places in the docs that recommend sh.startBalancer() (e.g., after backing up a sharded cluster) are probably not optimal: startBalancer turns the balancer on and waits for balancing to commence, when users probably merely want to turn the balancer back on.
There's an alternate method, sh.setBalancerState(), which can just turn the balancer on. Perhaps the docs could recommend that. Alternatively, this could turn into a Core Server issue in the shell to have a tidier method to enable balancing after a stop.
- is related to
-
DOCS-6721 Recommend sh.startBalancer() over sh.setBalancerState(true) for re-enabling balancing
- Closed
-
SERVER-21766 Remove waiting for balancer lock behavior from sh.startBalancer
- Closed
- related to
-
SERVER-9979 Don't wait for balancer to start when calling startBalancer from the shell
- Closed