The problem occurs when trying to stop cluster (either with tt stop or with tt restart) at the moment bootstrap have not been completed. In such a scenario once the replica instance is stopped with either of the mentioned commands the corresponding master instance exits instantly with the critical failure as below:
F> can't initialize storage: Can't check who replica <...> at <...> chose its bootstrap leader
As a result tt stop/restart for the master instance doesn't produce the termination-message:
The Instance small_cluster_app:storage-master (PID = <...>) has been terminated
because such kind of message is produced only for a running instance.
This commit introduces waiting for the cluster bootstrap completion after it is started or restarted.
The problem occurs when trying to stop cluster (either with
tt stopor withtt restart) at the moment bootstrap have not been completed. In such a scenario once the replica instance is stopped with either of the mentioned commands the corresponding master instance exits instantly with the critical failure as below:As a result
tt stop/restartfor the master instance doesn't produce the termination-message:because such kind of message is produced only for a running instance.
This commit introduces waiting for the cluster bootstrap completion after it is started or restarted.
Closes #1000