in manage share scheduler process, if failed to scheduler, the share size is 1, and have not commit quota usages, i think the
default share size is 1 is unreasonable, because we don't know the real size of the share in storage。so the default
size should be 0,then when admin force delete or unmange this manage_error share, will also quota cuts the usages, this is also unreasonable, this may lead to quota error.
in share mange layer,in function of manage_share(),if failed to get the real size of share, will raise, and also will not commit quota usages, then the size of share be set to 1. after admin force delete or umanage this share also lead to quota logic error.
in order to solve this problem, we should add to more status of share, they are STATUS_MANAGE_ERROR_DELETING:manage_error_deleting and
STATUS_MANAGE_ERROR_UNMANAGING:manage_error_unmanage_starting
when admin force delete or umanage the share,we shoud determine whether the status of share in above two, if in, we shoud skip quota usages cuts.
Fix proposed to branch: master /review. opendev. org/741588
Review: https:/