Merge pull request #333 from arangodb/bug-fix/upgrade-coordinator-downtime

Coordinator upgrade downtime

Author: neunhoef

pkg/deployment/reconcile/action_wait_for_member_up.go

@@ -167,6 +167,16 @@ func (a *actionWaitForMemberUp) checkProgressCluster(ctx context.Context) (bool,
 		log.Debug().Str("status", string(sh.Status)).Msg("Member set status not yet good")
 		return false, false, nil
 	}
+	// Wait for the member to become ready from a kubernetes point of view
+	// otherwise the coordinators may be rotated to fast and thus none of them
+	// is ready resulting in a short downtime
+	if m, found := a.actionCtx.GetMemberStatusByID(a.MemberID()); !found {
+		log.Error().Msg("No such member")
+		return false, true, nil
+	} else if !m.Conditions.IsTrue(api.ConditionTypeReady) {
+		log.Debug().Msg("Member not yet ready")
+		return false, false, nil
+	}
 	return true, false, nil
 }