fix: initialize new drwMutex for each attempt in 'for {' loop. (#14009)

It is possible that GetLock() call remembers a previously
failed releaseAll() when there are networking issues, now
this state can have potential side effects.

This PR tries to avoid this side affect by making sure
to initialize NewNSLock() for each GetLock() attempts
made to avoid any prior state in the memory that can
interfere with the new lock grants.

cmd/iam.go

@@ -213,9 +213,6 @@ func (sys *IAMSys) Init(ctx context.Context, objAPI ObjectLayer, etcdClient *etc
 	// Indicate to our routine to exit cleanly upon return.
 	defer cancel()
 
-	// Hold the lock for migration only.
-	txnLk := objAPI.NewNSLock(minioMetaBucket, minioConfigPrefix+"/iam.lock")
-
 	// allocate dynamic timeout once before the loop
 	iamLockTimeout := newDynamicTimeout(5*time.Second, 3*time.Second)
 
@@ -223,6 +220,9 @@ func (sys *IAMSys) Init(ctx context.Context, objAPI ObjectLayer, etcdClient *etc
 
 	// Migrate storage format if needed.
 	for {
+		// Hold the lock for migration only.
+		txnLk := objAPI.NewNSLock(minioMetaBucket, minioConfigPrefix+"/iam.lock")
+
 		// let one of the server acquire the lock, if not let them timeout.
 		// which shall be retried again by this loop.
 		lkctx, err := txnLk.GetLock(retryCtx, iamLockTimeout)