diff --git a/weed/shell/command_ec_balance.go b/weed/shell/command_ec_balance.go
index 9b5f2e9f0e..3f9685104b 100644
--- a/weed/shell/command_ec_balance.go
+++ b/weed/shell/command_ec_balance.go
@@ -24,7 +24,7 @@ func (c *commandEcBalance) Name() string {
 func (c *commandEcBalance) Help() string {
 	return `balance all ec shards among all racks and volume servers
 
-	ec.balance [-c EACH_COLLECTION|<collection_name>] [-apply] [-dataCenter <data_center>] [-shardReplicaPlacement <replica_placement>] [-diskType <disk_type>] [-volumeIds <id>[,<id>...]]
+	ec.balance [-collection EACH_COLLECTION|<collection_name>] [-apply] [-dataCenter <data_center>] [-shardReplicaPlacement <replica_placement>] [-diskType <disk_type>] [-volumeIds <id>[,<id>...]]
 
 	Options:
 	  -diskType: the disk type for EC shards (hdd, ssd, or empty for default hdd)
diff --git a/weed/shell/command_ec_rebuild.go b/weed/shell/command_ec_rebuild.go
index c812b77e03..4369c4a0c9 100644
--- a/weed/shell/command_ec_rebuild.go
+++ b/weed/shell/command_ec_rebuild.go
@@ -27,7 +27,7 @@ func (c *commandEcRebuild) Name() string {
 func (c *commandEcRebuild) Help() string {
 	return `find and rebuild missing ec shards among volume servers
 
-	ec.rebuild [-c EACH_COLLECTION|<collection_name>] [-apply] [-maxParallelization N] [-diskType=<disk_type>] [-checksum]
+	ec.rebuild [-collection EACH_COLLECTION|<collection_name>] [-apply] [-maxParallelization N] [-diskType=<disk_type>] [-checksum]
 
 	Before rebuilding, asks volume servers to recover any shards left unmounted by
 	a missing .ecx index (the index resides only on a peer server). Such shards are