This can happen if you used to have a LUN on address 1, then removed it and presented a new one. The runtime name would be still incremented, but would be discovered first.
If so, a reboot would fix this.
Thanks Matt for a quick response. I am not sure if there was a Lun on address 1 but could be. would rescan of vmbha fix this or reboot is necessary? I have 8 esx hosts in a cluster and rebooting all of them is not something I want to go through.