About this article: This article is reviewed against the cited sources. This is source research, not lab validation. Check the stated versions and your own requirements before making changes.
A raw device mapping (RDM) gives a VM direct access to a whole SAN LUN through a small mapping file on a VMFS datastore. You add one from Edit Settings: TechDocs says to "click the Add New Device button and select RDM Disk from the drop-down menu", then "In the Select Target LUN dialog box, select the target LUN for the raw device mapping and click OK." When that list is empty, or the LUN you want is missing, there are two different faults. Either the ESXi host does not see the LUN, or the host sees it and vCenter hides it. Find out which before you touch zoning or the array.
What the error means
The Select Target LUN list is built by vCenter from what the host reports, after filtering. KB 344351 covers both halves. Its symptom list includes "Creating Raw Device Mappings (RDMs) for use with Virtual Machines." Its background: "VMware vCenter facilitates numerous storage filters to determine using devices for potentially destructive operations such as formatting, creating extents, and creating Raw Device Mappings (RDMs) for use with VMs."
The RDM filter is config.vpxd.filter.rdmFilter. It hides a LUN in two cases: "The LUN is already being used as an RDM on the same host." and "The LUN is already being used as an RDM on a different host." The VMFS filter, config.vpxd.filter.vmfsFilter, hides a LUN that is "already being used as a datastore". KB 344351 adds that non-uniform presentation across hosts can leave candidate lists "either making the candidate list partial, or empty."
The LUN must also qualify for an RDM at all. TechDocs: "You cannot map to a disk partition. RDMs require the mapped device to be a whole LUN." And "The RDM is not available for direct-attached block devices or certain RAID devices. The RDM uses a SCSI serial number to identify the mapped device."
Likely causes
| Cause | What you see | Source |
|---|---|---|
| LUN not zoned or masked to this host | LUN missing from esxcli storage core device list |
KB 344351, KB 311466 |
| New LUN, no rescan yet | LUN appears after a rescan | TechDocs rescan, KB 308546 |
| LUN already in use as an RDM | Host lists the device; wizard does not | KB 344351 RDM filter |
| Shared RDM still attached to a running cluster node | Device "Connected", Target LUN list empty | KB 435880 |
| LUN already a VMFS datastore | Host lists it; wizard does not | KB 344351 VMFS filter |
| LUN IDs differ between hosts | Lists partial or empty; vMotion fails | KB 344351, KB 309797 |
| iSCSI network set up wrongly | Host "fails to discover the LUNs" | TechDocs iSCSI, KB 317719 |
| Local or RAID device | Device cannot be mapped | TechDocs RDM limits |
Prove the cause before changing anything
These steps only read, apart from the rescan, which KB 308546 says "does not cause a service interruption."
1. Ask the host whether it sees the LUN. On the ESXi host shell, run:
esxcli storage core device list
KB 344351: "The above command will list all the devices/LUN presented to the ESXi host. If a LUN/Device is not visible in the output of the command, then the device is not presented correctly or there are mapping issue's." Look for the NAA identifier your storage team gave you (naa. followed by hex digits). A good result lists a device entry carrying that identifier. A bad result has no entry with it: the host is not given the LUN, and no vCenter setting will fix that. Run the command on every host in the cluster and compare. KB 344351 requires that "the LUN ID for each respective LUN must match between all VMware ESXi Servers", and the filters compare hosts, so a LUN listed on one host and missing on another is itself the finding.
2. If it is missing, rescan. Low-impact step: KB 308546: "Performing a rescan does not cause a service interruption." TechDocs lists when a manual rescan is needed, including "Create new LUNs on a SAN.", "Change the path masking on a host." and "Add or remove discovery or static addresses (iSCSI only)." KB 308546 gives the command for all adapters:
esxcli storage core adapter rescan --all
or one adapter with esxcli storage core adapter rescan -A <vmkernel SCSI adapter name>; esxcli storage core adapter list lists them. In the vSphere Client: "Right-click on the host and navigate to Storage > Rescan Storage". Progress is written "in the /var/log/vmkernel.log and /var/log/vobd.log log files". Run step 1 again.
3. If it is still missing, check the paths. KB 311466: "The command esxcli storage core path list can be used to list all paths currently detected." For Fibre Channel: "If you cannot see the desired WWPN and WWNN addresses for your storage devices, you may have a zoning or Fiber Channel network or switch issue." If the array answers but the LUN is not in the list, "this may indicate an issue associating the Initiator with the device. The configuration that determines the association between storage device and ESXi host is stored on the storage array, and the array vendor may need to be engaged."
4. For iSCSI, test the network. KB 311466: "The command vmkping -I <interface> <destination IP> can be used to confirm IP communication to desired target." and nc -s <source ip> <destination IP> <port> with "port 860 and 3260". Then compare your design with TechDocs: "If you use separate vSphere switches, you must connect them to different IP subnets. Otherwise, VMkernel adapters might experience connectivity problems and the host fails to discover the LUNs." KB 317719 lists "Unable to see any storage from the storage device." when port binding is used where it should not be, and "Unable to see storage presented to the ESXi/ESX host." when it is missing where it is needed.
5. If the host sees the LUN, check who uses it. For each VM that could hold the LUN, open Edit Settings and look for an RDM hard disk: KB 424749 reads an RDM disk's NAA ID in the VM's Edit Settings. To see whether a datastore sits on the LUN, open the datastore's Configure tab in the vSphere Client and read its device backing; a LUN that already holds a VMFS datastore is hidden by the VMFS filter. KB 435880's case: "Although the LUNs are visible as "Connected" in the ESXi storage device list, the "Target LUN" selection dialog remains empty". It "also occurs if the shared RDM is still actively attached to a secondary cluster node that is currently powered ON, as the ESXi host places a strict lock on the .vmdk pointer file."
6. Compare LUN IDs across hosts. KB 344351: "the LUN ID for each respective LUN must match between all VMware ESXi Servers." KB 309797: "Ensure all hosts see the same LUN ID and VML ID for the RDM LUN(s)". Run step 1 on each host and compare the device's vml. identifier; KB 318949 shows how a LUN ID change alters the VML ID.
Example (illustrative): a new LUN is zoned to all four hosts in a cluster, and the RDM wizard is empty on one VM. esxcli storage core device list shows the NAA ID on every host. A search finds the same NAA ID mapped as an RDM on a test VM created last month. The RDM filter is hiding it, as designed.
Fix
Not presented to the host. Ask the storage team to present the LUN to every host in the cluster, then rescan with Rescan Storage or esxcli storage core adapter rescan --all (a low-impact step: KB 308546 says it "does not cause a service interruption"). KB 311466: if the array answers but the LUN is not in the list, "The configuration that determines the association between storage device and ESXi host is stored on the storage array, and the array vendor may need to be engaged." For a different case, an existing VM whose RDM shows a disk size of 0 MB and fails to power on with "Unable to enumerate all disks", KB 424749 gives the same storage-side step: "Coordinate with the storage team to verify LUN masking and zoning for the affected NAA ID." Risk: a zoning or masking change on the wrong LUN can expose data to the wrong host or take a datastore away; check the NAA ID with the storage team first.
iSCSI design. Put VMkernel ports in the layout the documentation describes. TechDocs: "Physical network adapters must be on the same subnet as the storage system they connect to." KB 317719 for port binding: "All VMkernel ports used for iSCSI connectivity must reside in the same broadcast domain and IP subnet." Disruptive step: changing port binding or VMkernel ports on a host with live iSCSI datastores can drop paths; put the host in maintenance mode first.
Already used as an RDM. Decide which VM should own the LUN. If a stale mapping is the cause, remove it from the old VM. For a cluster disk, follow KB 435880 (written for ESXi 8.x): "Ensure the RDM pointer file is completely removed from all secondary cluster nodes.", remove the 'Multi-writer' flag, then "Perform a Rescan Storage and Rescan Adapter on all ESXi hosts within the cluster" (a low-impact step, per KB 308546). On the secondary node, add the primary's pointer file with "Add New Device > Existing Hard Disk". Disruptive step: KB 435880 says releasing the lock may need "a maintenance window" to power off the secondary node.
Already a datastore. A LUN holding a VMFS datastore is not offered for an RDM. Use another LUN. Risk: do not remove a datastore to free its LUN without checking every VM and file on it.
LUN IDs differ. KB 344351: "schedule downtime and configure them for consistent or uniform presentation, ensuring each managed ESXi hosts see each respective LUN with the same LUN ID. If necessary, work with your storage array vendor". Disruptive step: changing a host LUN ID changes the VML ID; KB 318949 shows running VMs keeping the old one.
Local or RAID device. Use a SAN LUN or a virtual disk.
KB 344351 describes the filters as guarding "potentially destructive operations such as formatting, creating extents, and creating Raw Device Mappings (RDMs)". Fix the condition a filter reports instead of switching it off.
Confirm the fix
esxcli storage core device listshows the NAA ID on every host that may run the VM.- The Select Target LUN dialog lists the LUN, and the RDM disk is added.
- The RDM disk is added, and in Edit Settings it shows its real size. (An existing VM whose RDM shows 0 MB is a different fault, described in KB 424749.)
- vMotion of the VM to another host works.
What to watch afterwards
- Uniform LUN presentation for every new LUN across all hosts.
- vmkernel.log for path messages after zoning changes.
- Stale RDMs on retired or test VMs that keep a LUN hidden.
- iSCSI layout after network changes.
Read next
Related articles you may want next.
- VMwareSoftware iSCSI Paths Dead After a NIC Change: Check the 1:1 Uplink Policy on Bound PortsESXi 8.0 software iSCSI paths dead after a NIC change? For same-subnet port binding, record each bound port's uplinks, logins, sessions and paths before changing anything.Read article
- TroubleshootingAPD or PDL: classify an inaccessible ESXi datastoreDistinguish ESXi APD from PDL using device-specific evidence, then interpret the configured HA response without destructive recovery tests.Read article
- VMwarevSAN iSCSI Target Works for a Guest but Not an ESXi Host: Check the InitiatorAn ESXi host discovers a vSAN iSCSI target but no device appears, while a guest initiator works. Check the initiator type before treating it as a network fault.Read article
Risks and limits
- Zoning and masking are set on the switch and array; their screens differ by vendor.
- RDMs in physical compatibility mode cannot use VM snapshots, per TechDocs.
- Physical-mode RDMs "cannot be cloned, made into a template, or migrated if the migration involves copying the disk."
NetLeverage runs its own hosting on VMware vSphere, vCenter and ESXi. Talk to an engineer.
Sources
- Add an RDM Disk to a Virtual Machine — Broadcom TechDocs, Last Updated October 28, 2026, accessed 17 November 2026
- RDM Considerations and Limitations — Broadcom TechDocs, Last Updated October 28, 2026, accessed 17 November 2026
- Rescan Operations for ESXi Storage — Broadcom TechDocs, Last Updated October 28, 2026, accessed 17 November 2026
- Multiple Network Adapters in iSCSI or iSER Configuration — Broadcom TechDocs, Last Updated October 28, 2026, accessed 17 November 2026
- Performing a rescan of the storage on an ESXi host (KB 308546) — Broadcom, Updated On 11-10-2026, accessed 17 November 2026
- Understanding storage device detection in ESXi (KB 311466) — Broadcom, Updated On 03-04-2025, accessed 17 November 2026
- Cannot see some or all storage devices in VMware vCenter Server (KB 344351) — Broadcom, Updated On 09-22-2026, accessed 17 November 2026
- Considerations for using software iSCSI port binding in ESXi (KB 317719) — Broadcom, Updated On 09-28-2026, accessed 17 November 2026
- When attempting to add a Raw Device Mapping (RDM) to the VMs, the Target LUN list appears empty. (KB 435880) — Broadcom, Updated On 08-19-2026, accessed 17 November 2026
- Error: Unable to enumerate all disks for virtual machines with RDM disks (KB 424749) — Broadcom, Updated On 08-11-2026, accessed 17 November 2026
- Migrating a virtual machine with one or more RDMs attached fails with the error: Target disc should be different than the source disc, unable to access file (KB 309797) — Broadcom, Updated On 04-08-2025, accessed 17 November 2026
- VM with RDMs attached fails to power on after migrations or HA failover (KB 318949) — Broadcom, Updated On 10-12-2026, accessed 17 November 2026
This article was originally published on .