VMware vSphere / ESXi error reference · with fixes
Search a VMware ESXi or vCenter error by its code, message or symptom and get the cause and the exact steps to resolve it — the esxcli or PowerCLI command, the log to read, the service to restart. Click any error to expand the full resolution and a verification step. Filter by category to browse PSOD, storage, networking, vMotion, HA, snapshots and host-connectivity problems. Runs entirely in your browser.
How to use this VMware ESXi error reference
VMware surfaces failures in many forms: a purple diagnostic screen (PSOD) on the ESXi host, a red banner in the vSphere Client, a cryptic esxcli return, or a status like Host not responding in vCenter. The same underlying problem often appears with different wording depending on where you hit it. This reference groups the common ones across ESXi, vCenter and vSphere, and for each gives the root cause plus a numbered fix you can follow immediately. Type any part of the message into the search box, or filter by category, then expand the card to read the resolution and a verification step.
Most vSphere problems fall into a few buckets: a host crash (PSOD), boot and install failures, host-to-vCenter connectivity, storage (APD, PDL, datastore locks, VMFS), networking, migration and availability (vMotion, HA, DRS), snapshots, VM power-on and VMware Tools. Knowing the bucket usually points straight at the fix and the right log to read.
The logs and commands that solve most VMware issues
- vmkernel.log (
/var/log/vmkernel.log) for host, driver, storage and PSOD-adjacent events. - hostd.log and vpxa.log for management-agent and vCenter-connectivity problems.
- vmware.log inside each VM folder on the datastore for per-VM power-on and disk errors.
- Restart the management agents safely from the host shell:
/etc/init.d/hostd restartand/etc/init.d/vpxa restart. - Inspect storage paths with
esxcli storage core path listand adapters withesxcli network nic list.
PSOD: reading a purple screen
A Purple Screen of Death halts the ESXi host and prints the exception, the failing module and a backtrace. The exception type is the biggest clue: a #PF Exception 14 (page fault) usually points at a driver; a hardware LINT1/NMI points at memory or a faulty component; a VMFS heap message points at storage exhaustion. Configure a coredump target (a dump collector or a local partition) so the host writes a dump you can analyse, then update the implicated driver, firmware or hardware. Most PSODs are driver or hardware faults, not VMware bugs.
Storage: APD versus PDL
Two storage states cause the most confusion. APD (All Paths Down) is a transient loss of all paths to a device where the array may come back, so ESXi waits and retries. PDL (Permanent Device Loss) is the array telling ESXi the device is gone for good, identified by specific SCSI sense codes. The fix differs: APD is usually a fabric, zoning or array-controller problem to restore; PDL means you remove the dead device and remediate the datastore. Misreading one for the other wastes hours.
Privacy and how this tool runs
The whole error dataset is embedded in the page and the search runs in your browser with plain JavaScript. Nothing you type is sent to a server, logged or stored. The reference works offline once the page has loaded.
Frequently asked questions
What causes a VMware ESXi purple screen (PSOD)?
Most PSODs are caused by a faulty or mismatched driver, failing hardware (memory, CPU, a PCIe card), or storage heap exhaustion, not by ESXi itself. Read the exception on the screen: a #PF Exception 14 points at a driver, a LINT1 or NMI points at hardware. Configure a coredump target, then update the implicated driver and firmware against the VMware HCL.
How do I fix an ESXi host showing as Not Responding in vCenter?
Restart the management agents from the host shell with /etc/init.d/hostd restart and /etc/init.d/vpxa restart, confirm the management network and that vCenter can reach the host on port 902 and 443, then right-click the host and choose Reconnect. If the agents will not start, check disk space on the host scratch and the hostd.log.
What is the difference between APD and PDL?
APD (All Paths Down) is a temporary loss of all paths where the device may return, so ESXi keeps retrying. PDL (Permanent Device Loss) is the storage array signalling the device is gone permanently via SCSI sense codes. APD is a fabric or array problem to restore; PDL requires removing the dead device and remediating affected datastores and VMs.
Why does vMotion fail with a CPU incompatibility error?
The source and destination hosts expose different CPU instruction sets, so the running VM cannot move without risking a crash. Enable Enhanced vMotion Compatibility (EVC) on the cluster at the highest baseline both CPUs support, which masks the newer features. The VM must be powered off once to pick up an EVC mode it was started outside of.
How do I power on a VM that says the file is locked?
Another host or a stale process holds a lock on the VMDK or the .vmx. Find the lock owner with vmkfstools -D on the locked file (the MAC in the output identifies the host), confirm the VM is not running elsewhere, and clear the stale lock by restarting the holding host management agents or, as a last resort, the holding host. Never delete .lck files blindly.
Why can’t I delete or consolidate a snapshot?
Usually insufficient free space on the datastore for the merge, or an open file handle (a backup job) on the delta disk. Free space to at least the size of the delta chain, stop any backup touching the VM, then use Snapshot > Consolidate. Check vmware.log in the VM folder for the file the merge is blocked on.
