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.
When Consolidate or Delete All fails with "There is not enough space on the file system for the selected operation" or "File too large", the ESXi host could not find enough free space on the datastore to merge the snapshot files into the base disk. Broadcom's guidance is to have free space of at least 1.5 times the total size of the VM's snapshot files before trying again. Measure the snapshot files, free or add the space, then consolidate once and let it finish. Do not cancel it part way.
What the error means
A VM running on a snapshot writes its changes to delta files (-000001.vmdk and so on) instead of the base disk. Consolidation merges those delta files back into the base disk. Broadcom states that "the ESXi host must have sufficient free disk space on the virtual machine's assigned datastore" to do this (KB 392530), and that ESXi "requires temporary free space (overhead and metadata) to merge delta disks into a base disk" (KB 318352).
Where the space goes depends on whether the VM is running. Broadcom describes it this way in KB 318352:
- While powered on, Delete All "creates a temporary snapshot to buffer incoming I/O while older snapshots commit to the base disk". That temporary file grows with the VM's writes.
- While powered off, "No temporary snapshot or delta file is created because there are no active writes to track."
- Thin base disks "expand as snapshot data is committed, growing up to the maximum Provisioned Size".
The messages an administrator sees:
- vSphere Client: "Insufficient disk space on datastore '<datastore name>'. Consolidation failed for disk node 'scsi#:#': There is not enough space on the file system for the selected operation." (KB 392530)
- vSphere Client, other form: "An error occurred while consolidating disks: File too large", with the banner "Virtual machine disks consolidation is needed" (KB 398339).
/var/run/log/hostd.log: "Consolidate Disks translated error to vim.fault.NoDiskSpace" (KB 423621).- The VM's
vmware.log: "ConsolidateCheckSpaceNeeded: Not enough space to combine source disk" (KB 398339).
One trap: on some builds the task reports success while the disks are still on snapshot files (KB 398339). KB 423621 says this was fixed in ESXi 7.0 Update 3o (build 22348816) and 8.0 Update 1c (build 22088125). On older builds, check the disk file names, not the task result.
Likely causes
| Cause | What you see | How to confirm |
|---|---|---|
| Snapshot files larger than the free space can absorb | "not enough space on the file system" or "File too large" | Total of the delta files (command below) compared with the datastore's free space; less than 1.5 times means not enough |
| Snapshot kept for days or weeks | Large -00000N or -sesparse files; old snapshot dates |
vim-cmd vmsvc/snapshot.get shows the creation date; Broadcom advises no single snapshot be kept beyond 72 hours |
| Other files filling the datastore | Free space near zero; ISOs, old templates or unused disks present | Datastore browser, or esxcli storage filesystem list Free column |
| Datastore at 0% free | Files cannot even be deleted | Free column at or near 0; array-level snapshots may hold the last space (KB 392530) |
| Thin base disk growing during the merge | Free space falls during consolidation, even with the VM off | Provisioned size of the thin base disk is larger than its current size |
Prove the cause before changing anything
These checks only read. They run in an SSH session on the host that runs the VM.
1. Confirm which files the VM is running on. KB 316575 (Consolidating/Committing snapshots in VMware ESXi) gives:
grep file /vmfs/volumes/datastoreName/VMname/VMname.vmx
A disk line ending in -000001.vmdk (or any -00000N.vmdk) means that disk is still on a snapshot. A line naming only VMname.vmdk means it is on the base disk.
2. List the snapshots and their dates:
vim-cmd vmsvc/getallvms
vim-cmd vmsvc/snapshot.get [Vmid]
Take the Vmid from the first command. The second prints each snapshot's name and "Snapshot Created On". If the list is empty but the disks still point at -00000N.vmdk files, the VM needs consolidation rather than snapshot deletion. Broadcom also notes that snapshots taken by backup software through the API "may not appear in the Snapshot Manager" (KB 318825), so the file names are the reliable check.
3. Add up the snapshot files. KB 316414 (How to calculate current snapshot size) prints the total for one VM folder:
cd /vmfs/volumes/<datastore_name>/<VM_name>/
ls -la | grep -E 'delta|sesparse' | awk '{ SUM += $5/1024/1024 } END {printf "Total Snapshot Size: %.2f MB\n", SUM+0}'
The output is one line, for example (illustrative) Total Snapshot Size: 312480.00 MB. If the VM's disks sit on more than one datastore, run it in each folder.
4. Read the datastore's free space. KB 311779 (Device or Resource Busy) shows esxcli storage filesystem list, which prints each mounted datastore with Size and Free columns in bytes:
esxcli storage filesystem list
5. Compare. Multiply the snapshot total by 1.5. Broadcom's wording: "verify that the storage environment has at least 1.5x the total size of the snapshot available as insufficient capacity will result in repeated operation failure and potential data loss" (KB 392530). KB 398339 says the same: "the available free space must be at least 1.5 times the total size of all snapshot files (.vmdk)". If Free is below that figure, space is the cause.
6. Check the logs to match the error. In /var/run/log/hostd.log, look for Consolidate Disks message and NoDiskSpace. In the VM's vmware.log, look for ConsolidateCheckSpaceNeeded or Online consolidation failed.
Example (illustrative): VM app01 on datastore ds-gold-02 has three snapshots, the oldest 19 days old. The ls | awk total is 312,480 MB (about 305 GB). The 1.5 times target is about 458 GB. esxcli storage filesystem list shows 140 GB free on ds-gold-02. About another 320 GB must be freed or added first.
Fix
Work in this order. Each option frees or adds space; choose by what the datastore and the array allow.
1. Delete what does not belong on the datastore. Broadcom lists "ISOs, templates, unwanted virtual machines, or unused virtual disks" (KB 392530), and KB 318352 adds "old log bundles and ISO images". Risk: a deleted file cannot be recovered from the datastore. Confirm a disk is not attached to any VM (check every VM's settings, not just the obvious one) before removing it.
2. If the datastore is at 0% free, KB 392530 says you "may be unable to delete files directly from the datastore", and tells you to check the storage array "for array-level snapshots that may be consuming the final remaining capacity". Risk: an array snapshot may be someone's recovery point; agree with the storage owner before deleting it.
3. Move VMs with Storage vMotion. Move other VMs off the datastore if another datastore has room (KB 398339). KB 392530 also lists Storage vMotion of the affected VM's own disks to a datastore with more free space. KB 398339 says that if online consolidation fails, migrate the disk with the snapshot chain, or clone the VM, to another datastore while the VM is powered off.
4. Grow the datastore. KB 392530 says to "Expand the LUN to at least the recommended size based on the the error message details and extend the VMFS datastore". Its note: "SIOC does not support multiple extents. If you need to add an extent you will need to disable SIOC on the LUN." Risk: growing a LUN is an array change. Prefer growing the existing extent; an extra extent ties the datastore to two LUNs.
In the vSphere Client, right-click the datastore and choose Increase Datastore Capacity. A device whose Expandable column reads YES expands the existing extent; NO adds an extent (TechDocs). That page then says: "It is recommended to perform a cluster level rescan at vCenter Server after you perform any form of expansion done at the host level." Rescan every cluster that sees the device (right-click the host > Storage > Rescan Storage). Broadcom states that "Performing a rescan does not cause a service interruption" (KB 308546). 5. Consolidate again once Free is above the 1.5 times figure. In the vSphere Client: right-click the VM, Snapshots > Consolidate, and "Monitor the task in the vSphere Client to ensure successful completion" (KB 398339). If snapshots still exist, delete them through Snapshot Manager instead.
Disruptive step: consolidation slows the VM while it runs, and it can take hours or days. KB 316414 gives worked examples: 200 GB of snapshots at 2 MB/s takes about 26 hours, and 2 TB at 2 MB/s about 11.5 days; your storage may be faster or slower. Broadcom is firm on one point: "Snapshot consolidation should never be interrupted, as doing so can cause irreversible corruption of the virtual disks being consolidated" (KB 316414). Start it in a window long enough to finish and do not cancel it.
6. Consolidate with the VM powered off if free space is short and the VM can stop. KB 316414 recommends shutting the VM down for consolidation "to maintain data integrity during the process", and KB 343510 says that if space runs out while committing, "power off the VM and perform the same task" (KB 343510). Disruptive step: the VM is down until the merge completes. A thin base disk still grows as data is committed, so free space is still needed.
7. Last option: copy the VM out to another datastore. KB 398339 gives: "Power off the virtual machine. Migrate the disk with the snapshot chain to another datastore to consolidate, or clone the virtual machine to another datastore. Cloning automatically consolidates snapshots into the new destination disk." KB 392530 adds: "Delete the original VM once the new VM is deemed up and operational." Disruptive step: the VM is down for the copy. Risk: the clone is a new VM. Keep the original until the clone has started and been checked.
Confirm the fix
- Re-run the
grep filecheck on the.vmx. Every disk line names a base disk, with no-00000Nsuffix. vim-cmd vmsvc/snapshot.get [Vmid]returns no snapshots.- The VM no longer shows "Virtual machine disks consolidation is needed".
- The
ls | grep -E 'delta|sesparse'command lists nothing in the VM folder. - The datastore's Free column has gone up by roughly the size of the old delta files (less any growth of thin base disks).
While a long consolidation runs, KB 316414 shows how to watch the files change:
watch -n 2 -d 'ls -lth | grep -E "delta|flat|sesparse"'
The modified time on the files keeps moving while the merge works.
What to watch afterwards
- Snapshot age. Broadcom's guidance: "Do not retain a single snapshot for more than 72 hours", and use "only 2 to 3 snapshots" in a chain for performance (KB 318825). An alarm on snapshot age catches the next one before it fills the datastore.
- Backup software. A backup job that leaves its snapshot behind lets a delta file grow every day. Check VM folders for
-00000Nfiles after backup runs. - Datastore free space. KB 318856 names 200 MB as the level to act on and 100 MB as critical for VMFS itself; consolidation needs far more than that. KB 318352 recommends "a 20% free space buffer" for thin provisioning and consolidation.
Read next
Related articles you may want next.
- TroubleshootingvCenter says "Virtual machine disks consolidation is needed": why, and scenario-specific recovery decisions to clear itInvestigate disk consolidation warnings, identify conflicting holders and understand scenario-specific recovery options and escalation boundaries.Read article
- VMwareThe -000001.vmdk and delta files in a VM folder: what each one isLearn what each file in a VM folder is, how a healthy snapshot chain looks, and how to spot a broken one using file listings and named descriptors.Read article
- TroubleshootingA snapshot removal or consolidation task is stuck, or fails with a lock errorFind which host, process or backup proxy holds the disk file behind a stuck snapshot removal or consolidation, prove it, and clear it through scoped recovery decisions.Read article
Risks and limits
- Broadcom's pages are not identical on space. KB 318352 says a powered-off consolidation "requires no additional datastore space", yet it also says thin base disks grow while the data commits, and KB 392530 and 398339 ask for 1.5 times the snapshot size without separating powered-on from powered-off. Plan for the 1.5 times figure.
- The 1.5 times figure is Broadcom's rule of thumb, not a calculation of exact need. A busy VM consolidated while running can need more, because the temporary snapshot keeps growing.
- Version scope differs by page. KB 392530 lists "VMware vSphere ESXi (All Versions)", which carries the 7.x, 8.x and 9.x scope above; KB 398339 lists ESXi and vCenter 8.x and 9.x; KB 318352 lists ESXi 7.x and 8.x.
NetLeverage runs its own hosting on VMware vSphere, vCenter and ESXi. Talk to an engineer.
Sources
- Unable to poweron or consolidate a VM due to insufficient disk space on datastore for VMware vSphere ESXi (KB 392530) — Broadcom, Updated On 11-05-2026, accessed 17 November 2026
- VM Consolidation Failure Due to 'File Too Large' Error - vSphere ESXi (KB 398339) — Broadcom, Updated On 10-27-2026, accessed 17 November 2026
- How to calculate current snapshot size, estimate consolidation times, and understand performance factors affecting the consolidation process (KB 316414) — Broadcom, Updated On 10-22-2026, accessed 17 November 2026
- Virtual machine stops with "There is no more space for virtual disk" on NFS or VMFS Datastores (KB 318352) — Broadcom, Updated On 11-12-2026, accessed 17 November 2026
- "An error occurred while consolidating disks" when deleting one or more snapshots or consolidate disks (KB 343510) — Broadcom, Updated On 09-22-2026, accessed 17 November 2026
- Consolidating/Committing snapshots in VMware ESXi (KB 316575) — Broadcom, Updated On 08-11-2026, accessed 17 November 2026
- Best practices for using VMware snapshots in the vSphere environment (KB 318825) — Broadcom, Updated On 11-15-2026, accessed 17 November 2026
- Consolidate Snapshot disks fails but completes without any errors in the vSphere Client (KB 423621) — Broadcom, Updated On 12-24-2025, accessed 17 November 2026
- Error: "Device or Resource Busy" or "Device in use" when attempting to delete/unmount/disconnect a Datastore (KB 311779) — Broadcom, Updated On 09-12-2026, 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
- Increase Capacity of vSphere VMFS Datastore (vSphere 8.0 Storage) — Broadcom TechDocs, Last Updated October 28, 2026, accessed 17 November 2026
- Troubleshooting ESXi datastore or VMFS volume that is full or near capacity (KB 318856) — Broadcom, Updated On 07-22-2026, accessed 17 November 2026
This article was originally published on .