lima-vm/lima · error

failed to attach disk %#q: %w

Error message

failed to attach disk %#q: %w

What it means

After locating the disk, attachDisks calls disk.LockForInstance(inst.Dir) to take an exclusive lock binding the disk to this instance; failure (disk already locked/attached to another running instance, stale lock, permission problem) is wrapped here.

Source

Thrown at pkg/driver/vz/vm_darwin.go:593

		isoDev, err := vz.NewUSBMassStorageDeviceConfiguration(isoAttachment)
		if err != nil {
			return err
		}
		configurations = append(configurations, isoDev)
	}

	diskUtil := proxyimgutil.NewDiskUtil(ctx)

	for _, d := range inst.Config.AdditionalDisks {
		diskName := d.Name
		disk, err := store.InspectDisk(diskName, d.FSType)
		if err != nil {
			return fmt.Errorf("failed to run load disk %#q: %w", diskName, err)
		}

		logrus.Infof("Mounting disk %#q on %#q", diskName, disk.MountPoint)
		if err = disk.LockForInstance(inst.Dir); err != nil {
			return fmt.Errorf("failed to attach disk %#q: %w", diskName, err)
		}
		extraDiskPath := filepath.Join(disk.Dir, filenames.DataDisk)
		// ConvertToRaw is a NOP if no conversion is needed
		logrus.Debugf("Converting extra disk %#q to a raw disk (if it is not a raw)", extraDiskPath)

		if err = diskUtil.Convert(ctx, raw.Type, extraDiskPath, extraDiskPath, nil, true); err != nil {
			return fmt.Errorf("failed to convert extra disk %#q to a raw disk: %w", extraDiskPath, err)
		}
		extraDiskPathAttachment, err := vz.NewDiskImageStorageDeviceAttachmentWithCacheAndSync(extraDiskPath, false, diskImageCachingMode, vz.DiskImageSynchronizationModeFsync)
		if err != nil {
			return fmt.Errorf("failed to create disk attachment for extra disk %#q: %w", extraDiskPath, err)
		}
		extraDisk, err := vz.NewVirtioBlockDeviceConfiguration(extraDiskPathAttachment)
		if err != nil {
			return fmt.Errorf("failed to create new virtio block device config for extra disk %#q: %w", extraDiskPath, err)
		}
		configurations = append(configurations, extraDisk)
	}

View on GitHub (pinned to dd909d0973)

Solutions

1. Stop the other instance currently using the disk (`limactl stop <other-instance>`) or detach the disk from it
2. Remove the stale lock file in the disk directory if no instance is running, then retry
3. Recreate or rename the disk if the lock state is unrecoverable

Example fix

// before: two lima.yaml files both with
additionalDisks: ["shared"]
// after
$ limactl stop instance-a   # release shared disk before starting instance-b

running, _ := exec.Command("limactl", "list").Output()
if strings.Contains(string(running), "Running") {
    // ensure no other running instance references the same additionalDisks entry
}

if err := startInstance(cfg); err != nil {
    if strings.Contains(err.Error(), "failed to attach disk") {
        // stop the conflicting instance or clear the stale lock
    }
}

When it happens

Trigger: Starting an instance while the same additional disk is already attached to another running instance; leftover lock file in the disk directory from a crash; filesystem permissions preventing lock creation.

Common situations: Running two instances sharing one additionalDisk; an earlier VM crashed leaving a lock; running limactl as different users against the same LIMA_HOME.

Related errors

• failed to run load disk %#q: %w
• failed to convert extra disk %#q to a raw disk: %w
• in use by instance %#q
• failed to unlock for reuse in the same instance: %w
• disk format %#q not supported, use `qcow2` or `raw` instead

AI-assisted analysis of lima-vm/lima@dd909d0973 (2026-09-01). Data as JSON: /api/errors/47af1e4723d6f790. Report an issue: GitHub.