Import from vmware failed - NFS error

[luboss79]

Hello,
I am trying to use the Import-Export tool in ACS with configured NFS primary storage. I'm getting the NFS error:

2026-03-17 12:26:59,607 ERROR [c.c.h.g.VMwareGuru] (API-Job-Executor-8:[ctx-59095062, job-17951, ctx-7c2428ef]) (logid:59fd44a1) Error create template file of the VM: vm-export-test from vCenter vcenter.ourdomain.sk: Unable to create mount point for nfs://172.0.0.10//mnt/NFS_DATA/psm-ps-migration com.cloud.utils.exception.CloudRuntimeException: Unable to create mount point for nfs://172.0.0.10//mnt/NFS_DATA/psm-ps-migration

Here are the steps to reproduce my issue:

1. Select Migrate existing instance to KVM
2. Choose the External VMware vCenter - add the connect parameters
3. VM instance for export is visible in Instance Listing
4. Choose the VM - Import Instance
5. Fill the name, choose the compute offering
6. Click OK

ACS version: 4.22.0
KVM Hosts: Oracle Linux 8.10
Primary storage: NFS

The question is: Please explain the log, does it mean that the NFS storage (ACS primary) is beeing mounted on the vcenter server? If so, is it mandatory prerequisite for importing the instance from vmware?

Thank you

[nvazquez]

Hi @luboss79 can you share the management server logs around the failure time or the strack strace of the error?

[luboss79]

Hi,
of course, the log file is attached.
Thank you
cloudstack-management-vmware-export.log

[luboss79]

Hi @nvazquez, here is the full log with the stack trace. Please can you check it?
Thank you
impor_vmware_vm_error.txt

[nvazquez]

Thanks @luboss79 sorry for the delay. I do see a problem with a specific VM on the Vmware side while retrieving its nics.

2026-03-19 14:06:50,867 INFO  [c.c.h.v.u.VmwareHelper] (API-Job-Executor-34:[ctx-f1a5194f, job-17995, ctx-7007e9da]) (logid:41abb196) Unable to retrieve guest nics for instance: vm-export-test from VMware tools as tools status: TOOLS_NOT_RUNNING
2026-03-19 14:06:50,874 ERROR [c.c.h.v.u.VmwareHelper] (API-Job-Executor-34:[ctx-f1a5194f, job-17995, ctx-7007e9da]) (logid:41abb196) com.vmware.vim25.VirtualVmxnet3 com.vmware.vim25.VirtualEthernetCardDistributedVirtualPortBackingInfo 00:50:56:99:c9:b7
2026-03-19 14:06:50,875 DEBUG [c.c.h.v.u.VmwareHelper] (API-Job-Executor-34:[ctx-f1a5194f, job-17995, ctx-7007e9da]) (logid:41abb196) NIC com.vmware.vim25.VirtualVmxnet3@58e44ff6 is connected to dvSwitch 7a 80 19 50 4f f6 5b fe-b4 93 a9 a7 91 98 06 3e pg dvportgroup-48651 port 193
2026-03-19 14:06:50,910 INFO  [c.c.h.v.u.VmwareHelper] (API-Job-Executor-34:[ctx-f1a5194f, job-17995, ctx-7007e9da]) (logid:41abb196) Unable to retrieve unmanaged instance nic info. Client received SOAP Fault from server:
Got <criteria> tag when expecting <_this> tag

while parsing call information for method FetchDVPorts
at line 1, column 110

while parsing SOAP body
at line 1, column 102

while parsing SOAP envelope
at line 1, column 38

while parsing HTTP request for method fetchPorts
on object of type vim.DistributedVirtualSwitch
at line 1, column 0 Please see the server log to find more detail regarding exact cause of the failure.

Can you examine the network configuration for the VM vm-export-test on vCenter and check if there are any issues with its nics?

[luboss79]

There is a standard vmware nic available - see the picture. During the export the vm is down, because the export tool doesn't work when the vm is running. Also the vmware tools is installed on the vm.

[luboss79]

We finally fixed this issue by settings the following ownership on the mnt directory on the KVM servers:

sudo chown cloud:cloud /var/cloudstack/mnt/; sudo chmod 755 /var/cloudstack/mnt

The question is why this should be set manually, I would expect that the migration process takes care of this.

Now we have another issues using the export-import tool - discussed in the separate thread #12888