Multiple errors while adding Esxi virtualmachines to proxmox

[markleakspservice]

Hello everyone, im using proxmox version 8.2 and im receiving multiple errors while I end the EsXi import tool configuration, as I click on the button "add" the first time I received this error "create storage failed: failed to spawn fuse mount, process exited with status 65280 (500)" when i re-try i received another error "Connection error - Timeout" or "Connection error 596: Connection timed out" or "create storage failed: cfs-lock 'file-storage_cfg' error: got lock request timeout (500)", both proxmox and vCentral communicate between each other and are in the same network , i checked the server log on vCentral and i saw that proxmox connects to it and gets the virtual machines but when i switched to proxmox tab i got the aboves errors.

Thanks in advance.

 

[dcsapak]

is there anything in the journal/syslog?

do you by any chance have a datastore with a '+' (or other special characters) in the name?

 

[markleakspservice]

is there anything in the journal/syslog?

do you by any chance have a datastore with a '+' (or other special characters) in the name?

hi, this is what i got when i tried to use the esxi import tool:

May 21 14:00:16 pve systemd[1]: Started pve-esxi-fuse-esxiTest.scope.
May 21 14:01:32 pve esxi-folder-fus[3663]: pve esxi-folder-fuse[3663]: error looking up file or directory: error looking up file or directory: http error code 500
May 21 14:01:32 pve systemd[1]: pve-esxi-fuse-esxiTest.scope: Deactivated successfully.
May 21 14:01:32 pve systemd[1]: run-pve-import-esxi-esxiTest-mnt.mount: Deactivated successfully.
May 21 14:01:42 pve systemd[1]: Started pve-esxi-fuse-esxiTest.scope.

I checked my datastores and none of them had special characters in the name

 

[dcsapak]

May 21 14:01:32 pve esxi-folder-fus[3663]: pve esxi-folder-fuse[3663]: error looking up file or directory: error looking up file or directory: http error code 500

this looks like the vsphere/vcenter has some internal error, maybe you're running into the rate limit? whats your vsphere/vcenter version ?

 

[markleakspservice]

this looks like the vsphere/vcenter has some internal error, maybe you're running into the rate limit? whats your vsphere/vcenter version ?

i checked the version and it is the :

7.0.3.01400​

 

[dcsapak]

can you check if `/run/pve/import/esxi/<storageid>/manifest.json` exists and if yes maybe post it here? CAUTION: it contains a list of vms of the esxi host and it's disks, so anonymize if necessary
what interests me from there are the paths + datastore names if they contain anything that might be a problem

 

[markleakspservice]

can you check if `/run/pve/import/esxi/<storageid>/manifest.json` exists and if yes maybe post it here? CAUTION: it contains a list of vms of the esxi host and it's disks, so anonymize if necessary
what interests me from there are the paths + datastore names if they contain anything that might be a problem

both in Vsphere and Proxmox i dont have the manifest.json, i first tried in proxmox and the directory

/run/pve/import/esxi/

is empty, then i tried on Vsphere and i could not find the complete directory. (Im new in this field)

thanks in advance.

 

[dcsapak]

you could try to manually generate the manifest.json with

/usr/libexec/pve-esxi-import-tools/listvms.py <ip-of-esxi> <username> <path-to-password-file> --skip-cert-verification > /tmp/manifest.json

the manifest.json will then be in /tmp/

 

[markleakspservice]

you could try to manually generate the manifest.json with

/usr/libexec/pve-esxi-import-tools/listvms.py <ip-of-esxi> <username> <path-to-password-file> --skip-cert-verification > /tmp/manifest.json

the manifest.json will then be in /tmp/

Ok, i have it now, but i dont really know what should i anonymize.

thanks in advance

 

[azzii]

I have the same problem. And it is also described in the topic:
https://forum.proxmox.com/threads/esxi-import-timeout.149126

/usr/libexec/pve-esxi-import-tools/listvms.py generate manifest correctly. tcpdump shows active packet exchange with the ESX host even after an error occurs. Judging by the log on the ESX side, proxmox successfully logs in to the host and even receives VMX files.

 

[pedro-udf]

I have the same problem.
Manifest looks fine. I even have the esxi datastores mounted on /run/pve/import/esxi/vmwarec1h3/mnt and can browse the vmdk on the host (only on one node of the cluster)

But I cannot see the "esx storage" on the UI, so I can't start to import VMs.

 

[pedro-udf]

I have the same problem.
Manifest looks fine. I even have the esxi datastores mounted on /run/pve/import/esxi/vmwarec1h3/mnt and can browse the vmdk on the host (only on one node of the cluster)

But I cannot see the "esx storage" on the UI, so I can't start to import VMs.

It's fixed now. I think libpve-storage-perl:amd64 version 8.2.3 fixed this.

 

[azzii]

It's not fixed in libpve-storage-perl:amd64. It also throws an error "Connection error - Timeout."

 

[Primox]

I had the same issue yesterday and my issue is a Datastore on ESXi that fail to connect. Note that no vm are related to this datastore in fact after deleting it all goes well.

 

[Robert.Roygaard]

I still do have the same issue, also removed a failed Datastore from the ESX inventory but that didn't help at all.

 

[dcsapak]

It's not fixed in libpve-storage-perl:amd64. It also throws an error "Connection error - Timeout."

I had the same issue yesterday and my issue is a Datastore on ESXi that fail to connect. Note that no vm are related to this datastore in fact after deleting it all goes well.

I still do have the same issue, also removed a failed Datastore from the ESX inventory but that didn't help at all.

hi,

if you have a similar issue, please add more information to your post, e.g. the exact error message, the logs, and things asked from previous posters, like this:

can you check if `/run/pve/import/esxi/<storageid>/manifest.json` exists and if yes maybe post it here? CAUTION: it contains a list of vms of the esxi host and it's disks, so anonymize if necessary
what interests me from there are the paths + datastore names if they contain anything that might be a problem

and this

you could try to manually generate the manifest.json with

/usr/libexec/pve-esxi-import-tools/listvms.py <ip-of-esxi> <username> <path-to-password-file> --skip-cert-verification > /tmp/manifest.json

the manifest.json will then be in /tmp/

 

[Robert.Roygaard]

I've fixed the issue by deleting an old host that had strange chars in his name. Also make sure to use the ESXi host and NOT the VCenter-Server you use as the WebUI.