Hi,

Thanks for the report. I'm checking with the XOA team to see if this is a known issue.

Regards, Dan

Yes. In case you missed it, this is from https://docs.xcp-ng.org/installation/upgrade/ :

@User-cxs said:

when I want to shut down a node with VMs on it, the HA will cause the following

We are still waiting for your question.

P.S. The solution is in the GH issue that I previously linked.

@kruess I'm unsure why you are encountering this error. Did you eject all ISOs as shown in the documentation --

Eject CDs from your VMs before upgrading to avoid issues: xe vm-cd-eject --multiple

@kruess I can see numerous instances of VM_HOST_INCOMPATIBLE_VERSION in your xensource log. It is difficult to know the exact cause without more details.

• Was the pool master restarted after the patches were applied?
• Are you trying to start the VM on the master?
• Are you upgrading directly from XS 7.1.2 to XCP 8.3?

@User-cxs You are probably experiencing this issue.

Just a guess because you didn't actually ask a question.

@edsilber said:

you have held broken packages

Have you tried running yum clean and then retrying the install process?

You are running XO from sources, correct? What OS version?

Edit:

Oh, I see you are running a very old deployment of XOA. Deploy a new instance of the appliance and then the package issues will be resolved.

@User-cxs Are you saying that the initial problem was due to the appliance not being registered?

@User-cxs Can you show us the output of ip a on your XOA VM?

@ph7 Do you know which commit you were on previously? Also, it may help if you can share the full backup log.

Hi,

You may want to follow this thread for the VDI not visible issue.

Dan

Just go to Settings > Servers tab in XO and remove the entry associated with this pool.

@tmnguyen You can open a support ticket and request that we reactivate your XOSTOR trial licenses to match your existing XOA trial.

@Bytevenidos You can try that, but it most likely won't work.

I've confirmed with the dev team that the end point to mount an ISO doesn't currently exist for the REST API. However, it does exist in the older JSON-RPC API, which you can access using xo-cli --

vm.insertCd id=<string> cd_id=<string> [force=<boolean>]

@olivierlambert It's even older than that.

https://github.com/vatesfr/xen-orchestra/blob/master/CHANGELOG.v5.md#5540-2020-12-29

https://xen-orchestra.com/blog/xen-orchestra-5-54/

That XOA version was released over 5 years ago. You really should share these kind of details in advance to avoid nonproductive exchanges. I recommend that you revisit the reason that they are "stuck" on this old version.

In current releases of XOA, you can resolve the issue by running a successful backup job that includes the affected VM.

To manually resolve the issue, you need to clear the left over keys in the blocked-operation parameter. For example, this will clear these keys for a VM with UUID be0a9812-fd14-be75-e2fa-40c31ce8875c --

vm_uuid=be0a9812-fd14-be75-e2fa-40c31ce8875c
xe vm-param-remove uuid=$vm_uuid param-name=blocked-operations param-key=migrate_send
xe vm-param-remove uuid=$vm_uuid param-name=blocked-operations param-key=pool_migrate

@Bytevenidos I don't see an endpoint in the REST API for mounting an ISO. However, you should be able to mount the ISO during the VM creation. I will ask around about using the REST API to mount and eject ISOs.

@Bytevenidos Are you up-to-date?

You attach a VDI to a VM by creating a VBD. Take a look at the POST /vbds endpoint.

@allySalami I'm not sure how that could be. You checked the VM's Advanced tab, correct? Are you running XOA or XO from sources? What is your current version / commit?