[pve-devel] [PATCH qemu-server 15/22] vm start/commandline: activate volumes before config_to_command()

Fiona Ebner f.ebner at proxmox.com
Mon Jun 16 15:27:43 CEST 2025


Am 16.06.25 um 15:15 schrieb Fabian Grünbichler:
>> Fiona Ebner <f.ebner at proxmox.com> hat am 16.06.2025 14:57 CEST geschrieben:
>> Am 16.06.25 um 14:46 schrieb Fabian Grünbichler:
>>> variant B:
>>>
>>> move decision which driver to use to plugin via qemu_blockdev_options
>>
>> Yes.
>>
>>> how would the default/fallback look like here? without activation, we
>>> can't rely on path since that doesn't currently require being activated,
>>> so we can't use it to distinguish block devices from files prior to
>>> activation..
>>
>> As written above, we could still try to probe the file like we currently
>> do. If that works, great, we don't lose anything compared to variant A.
>> If that fails, we look at whether path is set. If yes, then assume it's
>> a file-based storage and use the 'file' driver. Otherwise, use the
>> 'host_device' driver.
> 
> but this assumption will be wrong half the time? if we don't activate
> before deciding the driver, then we can only guess that a given path
> is a file, and not a block device? if we have to activate before guessing,
> then we are back at variant A (or at a mixed variant)?

Many plugins do not require activation. And I mean decide based on
$scfg->{path}. Might be a good enough default implementation still.

But I'm also fine with doing the activation up-front. We'll just need to
fix the locking for showcmd.




More information about the pve-devel mailing list