On 3/18/20 11:26 AM, Mira Limbeck wrote:
>
>> Thomas Lamprecht <t.lampre...@proxmox.com> hat am 18. März 2020 10:32
>> geschrieben:
>>
>>
>> On 3/18/20 10:11 AM, Fabian Grünbichler wrote:
>>>>> @@ -594,10 +597,16 @@ sub phase2 {
>>>>> }
>>>>>
>>>>> my $spice_port;
>>>>> + my $tunnel_addr = [];
>>>>> + my $sock_addr = [];
>>>>> + # version > 0 for unix socket support
>>>>> + my $nbd_protocol_version = 1;
>>>>> + my $input = "nbd_protocol_version: $nbd_protocol_version\n";
>>>>> + $input .= "spice_ticket: $spice_ticket\n" if $spice_ticket;
>>>> I know it's already been applied, but this breaks new->old migration
>>>> (with SPICE) for no reason, doesn't it? I know we don't always support
>>>> it, but I don't see why we need the break here..
>>>>
>>>> I.e. if we just send the spice-ticket w/o a prefix an old node will be
>>>> perfectly happy and just send us back the TCP migration URI (and with
>>>> the fallback in place a new node will also be happy), but if we do
>>>> prefix it the remote will pick up the entire "spice_ticket: xxx" as the
>>>> ticket.
>>> yes, we could revert that back to the old behaviour and add a FIXME: add
>>> prefix 'spice_ticket: ' with 7.0?
>>>
>>
>> yeah, please. And nice work catching this Stefan!
> Yes, looked over it again and again. Even tested live migration with spice
> and virt-viewer, but in this case probably missed the new -> old migration.
> Should have waited till today before sending it.
>
I'd like to bump qemu-server soon and am missing from the pile of known issues,
can you please take a look at this (tomorrow)? That'd be great :)
_______________________________________________
pve-devel mailing list
pve-devel@pve.proxmox.com
https://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
