Am 03.03.21 um 17:53 schrieb Thomas Lamprecht:
On 03.03.21 12:01, Fabian Ebner wrote:
as otherwise the empty string is printed with 'qm status --verbose' when
it's not a template.
Signed-off-by: Fabian Ebner
---
PVE/QemuServer.pm | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff
Am 04.03.21 um 09:04 schrieb Fabian Ebner:
Am 03.03.21 um 17:53 schrieb Thomas Lamprecht:
On 03.03.21 12:01, Fabian Ebner wrote:
as otherwise the empty string is printed with 'qm status
--verbose' when
it's not a template.
Signed-off-by: Fabian Ebner
---
PVE/QemuServer.pm | 2 +-
1 file
LGTM, for both qemu-server patches:
Reviewed-by: Stefan Reiter
Not sure about the formatting in the GUI, but I'm also the wrong person
to ask there. Maybe don't capitalize "Qemu", as we also don't do that
for "running"/"stopped"/... either?
On 01/03/2021 16:53, Fabian Ebner wrote:
to be re
Hidden behind "Advanced" options, as to not confuse inexperienced users.
Signed-off-by: Stefan Reiter
---
www/manager6/qemu/MachineEdit.js | 69
1 file changed, 69 insertions(+)
diff --git a/www/manager6/qemu/MachineEdit.js b/www/manager6/qemu/MachineEdit.js
ind
no functional change intended
Signed-off-by: Stefan Reiter
---
www/manager6/Makefile | 1 +
www/manager6/qemu/HardwareView.js | 16 +
www/manager6/qemu/MachineEdit.js | 38 +++
3 files changed, 40 insertions(+), 15 deletions(-)
create mode 1
The file is provided by pve-qemu-kvm.
Signed-off-by: Stefan Reiter
---
PVE/API2/Qemu/Machine.pm | 49
PVE/API2/Qemu/Makefile | 2 +-
2 files changed, 50 insertions(+), 1 deletion(-)
create mode 100644 PVE/API2/Qemu/Machine.pm
diff --git a/PVE/API2/Qe
as 'machine-types', so it is clear this refers to QEMU machines, not the
local machine (as one might think, this being a 'node' API call).
Signed-off-by: Stefan Reiter
---
Requires dependency bump on updated qemu-server.
PVE/API2/Nodes.pm | 6 ++
1 file changed, 6 insertions(+)
diff --git
Same rationale as the CPU flags file, avoids calling QEMU binary just to
query machine types.
Signed-off-by: Stefan Reiter
---
debian/parse-machines.pl | 21 +
debian/rules | 4 +++-
2 files changed, 24 insertions(+), 1 deletion(-)
create mode 100755 debian/pars
...for QEMU 5.2
Add a postinst with some inline perl (to avoid having to install a
seperate file to run) that conservatively does its best to set the
machine version for all VMs the bug might affect to 5.1.
Signed-off-by: Stefan Reiter
---
Requires Fabian's #3301 series:
https://lists.proxmox.c
Signed-off-by: Stefan Reiter
---
This is required to make the whole deal work, otherwise even setting back the
machine version to 5.1 or lower won't actually fix the issue.
...restore-device-paths-for-pre-5.1-vms.patch | 108 ++
debian/patches/series |
...and make machine version configurable in the GUI, to allow users to return to
the default behaviour of always using the latest one.
The workaround is based on a postinst script running when upgrading the
qemu-server package, which sets the machine version of all effected Windows VMs
to pc-(i440
On 01.03.21 16:53, Fabian Ebner wrote:
> to be re-used in the vmstatus() call.
>
> Signed-off-by: Fabian Ebner
> ---
> PVE/QemuServer/Machine.pm | 16 +++-
> 1 file changed, 11 insertions(+), 5 deletions(-)
>
>
applied, with Stefan's R-b, thanks!
_
On 01.03.21 16:53, Fabian Ebner wrote:
> Signed-off-by: Fabian Ebner
> ---
> PVE/QemuServer.pm | 32
> 1 file changed, 32 insertions(+)
>
>
applied, with Stefan's R-b, thanks!
___
pve-devel mailing list
pve-devel@lis
I suppose we should also detect old QEMU versions, so we don't set 5.1
machine versions for QEMU < 5.1. Unlikely, but easy to detect:
print "QEMU version too old for workaround" if
!min_version(PVE::QemuServer::kvm_user_version(), 5, 1);
or similar... I can send a v2 if required.
On 04/0
Didn't follow this issue too closely, but what about restoring a
pre-QEMU-ACPI-patch Windows VM backup in a post-QEMU-ACPI-patch world?
Those would need to get pinned as well, right? Doing the pinning itself
on restore shouldn't be difficult, but a good way to detect when it's
needed is needed.
15 matches
Mail list logo
