On 11/6/26 1:16 PM, Ben Grande wrote:
On 26-10-21 15:37:58, Ben Grande wrote:
Report the new VIR_DOMAIN_EVENT_STOPPED_REBOOT reason when reboot has
been requested. I have only tested this on Xen, so I made a patch per
VMM. I preferred to rely on the stopped reason, as there is no
discussion that it should be sent, while for the reboot event, there is
debate that it shouldn't always be sent.
CI:
https://gitlab.com/libvirt/libvirt/-/merge_requests/567
Previous discussion and patches:
https://lists.libvirt.org/archives/list/[email protected]/thread/ZB5ALEFMTONTYZMJ6R2H6JWTOFL4ETER/
https://lists.libvirt.org/archives/list/[email protected]/thread/KYAL45C54LYSVUITZIEOLAQZU4GBVRQA/
Ben Grande (9):
libvirt-domain: Add stopped reason of reboot
This patch doesn't build until 3 and 7 are applied, which breaks bisection. IMO
patches 1, 3 and 7 should be squashed into 1 patch, the driver changes into
another, and finally 1 NEWS patch with a single entry.
docs: libvirt-domain: briefly document reboot event
This patch is not correct since VIR_DOMAIN_EVENT_ID_REBOOT is no longer emitted
by the series. It should mention when 'on_reboot' isn't 'restart', drivers that
support it will emit a VIR_DOMAIN_EVENT_STOPPED lifecycle event with detail
VIR_DOMAIN_EVENT_STOPPED_REBOOT.
virsh: Add reboot to event enum
libxl: Propagate stopped reason of reboot
lxc_process: Propagate stopped reason of reboot
I'm not sure about this change. Will comment in the patch.
Regards,
Jim
