Tested this series by: - manually adding a resource to the config and running `ha-manager relocate ...` right after. With the patches applied, this no longer fails - creating a VM through `qm create <vmid> --ha-managed 1` - running `qmrestore` with the --start flag explicitly set to 1 or 0 or omitting it - running `qmrestore` with the --ha-managed flag explicitly set to 1 or 0 or omitting it - creating a CT through `pct create <vmid> <ostemplate> --ha-managed 1`
The changes made all of the above work as expected. Thanks for fixing these! Also had a look at the code and the changes lgtm! Consider this: Tested-by: Michael Köppl <[email protected]> Reviewed-by: Michael Köppl <[email protected]> On Mon Nov 10, 2025 at 3:36 PM CET, Daniel Kral wrote: > Some small fixes and improvements while playing around with the > HA-on-create feature related mainly to the CLI usage. Two qemu-server > patches only enhance the usage by allowing --ha-managed and --start > parameters for qmrestore. > > pve-ha-manager.git: > > Daniel Kral (1): > api: resources: fix uninitialized value in check_service_state > > src/PVE/API2/HA/Resources.pm | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > > qemu-server.git: > > Daniel Kral (3): > api: create_vm: use ha-manager command to add VM as an HA resource > qmrestore: allow starting a VM after it was restored successfully > qmrestore: allow adding a VM as an HA resource after it was restored > > src/PVE/API2/Qemu.pm | 10 ++++------ > src/PVE/CLI/qmrestore.pm | 12 ++++++++++++ > 2 files changed, 16 insertions(+), 6 deletions(-) > > > pve-container.git: > > Daniel Kral (1): > api: create_vm: use ha-manager command to add container as an HA > resource > > src/PVE/API2/LXC.pm | 4 +--- > 1 file changed, 1 insertion(+), 3 deletions(-) > > > Summary over all repositories: > 4 files changed, 22 insertions(+), 10 deletions(-) _______________________________________________ pve-devel mailing list [email protected] https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
