gave this series another spin in my test cluster and so far it seems to
work as described.
I did a very quick test if we see some noticeable "lag" when destroying
a guest by filling the `used_vmids.list` with IDs from 100..30000 and a
stepsize of 2, so close to 15k lines. But I didn't notice that it would
take longer.
A few small style nits are addressed, in the specific patches. But
nothing severe that would justify a v4 right away.
Consider this series:
Tested-By: Aaron Lauterer <a.laute...@proxmox.com>
Reviewed-By: Aaron Lauterer <a.laute...@proxmox.com>
On 2024-11-05 02:58, Severen Redwood wrote:
Hi everyone,
This is a small update to my previous patch series [1, 2]
adding optional support for preventing PVE from suggesting
previously used VM/CT IDs.
In particular, missing use statements have been added to the
patches for marking VM/CT IDs as used, as noticed by Aaron
[3]. Everything else remains the same as in v2.
Also note when building this patch series that the pve-cluster
patches must be applied and built first to prevent the build
of pve-manager from failing.
If anyone has any other feedback, please let me know.
Thanks,
Severen
[1]:
https://lore.proxmox.com/pve-devel/mailman.177.1728018639.332.pve-de...@lists.proxmox.com/
[2]:
https://lore.proxmox.com/pve-devel/mailman.624.1730255368.332.pve-de...@lists.proxmox.com/T/#t
[3]:
https://lore.proxmox.com/pve-devel/mailman.624.1730255368.332.pve-de...@lists.proxmox.com/T/#m7313a65f5b26f29ff0a2b4005069a7306311ccf6
_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
