netOrk keeps one list of NTP servers and a time zone per site and aligns
a device that differs (NetOrk/netork#2360, #2365). Both drivers now
implement napalm-device-types' TimeSyncMixin (4.35.0) and write both
values.
Neither API keeps time servers: /nodes/{node}/time knows only the zone,
and both products run chrony on Debian. The servers are therefore
PosixTimeSyncMixin's, over the driver's existing SSH session
(run_command, root): read from chrony's configuration, written to
/etc/chrony/sources.d/netork.sources with every other source commented
out, then a reload or a restart of chrony.
The zone is read with them from the host (timedatectl: the zone the
clock really runs in), and from GET /nodes/{node}/time only where the
host names none. It is written through PUT /nodes/{node}/time rather
than timedatectl set-timezone: Proxmox's own call validates the name and
keeps /etc/timezone, which its web UI and API show, in step with
/etc/localtime.
One device is one node: in a PVE cluster every node has its own clock,
and only the node this connection resolved is read and written.
get_ntp_servers() read GET /nodes/{node}/ntp, a path Proxmox VE does not
have, so it always reported no servers. It now reports what
get_time_sync() reads, on the backup server too.
A second driver in this package, ProxmoxBackupDriver. Its API and web UI
are on 8007, not on PVE's 8006, and its node is always localhost; open()
proves the login with GET /version, so a bad token fails there.
Roles: BackupServerDriver (napalm-device-types 4.25.0) first, then
StorageDriver for the disks and ZFS pools the server manages. It reads
datastores, the backup groups of every namespace with their newest
snapshot, a group's snapshots, the GC, verify, prune, sync and tape jobs
with how their last run ended, and the remotes without their secrets;
it starts garbage collection and jobs, verifies, protects and deletes
snapshots, and follows the tasks. As a host it reports facts,
interfaces, CPU and memory, disks, pools, its pbs-subscription and the
same packages, updates, host readers and sshd hardening as a PVE node.
A privilege-separated token without an ACL entry gets an empty list from
PBS, not a refusal, so every read checks /access/permissions and raises
PermissionError. A removable datastore that is not mounted is reported
as unmounted and left out of the group listing.
The fixture is what a PBS 4.2 answered, sanitised; groups, snapshots and
jobs it did not have yet follow the API schema.