configure-ovs-network: systemctl reload, not restart NetworkManager

[cgwalters]

A full process restart today can trigger bugs; xref
https://bugzilla.redhat.com/show_bug.cgi?id=2077605

NM is also managing the underlying host network state such as DHCP
leases, so blindly restarting it is problematic right now.

Let's try doing a reload instead.

/assign @jcaamano

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: cgwalters

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [cgwalters]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[jcaamano]

When we do systemctl restart NetworkManager, we can then use nm-online -s to wait until NetworkManager has auto-activated all the profiles so we are sure all interfaces are fully configured after that point (otherwise nm-online fails).

If we do systemctl reload NetworkManager, then nm-online -s succeeds immediately and we are not sure whats the configuration state of the interfaces, more than probably their configuration is still ongoing risking that further steps down the pipeline fail.

[cgwalters]

If we do systemctl reload NetworkManager, then nm-online -s succeeds immediately and we are not sure whats the configuration state of the interfaces, more than probably their configuration is still ongoing risking that further steps down the pipeline fail.

OK, but we can monitor the state via DBus right? There's also nmcli m which may be useful in shell script (though this thing really should have long since been in a compiled language).

Alternatively, we could poll nmcli c?

[openshift-ci]

@cgwalters: all tests passed!

Full PR test history. Your PR dashboard.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here.

[jcaamano]

I don't know about dbus. Monitoring nmcli connections or devices is error prone because we don't have a general idea what devices are on the system as virtual devices might not show on that list while disconnected.

Tried some other thing here: #3120

[cgwalters]

Closing in favor of #3120