remount: Also run Before=systemd-hwdb-update.service

[cgwalters]

I saw a race with the readonly sysroot with RHEL CoreOS because
we hadn't mounted /etc writable by the time systemd-hwdb-update.service
ran.

[openshift-ci-robot]

[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

[cgwalters]

One thing I'm not remembering here is why we don't just do all of ostree-remount in the initramfs. Or at least, if the rootfs is already writable, there's no reason to wait.

[cgwalters]

Oh right, it's because ostree-prepare-root.c doesn't link to glib today. It's not at all a problem for those of us shipping an initramfs to do so, but is more of a problem for the no-initramfs statically-linked crowd.

Hmm. I guess we could use a kernel argument, or honestly just don't use GKeyFile and just read the config line by line.

[akiernan]

It's not at all a problem for those of us shipping an initramfs to do so, but is more of a problem for the no-initramfs statically-linked crowd.

/me raises hand!

I guess my only concern with a kernel argument is ensuring it gets in there on upgrade, if folk are hardcoding their boot parameters.

[jlebon]

Hmm, interesting, this is causing Kola failures. E.g.:

--- FAIL: ostree.remote (150.79s)
        harness.go:779: Cluster failed starting machines: machine "3bd322f2-eaff-4351-8594-c8020baa67f3" failed basic checks: some systemd units failed:
systemd-udev-settle.service loaded failed failed udev Wait for Complete Device Initialization

[cgwalters]

Obsoleted by #2113