Age | Commit message (Collapse) | Author |
|
This commit removes automatically installing udev rules for hotplug
and adds a module to install them instead.
Automatically including the udev rules and checking if hotplug was
enabled consumed too many resources in certain circumstances. Moving the
rules to a module ensures we don't spend extra extra cycles on hotplug
if hotplug functionality isn't desired.
LP: #1946003
|
|
Adds a udev script which will invoke a hotplug hook script on all net
add events. The script will write some udev arguments to a systemd FIFO
socket (to ensure we have only instance of cloud-init running at a
time), which is then read by a new service that calls a new 'cloud-init
devel hotplug-hook' command to handle the new event.
This hotplug-hook command will:
- Fetch the pickled datsource
- Verify that the hotplug event is supported/enabled
- Update the metadata for the datasource
- Ensure the hotplugged device exists within the datasource
- Apply the config change on the datasource metadata
- Bring up the new interface (or apply global network configuration)
- Save the updated metadata back to the pickle cache
Also scattered in some unrelated typing where helpful
|
|
Cloud-init delivers udev rules on Azure to create the following symlinks:
- /dev/disk/cloud/azure_root
- /dev/disk/cloud/azure_root-part#
- /dev/disk/cloud/azure_resource
- /dev/disk/cloud/azure_resource-part#
Cloud-init cc_disk_setup expects presence of these dev links in order to
setup the mounted ephemeral disks. Gen1 instances udev rules match based
only a DEVICE_ID attribute that no longer exists on Gen2 instances.
Supplement existing Gen1 rules with matches on specitic SCSI target/lun
path 0:0:0 and 0:0:1 and generate links for azure_root and azure_resource
respectively.
LP: #1797480
|
|
This didn't really work. See bug for more info.
LP: #1577844
|
|
Instead of sleep and check loop, use 'udevadm settle' to wait.
since we run from a udev event, this is sufficient.
udev settle will exit when either of
a.) the file exists
b.) the udev event queue has all been processed.
c.) timeout is reached.
Since cloud-init-wait is being run as a udev event, 'b' cannot
be satisfied until it finishes. Thus, this essentially becomes a
inotify based wait for the file /run/cloud-init/network-config-ready
and no loops are needed.
|
|
since we run from a udev event, this is sufficient.
udev settle will exit when either of
a.) the file exists
b.) the udev event queue has all been processed.
Since cloud-init-wait is being run as a udev event, 'b' cannot
be satisfied until it finishes. Thus, this essentially becomes a
inotify based wait for the file /run/cloud-init/network-config-ready
and no loops are needed.
|
|
|
|
I had left this in to commit it, it was my first pass at
cloud-init doing the naming itself. That design was then
replaced with the idea for cloud-init to instead write systemd.rules
files.
|
|
Note, still broken as cloud-init local is not going to ever touch the
CI_NET_READY file (/run/cloud-init/network-config-ready).
So as this is , it will actually just block for 60 seconds and go on.
|
|
|
|
And install them in the Debian packaging.
|