| Commit message (Collapse) | Author | Age | Files | Lines |
|---|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Move isc-dhclient code to dhcp.py
In support of the upcoming deprecation of
isc-dhcp-client, this code refactors current
dhcp code into classes in dhcp.py. The
primary user-visible change should be the
addition of the following log:
dhcp.py[DEBUG]: DHCP client selected: dhclient
This code lays groundwork to enable
alternate implementations to live side by
side in the codebase to be selected with
distro-defined priority fallback. Note that
maybe_perform_dhcp_discovery() now selects
which dhcp client to call, and then runs the
corresponding client's dhcp_discovery()
method. Currently only class IscDhclient is
implemented, however a yet-to-be-implemented
class Dhcpcd exists to test fallback behavior
and this will be implemented in part two of
this series.
Part of this refactor includes shifting
dhclient service management from hardcoded
calls to the distro-defined manage_service()
method in the *BSDs. Future work is required
in this area to support multiple clients via
select_dhcp_client().
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Commit d1ffbea556a06105 enabled skipping python datasource detection on
OpenStack when no other datasources (besides DataSourceNone) can be discovered.
This allowed one to override detection, which is a requirement for OpenStack
Ironic which does not advertise itself to cloud-init.
Since no further datasources can be detected at this stage in the code, this
pattern can be generalized to other datasources to facilitate troubleshooting
or providing a general workaround to runtime detection bugs.
Additionally, this pattern can be extended to kernel commandline datasource
definition. Since kernel commandline is highest priority of the
configurations, it makes sense to override python code datasource
detection as well.
Include an integration test on LXD for this behavior that configures kernel
commandline and reboots to verify that the specified datasource is forced.
|
| |
|
|
|
|
|
|
|
|
|
| |
Bump system_cfg over ds network_config_source for Oracle DS,
so that if network config is defined under /etc/cloud,
it will be honored.
In a previous change, we moved the initramfs and system_cfg
bellow ds to favor ds, but this implied system-wide configs
were always not honored.
LP: #1956788
|
| |
|
| |
Usage was dropped in de7851b93c5a2d4658.
|
| |
|
|
|
|
|
|
| |
If /run/net* files aren't available, we use IMDS for configuring the
default interface. Rather than attempt a static configuration, grab
the MAC and let DHCP do the rest (as it does when /run/net*
is available).
LP: #1989686
|
| |
|
|
|
|
|
|
|
| |
Previous to 6270b50, if iSCSI config was not available, we used the
ephemeral DHCP4 address as the primary address on Oracle. After
6270b50, we instead used the IMDS address configuration. However, the
parsing of IMDS ignored the "subnetCidrBlock" field, causing the
resulting network config to have the wrong subnet.
LP: #1989686
|
| |
|
|
|
|
|
|
| |
In the case cloudinit.temp_utils points to a fs mounted as noexec
and needs_exe=True, fallback to use
os.join.path(Distro.usr_lib_exec, "cloud-init/clouddir) that
will be mounted with exec perms.
LP: #1962343
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Also refactor network context managers into net.ephemeral
Currently EC2 is the only IMDS to make use of this.
IPv6 requires a link local address on interfaces. A
link local address is sufficient for the EC2 IMDS,
so no dhcp6 assignment is required for early boot
IMDS queries.
The kernel assigns this address using RFC 4291 [1]
during link initialization, so all cloud-init needs
to do is ensure that link is up.
This means that even if dhcp4 fails, an ipv6-enabled
instance may still succeed at crawling metadata.
[1] https://datatracker.ietf.org/doc/html/rfc4291#section-2.5.6
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
For primary network config:
- Use `iSCSI` config if some `/run/net*` file exists, even if
`/run/initramfs/open-iscsi.interface` does not.
- If the instance is not an `iSCSI` one, then crawl the network
config from `IMDS` instead of falling back to "best guess".
- Remove unnecessary conditional use of dhcp.EphemeralDHCPv4
and use it always to crawl `IMDS`.
- Migrate tests to pytest.
- Extend unit test coverage.
- Add some types for mypy.
LP: #1967942
|
| |
|
|
|
|
|
|
| |
It was implemented as a namedtuple, because it was written
when the codebase supported Python 2 (where using an enum would have
introduced a new dependency). As enum is in the stdlib in all our
supported Python releases, we can now use it without that constraint.
LP: #1874875
|
| |
|
|
|
|
|
|
| |
This reverts commit b306633fd17e5ba0173ad3c41add59cb11884757.
While this ultimately seems like a better solution, currently the
file /etc/cloud/cloud.cfg.d/99-disable-network-config.cfg exists on
all Oracle launched instances which will prevent networking from
being properly initialized.
|
| |
|
|
|
|
|
|
|
|
|
| |
In 2c52e6e88b19f5db8d55eb7280ee27703e05d75f, the order of
reading network config was changed for Oracle due to initramfs
needing to take lower precedence than the datasource. However,
this also bumped system_cfg to a lower precedence than ds, which
means that any network configuration specified in /etc/cloud will not
be applied. system_cfg should instead be moved above ds so network
configuration in /etc/cloud takes precedence.
LP: #1956788
|
| |
|
|
|
| |
Applied Black and isort, fixed any linting issues, updated tox.ini
and CI.
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
Add connectivity_url to Oracle's EphemeralDHCPv4
On bionic, when trying to bring up the EphemeralDHCPv4, it's possible
that we already have a route defined, which will result in an error when
trying to add the DHCP route. Use the connectivity_url to check if we
can reach the metadata service, and if so, skip the EphemeralDHCPv4.
The has_url_connectivity function has also been modified to take
a dict of kwargs to send to readurl.
LP: #1939603
|
| |
|
|
|
|
|
|
|
| |
This just separates the reading of dmi values into its own file.
Some things of note:
* left import of util in dmi.py only for 'is_container'
It'd be good if is_container was not in util.
* just the use of 'util.is_x86' to dmi.py
* open() is used directly rather than load_file.
|
| | |
|
| |
|
|
| |
* v2 of the API is now default with fallback to v1.
* Refactored the Oracle datasource to fetch version, instance, and vnic metadata simultaneously.
|
| |
|
|
|
|
|
|
|
| |
The /opc/v1/ metadata endpoints[0] are universally available in Oracle
Cloud Infrastructure and the OpenStack endpoints are considered
deprecated, so we can refactor the data source to use the OPC endpoints
exclusively. This simplifies the datasource code substantially, and
enables use of OPC-specific attributes in future.
[0] https://docs.cloud.oracle.com/en-us/iaas/Content/Compute/Tasks/gettingmetadata.htm
|
| |
|
|
|
| |
* test_oracle: sort imports
* DataSourceOracle: sort imports
|
| |
|
|
| |
Bring it into line with the current state of the code (by dropping the
now-untrue caveat).
|
| |
|
|
|
|
|
|
|
|
| |
Add support for detecting netfailover[1] device 3-tuple in networking
layer. In the Oracle datasource ensure that if a provided network
config, either fallback or provided config includes a netfailover master
to remove any MAC address value as this can break under 3-netdev
as the other two devices have the same MAC.
1. https://www.kernel.org/doc/html/latest/networking/net_failover.html
|
| |
|
|
|
|
| |
When rendering secondary vnic configuration from IMDS, only emit
configuration for the IP and MTU values only. Add support to mutate
either a v1 or a v2 network_config input.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The Oracle platform provides networking configuration from two sources:
* the primary interface configuration comes from the initramfs, because
Oracle instance all iSCSI boot
* secondary interface configuration comes from an IMDS accessed over
HTTP
As we need to combine these two sources of network configuration, the
default "prefer initramfs config over data source config" behaviour
isn't appropriate; we would never get the IMDS interfaces via that
route. Instead, the Oracle data source has code to combine these two
sources, so we prefer its network configuration over the initramfs
configuration.
(This is not appropriate default behaviour, because _in general_ data
sources won't know how to merge initramfs-provided configuration into
their provided configuration, so switching this order for all data
sources would result in initramfs configuration being discarded on any
data source that implements network_config.)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Oracle Cloud Infrastructure's Instance Metadata Service provides network
configuration information for non-primary NICs. This commit introduces
support, on Virtual Machines[0], for fetching that network metadata,
converting it to v1 network-config[1] and combining it into the network
configuration generated for the primary interface.
By default, this behaviour is not enabled. Configuring the Oracle
datasource to `configure_secondary_nics` enables it:
datasource:
Oracle:
configure_secondary_nics: true
Failures to fetch and generate secondary NIC configuration will log a
warning, but otherwise will not affect boot.
[0] The expected use of the IMDS-provided network configuration is
substantially different on Bare Metal Machines, so support for that
will be addressed separately.
[1] This is v1 config, because cloudinit.net.cmdline generates v1 config
and we need to integrate the secondary NICs into that configuration.
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Previously "cmdline" network configuration could be either
user-specified network-config=... configuration data, or
initramfs-provided configuration data. Before data sources could modify
the order in which network config sources were considered, this
conflation didn't matter (and, indeed, in the default data source
configuration it will continue to not matter).
However, it _is_ desirable for a data source to be able to specify that
its network configuration should be preferred over the
initramfs-provided network configuration but still allow explicit
network-config=... configuration passed to the kernel cmdline to
continue to override both of those sources.
(This also modifies the Oracle data source to use read_initramfs_config
directly, which is effectively what it was using
read_kernel_cmdline_config for previously.)
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Add the following instance-data.json standardized keys:
* v1._beta_keys: List any v1 keys in beta development,
e.g. ['subplatform'].
* v1.public_ssh_keys: List of any cloud-provided ssh keys for the
instance.
* v1.platform: String representing the cloud platform api supporting the
datasource. For example: 'ec2' for aws, aliyun and brightbox cloud
names.
* v1.subplatform: String with more details about the source of the
metadata consumed. For example, metadata uri, config drive device path
or seed directory.
To support the new platform and subplatform standardized instance-data,
DataSource and its subclasses grew platform and subplatform attributes.
The platform attribute defaults to the lowercase string datasource name at
self.dsname. This method is overridden in NoCloud, Ec2 and ConfigDrive
datasources.
The subplatform attribute calls a _get_subplatform method which will
return a string containing a simple slug for subplatform type such as
metadata, seed-dir or config-drive followed by a detailed uri, device or
directory path where the datasource consumed its configuration.
As part of this work, DatasourceEC2 methods _get_data and _crawl_metadata
have been refactored for a few reasons:
- crawl_metadata is now a read-only operation, persisting no attributes on
the datasource instance and returns a dictionary of consumed metadata.
- crawl_metadata now closely represents the raw stucture of the ec2
metadata consumed, so that end-users can leverage public ec2 metadata
documentation where possible.
- crawl_metadata adds a '_metadata_api_version' key to the crawled
ds.metadata to advertise what version of EC2's api was consumed by
cloud-init.
- _get_data now does all the processing of crawl_metadata and saves
datasource instance attributes userdata_raw, metadata etc.
Additional drive-bys:
* unit test rework for test_altcloud and test_azure to simplify mocks
and make use of existing util and test_helpers functions.
|
|
|
This adds a Oracle specific datasource that functions with OCI.
It is a simplified version of the OpenStack metadata server
with support for vendor-data.
It does not support the OCI-C (classic) platform.
Also here is a move of BrokenMetadata to common 'sources'
as this was the third occurrence of that class.
|
