Fixes: https://github.com/talos-systems/talos/issues/3219
We already have `etcd leave`, which makes the node exclude itself from
etcd members.
But in case if the node can't remove itself because it doesn't have
connection to etcd we need this etcd remove-member cli, which basically removes
a node from a different node.
No unit tests for that as it's going to destroy the test cluster.
Signed-off-by: Artem Chernyshev <artem.0xD2@gmail.com>
This is required to upgrade from Talos 0.8.x to 0.9.x. After the cluster
is fully upgraded, control plane is still self-hosted (as it was
bootstrapped with bootkube).
Tool `talosctl convert-k8s` (and library behind it) performs the upgrade
to self-hosted version.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This uses API in `os-runtime` to pull the initial list of resources +
updates for resource by type.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This explains the intetion better: config is applied on reboot, and
allows to easily distinguish it from `apply-config --immediate` which
applies config immediately without a reboot (that is coming in a
different PR).
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Also fix recovery grpc handler to print panic stacktrace to the log.
Any API should follow the structure compatible with apid proxying
injection of errors/nodes.
Explicitly fail GenerateConfig API on worker nodes, as it panics on
worker nodes (missing certificates in node config).
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Our upgrades are safe by default - we check etcd health, take locks,
etc. But sometimes upgrades might be a way to recover broken (or
semi-broken) cluster, in that case we need upgrade to run even if the
checks are not passing. This is not a safe way to do upgrades, but it
might be a way to recover a cluster.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
There are several ways Talos node might be restarted or shut down:
* error in sequence (initiated from machined)
* panic in main goroutine (machined recovers panics)
* error in sequence (initiated via API, event caught by machined)
* reboot/shutdown via Talos API
Before this change, paths (1) and (2) were handled in machined, and no
disks were unmounted and processes killed, so technically all the
processes are running and potentially writing to the filesystems.
Paths (3) and (4) try to stop services (but not pods) and unmount
explicitly mounted filesystems, followed by reboot directly from
sequencer (bypassing machined handler).
There was a bug that user disks were never explicitly unmounted (but
they might have been unmounted if mounted on top `/var`).
This refactors all the reboot/shutdown paths to flow through machined's
main function: on paths (4) event is sent via event API from the
sequencer back to the machined and machined initiates proper shutdown
sequence.
Refactoring in machined leads to all the paths (1)-(4) flowing through
the same function `handle(error)`.
Added two additional checks before flushing buffers:
* kill all non-system processes, this also kills all mount namespaces
* unmount any filesystem backed by `/dev/*`
This ensures all filesystems are unmounted before buffers are flushed.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Control plane components are running as static pods managed by the
kubelets.
Whole subsystem is managed via resources/controllers from os-runtime.
Many supporting changes/refactoring to enable new code paths.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This brings in `os-runtime` package and exposes resources with first
iteration of read-only API.
Two Talos resources (and one controller) are implemented:
* legacy.Service resource tracks Talos 'service' `RUNNING` state
* config.V1Alpha1 stores current runtime config
Glue point between existing runtime and new os-runtime based runtime is
in `v1alpha2` implementation and `V1Alpha2()` sub-interfaces of existing
`Runtime`, `State`, `Controller` interfaces.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Before these changes, errors were always sent as strings, so if original
error was gRPC error (which is almost always the case for apid), it is
formatted as string and original fields (like code) are lost in the
formatted string.
With this change, apid sends errors as official `grpc.Status` protobuf
structure, and client decodes that into Go grpc.Status based error.
This change is backwards and forwards compatible.
This should fix more cases when integration tests were not able to
ignore grpc `transport is closing` errors when they were sent as strings
from the apid endpoint.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Idea is to add an option to perform "selective" reset: default reset
operation is to wipe all partitions (triggering reinstall), while spec
allows only to wipe some of the operations.
Other operations are performed exactly in the same way for any reset
flow.
Possible use case: reset only `EPHEMERAL` partition.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Regular upgrade path takes just one reboot, but it requires all the
processes to be stopped on the node before upgrade might proceed. Under
some circumstances and with potential Talos bugs it might not work
rendering Talos upgrades almost impossible.
Staged upgrades build upon regular install flow to run the upgrade on
the node reboot. Such upgrades require two reboots of the node, and it
requires two pulls of the installer image, but they should be much less
suspicious to the failure. Once the upgrade is staged, node can be
rebooted in any possible way, including hard reset and upgrade is
performed on the next boot.
New ADV format was implemented as well to allow to store install image
ref/options across reboots. New format allows for bigger values and
takes 50% of the `META` partition. Old ADV is still kept for
compatibility reasons.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
If the node time is out of sync, it can generate incorrect
configuration. And maintenance mode does not allow us starting ntp,
because there is no containerd.
By providing current UTC time of the machine where talosctl client is
running, it is possible to force GenerateConfiguration use correct time.
Signed-off-by: Artem Chernyshev <artem.0xD2@gmail.com>
Initial version which only allows setting CNI using preset, no custom
CNI urls are supported at the moment. Still need to figure out what kind
of UI can be used for that.
Signed-off-by: Artem Chernyshev <artem.0xD2@gmail.com>
This allows config to be written to disk without being applied
immediately.
Small refactoring to extract common code paths.
At first, I tried to implement this via the sequencer, but looks like
it's too hard to get it right, as sequencer lacks context and config to
be written is not applied to the runtime.
Fixes#2828
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Fixes: https://github.com/talos-systems/talos/issues/2766
This API is implemented in Maintenance and Machine services.
Can be used to generate configuration on the node, instead of using
talosctl to generate it locally.
To be used in interactive installer and talosctl gen config.
Signed-off-by: Artem Chernyshev <artem.0xD2@gmail.com>
Now maintenance service implements `MachineService` interface, stubbing
all not implemented methods.
Signed-off-by: Artem Chernyshev <artem.0xD2@gmail.com>
Fixes#2761
Service `osd` was merged into machined on Jul, 13th, before 0.6 release.
It's time to drop the backwards compatibility with clients before 0.6.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Instead of hosting a web service, we decided to implement a gRPC service
that exposes APIs that can be used in a client-side interactive installer.
Signed-off-by: Andrew Rynhard <andrew@rynhard.io>
Adds the ability to apply (replace) an existing node configuration with
a new one via the Machine API.
Fixes#2345
Signed-off-by: Seán C McCord <ulexus@gmail.com>
This moves `pkg/config`, `pkg/client` and `pkg/constants`
under `pkg/machinery` umbrella.
And `pkg/machinery` is published as Go module inside Talos repository.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This prevents reboots when some actions triggers sequence while another
sequence is still running.
Fixes#2209
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This builds a simple CLI UI for Talos cluster monitoring.
Some new APIs were added for monitoring based on Prometheus procfs
package.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This implements existing server-side health checks as defined in
`internal/pkg/cluster/checks` in Talos API.
Summary of changes:
* new `cluster` API
* `apid` now listens without auth on local file socket
* `cluster` API is for now implemented in `machined`, but we can move it
to the new service if we find it more appropriate
* `talosctl health` by default now does server-side health check
UX: `talosctl health` without arguments does health check for the
cluster if it has healthy K8s to return master/worker nodes. If needed,
node list can be overridden with flags.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This merges `osd` API into `machined`. API was copied from `osd` into
`machined`, and `osd` API was deprecated.
For backwards compatibility, `machined` still implements `osd` API, so
older Talos API clients can still talk to the node without changes.
Docs were updated. No functional changes.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
1. Add [xid-based](https://github.com/rs/xid) event IDs. Xids
are sortable and unique enough. Xids also encode event publishing
time with a second precision.
2. Add three ways to look back into event history: based on number of
events, on time and ID. Lookup via ID might be used to restart event
polling in case of broken API connection from the same moment.
3. Reimplement core event buffer with positions which are always
incremented instead of generation+index, this implementation is much
more simple (idea from circular buffer).
4. By default, Events API works the same - it shows no history and
starts streaming new events only.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This implements service events, adds test for events API based on
service events as they're the easiest to generate on demand.
Disabled validate test for 'metal' as it validates disk device against
local system which doesn't make much sense.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
Streaming APIs are not supposed to wrap response into `repeated`
container, as streaming allows to send as many responses back as
required.
Signed-off-by: Andrey Smirnov <smirnov.andrey@gmail.com>
This adds the ability to bootstrap a cluster using the API.
The API simply starts the bootkube service.
Signed-off-by: Andrew Rynhard <andrew@andrewrynhard.com>
This is a rewrite of machined. It addresses some of the limitations and
complexity in the implementation. This introduces the idea of a
controller. A controller is responsible for managing the runtime, the
sequencer, and a new state type introduced in this PR.
A few highlights are:
- no more event bus
- functional approach to tasks (no more types defined for each task)
- the task function definition now offers a lot more context, like
access to raw API requests, the current sequence, a logger, the new
state interface, and the runtime interface.
- no more panics to handle reboots
- additional initialize and reboot sequences
- graceful gRPC server shutdown on critical errors
- config is now stored at install time to avoid having to download it at
install time and at boot time
- upgrades now use the local config instead of downloading it
- the upgrade API's preserve option takes precedence over the config's
install force option
Additionally, this pulls various packes in under machined to make the
code easier to navigate.
Signed-off-by: Andrew Rynhard <andrew@andrewrynhard.com>
This PR introduces a new strategy for upgrades. Instead of attempting to
zap the partition table, create a new one, and then format the
partitions, this change will only update the `vmlinuz`, and
`initramfs.xz` being used to boot. It introduces an A/B style upgrade
process, which will allow for easy rollbacks. One deviation from our
original intention with upgrades is that this change does not completely
reset a node. It falls just short of that and does not reset the
partition table. This forces us to keep the current partition scheme in
mind as we make changes in the future, because an upgrade assumes a
specific partition scheme. We can improve upgrades further in the
future, but this will at least make them more dependable. Finally, one
more feature in this PR is the ability to keep state. This enables
single node clusters to upgrade since we keep the etcd data around.
Signed-off-by: Andrew Rynhard <andrew@andrewrynhard.com>
This introduces a health/ready api for networkd. This
will allow us to better determine the state of networkd
and allow for some level of monitoring.
Signed-off-by: Brad Beam <brad.beam@talos-systems.com>
This PR brings our protobuf files into conformance with the protobuf
style guide, and community conventions. It is purely renames, along with
generated docs.
Signed-off-by: Andrew Rynhard <andrew@andrewrynhard.com>
