* Originally, Kubelet and CAdvisor metrics used a labelmap
relabel to add Kubernetes SD node labels onto timeseries
* With https://github.com/poseidon/typhoon/pull/596 that
relabel was dropped since node labels aren't usually that
valuable. `__meta_kubernetes_node_name` was retained but
the field name is empty
* Favor just using Prometheus server-side `instance` in
queries that require some node identifier for aggregation
or debugging
Fix https://github.com/poseidon/typhoon/issues/823
* seccomp graduated to GA in Kubernetes v1.19. Support for
seccomp alpha annotations will be removed in v1.22
* Replace seccomp annotations with the GA seccompProfile
field in the PodTemplate securityContext
* Switch profile from `docker/default` to `runtime/default`
(no effective change, since docker is the runtime)
* Verify with docker inspect SecurityOpt. Without the profile,
you'd see `seccomp=unconfined`
Related: https://github.com/poseidon/terraform-render-bootstrap/pull/215
* Fix race condition for bootstrap-secrets SELinux context on non-bootstrap controllers in multi-controller FCOS clusters
* On first boot from disk on non-bootstrap controllers, adding bootstrap-secrets races with kubelet.service starting, which can cause the secrets assets to have the wrong label until kubelet.service restarts (service, reboot, auto-update)
* This can manifest as `kube-apiserver`, `kube-controller-manager`, and `kube-scheduler` pods crashlooping on spare controllers on first cluster creation
* Mention the first master branch SHA that introduced Terraform
v0.13 forward compatibility
* Link the migration guide on Github until a release is available
and website docs are published
* Fedora CoreOS now ships systemd-udev's `default.link` while
Flannel relies on being able to pick its own MAC address for
the `flannel.1` link for tunneled traffic to reach cni0 on
the destination side, without being dropped
* This change first appeared in FCOS testing-devel 32.20200624.20.1
and is the behavior going forward in FCOS since it was added
to align FCOS network naming / configs with the rest of Fedora
and address issues related to the default being missing
* Flatcar Linux (and Container Linux) has a specific flannel.link
configuration builtin, so it was not affected
* https://github.com/coreos/fedora-coreos-tracker/issues/574#issuecomment-665487296
Note: Typhoon's recommended and default CNI provider is Calico,
unless `networking` is set to flannel directly.
* Set Kubelet cgroup driver to systemd when Flatcar Linux edge
is chosen
Note: Typhoon module status assumes use of the stable variant of
an OS channel/stream. Its possible to use earlier variants and
those are sometimes tested or developed against, but stable is
the recommendation
* In v1.18.3, the `os_stream` variable was added to select
a Fedora CoreOS image stream (stable, testing, next) on
AWS and Google Cloud (which publish official streams)
* Remove `os_image` variable deprecated in v1.18.3. Manually
uploaded images are no longer needed
* DigitalOcean introduced Virtual Private Cloud (VPC) support
to match other clouds and enhance the prior "private networking"
feature. Before, droplet's belonging to different clusters (but
residing in the same region) could reach one another (although
Typhoon firewall rules prohibit this). Now, droplets in a VPC
reside in their own network
* https://www.digitalocean.com/docs/networking/vpc/
* Create droplet instances in a VPC per cluster. This matches the
design of Typhoon AWS, Azure, and GCP.
* Require `terraform-provider-digitalocean` v1.16.0+ (action required)
* Output `vpc_id` for use with an attached DigitalOcean
loadbalancer
* Change the Calico VXLAN interface for MTU from 1450 to 1410
* VXLAN on Azure should support MTU 1450. However, there is
history where performance measures have shown that 1410 is
needed to have expected performance. Flatcar Linux has the
same MTU 1410 override and note
* FCOS 31.20200323.3.2 was known to perform fine with 1450, but
now in 31.20200517.3.0 the right value seems to be 1410
* Fedora CoreOS `kubelet.service` can start before the hostname
is set. Kubelet reads the hostname to determine the node name to
register. If the hostname was read as localhost, Kubelet will
continue trying to register as localhost (problem)
* This race manifests as a node that appears NotReady, the Kubelet
is trying to register as localhost, while the host itself (by then)
has an AWS provided hostname. Restarting kubelet.service is a
manual fix so Kubelet re-reads the hostname
* This race could only be shown on AWS, not on Google Cloud or
Azure despite attempts. Bare-metal and DigitalOcean differ and
use hostname-override (e.g. afterburn) so they're not affected
* Wait for nodes to have a non-localhost hostname in the oneshot
that awaits /etc/resolve.conf. Typhoon has no valid cases for a
node hostname being localhost (not even single-node clusters)
Related Openshift: https://github.com/openshift/machine-config-operator/pull/1813
Close https://github.com/poseidon/typhoon/issues/765
* Remove node label `node.kubernetes.io/master` from controller nodes
* Use `node.kubernetes.io/controller` (present since v1.9.5,
[#160](https://github.com/poseidon/typhoon/pull/160)) to node select controllers
* Rename controller NoSchedule taint from `node-role.kubernetes.io/master` to
`node-role.kubernetes.io/controller`
* Tolerate the new taint name for workloads that may run on controller nodes
and stop tolerating `node-role.kubernetes.io/master` taint
* Kubelet `--lock-file` and `--exit-on-lock-contention` date
back to usage of bootkube and at one point running Kubelet
in a "self-hosted" style whereby an on-host Kubelet (rkt)
started pods, but then a Kubelet DaemonSet was scheduled
and able to take over (hence self-hosted). `lock-file` and
`exit-on-lock-contention` flags supported this pivot. The
pattern has been out of favor (in bootkube too) for years
because of dueling Kubelet complexity
* Typhoon runs Kubelet as a container via an on-host systemd
unit using podman (Fedora CoreOS) or rkt (Flatcar Linux). In
fact, Typhoon no longer uses bootkube or control plane pivot
(let alone Kubelet pivot) and uses static pods since v1.16.0
* https://github.com/poseidon/typhoon/pull/536
* Enable terraform-provider-ct `strict` mode for parsing
Container Linux Configs and snippets
* Fix Container Linux Config systemd unit syntax `enable`
(old) to `enabled`
* Align with Fedora CoreOS which uses strict mode already
* Build Kubelet container images internally and publish
to Quay and Dockerhub (new) as an alternative in case of
registry outage or breach
* Use our infra to provide single and multi-arch (default)
Kublet images for possible future use
* Docs: Show how to use alternative Kubelet images via
snippets and a systemd dropin (builds on #737)
Changes:
* Update docs with changes to Kubelet image building
* If you prefer to trust images built by Quay/Dockerhub,
automated image builds are still available with unique
tags (albeit with some limitations):
* Quay automated builds are tagged `build-{short_sha}`
(limit: only amd64)
* Dockerhub automated builts are tagged `build-{tag}`
and `build-master` (limit: only amd64, no shas)
Links:
* Kubelet: https://github.com/poseidon/kubelet
* Docs: https://typhoon.psdn.io/topics/security/#container-images
* Registries:
* quay.io/poseidon/kubelet
* docker.io/psdn/kubelet
* Promote DigitalOcean from alpha to beta for Fedora
CoreOS and Flatcar Linux
* Upgrade mkdocs-material and PyPI packages for docs
* Replace docs mentions of Container Linux with Flatcar
Linux and move docs/cl to docs/flatcar-linux
* Deprecate CoreOS Container Linux support. Its still
usable for some time, but start removing docs
* Set a consistent MCS level/range for Calico install-cni
* Note: Rebooting a node was a workaround, because Kubelet
relabels /etc/kubernetes(/cni/net.d)
Background:
* On SELinux enforcing systems, the Calico CNI install-cni
container ran with default SELinux context and a random MCS
pair. install-cni places CNI configs by first creating a
temporary file and then moving them into place, which means
the file MCS categories depend on the containers SELinux
context.
* calico-node Pod restarts creates a new install-cni container
with a different MCS pair that cannot access the earlier
written file (it places configs every time), causing the
init container to error and calico-node to crash loop
* https://github.com/projectcalico/cni-plugin/issues/874
```
mv: inter-device move failed: '/calico.conf.tmp' to
'/host/etc/cni/net.d/10-calico.conflist'; unable to remove target:
Permission denied
Failed to mv files. This may be caused by selinux configuration on
the
host, or something else.
```
Note, this isn't a host SELinux configuration issue.
Related:
* https://github.com/poseidon/terraform-render-bootstrap/pull/186
* Enable bootstrap token authentication on kube-apiserver
* Generate the bootstrap.kubernetes.io/token Secret that
may be used as a bootstrap token
* Generate a bootstrap kubeconfig (with a bootstrap token)
to be securely distributed to nodes. Each Kubelet will use
the bootstrap kubeconfig to authenticate to kube-apiserver
as `system:bootstrappers` and send a node-unique CSR for
kube-controller-manager to automatically approve to issue
a Kubelet certificate and kubeconfig (expires in 72 hours)
* Add ClusterRoleBinding for bootstrap token subjects
(`system:bootstrappers`) to have the `system:node-bootstrapper`
ClusterRole
* Add ClusterRoleBinding for bootstrap token subjects
(`system:bootstrappers`) to have the csr nodeclient ClusterRole
* Add ClusterRoleBinding for bootstrap token subjects
(`system:bootstrappers`) to have the csr selfnodeclient ClusterRole
* Enable NodeRestriction admission controller to limit the
scope of Node or Pod objects a Kubelet can modify to those of
the node itself
* Ability for a Kubelet to delete its Node object is retained
as preemptible nodes or those in auto-scaling instance groups
need to be able to remove themselves on shutdown. This need
continues to have precedence over any risk of a node deleting
itself maliciously
Security notes:
1. Issued Kubelet certificates authenticate as user `system:node:NAME`
and group `system:nodes` and are limited in their authorization
to perform API operations by Node authorization and NodeRestriction
admission. Previously, a Kubelet's authorization was broader. This
is the primary security motivation.
2. The bootstrap kubeconfig credential has the same sensitivity
as the previous generated TLS client-certificate kubeconfig.
It must be distributed securely to nodes. Its compromise still
allows an attacker to obtain a Kubelet kubeconfig
3. Bootstrapping Kubelet kubeconfig's with a limited lifetime offers
a slight security improvement.
* An attacker who obtains the kubeconfig can likely obtain the
bootstrap kubeconfig as well, to obtain the ability to renew
their access
* A compromised bootstrap kubeconfig could plausibly be handled
by replacing the bootstrap token Secret, distributing the token
to new nodes, and expiration. Whereas a compromised TLS-client
certificate kubeconfig can't be revoked (no CRL). However,
replacing a bootstrap token can be impractical in real cluster
environments, so the limited lifetime is mostly a theoretical
benefit.
* Cluster CSR objects are visible via kubectl which is nice
4. Bootstrapping node-unique Kubelet kubeconfigs means Kubelet
clients have more identity information, which can improve the
utility of audits and future features
Rel: https://kubernetes.io/docs/reference/command-line-tools-reference/kubelet-tls-bootstrapping/
Rel: https://github.com/poseidon/terraform-render-bootstrap/pull/185
* DigitalOcean firewall rules should reference Terraform tag
resources rather than using tag strings. Otherwise, terraform
apply can fail (neeeds rerun) if a tag has not yet been created
* Race: During initial bootstrap, static control plane pods
could hang with Permission denied to bootstrap secrets. A
manual fix involved restarting Kubelet, which relabeled mounts
The race had no effect on subsequent reboots.
* bootstrap.service runs podman with a private unshared mount
of /etc/kubernetes/bootstrap-secrets which uses an SELinux MCS
label with a category pair. However, bootstrap-secrets should
be shared as its mounted by Docker pods kube-apiserver,
kube-scheduler, and kube-controller-manager. Restarting Kubelet
was a manual fix because Kubelet relabels all /etc/kubernetes
* Fix bootstrap Pod to use the shared volume label, which leaves
bootstrap-secrets files with SELinux level s0 without MCS
* Also allow failed bootstrap.service to be re-applied. This was
missing on bare-metal and AWS
* Initial support for Flatcar Linux on Azure used the Flatcar
Linux Azure Marketplace images (e.g. `flatcar-stable`) in
https://github.com/poseidon/typhoon/pull/664
* Flatcar Linux Azure Marketplace images have some unresolved
items https://github.com/poseidon/typhoon/issues/703
* Until the Marketplace items are resolved, revert to requiring
Flatcar Linux's images be manually uploaded (like GCP and
DigitalOcean)
* No change to Fedora CoreOS modules
* For Container Linx AWS and Azure, change the `os_image` default
from coreos-stable to flatcar-stable
* For Container Linux GCP and DigitalOcean, change `os_image` to
be required since users should upload a Flatcar Linux image and
set the variable
* For Container Linux bare-metal, recommend users change the
`os_channel` to Flatcar Linux. No actual module change.
* Add "lb" outbound rule for worker TCP _and_ UDP traffic
* Fix Azure worker nodes clock synchronization being inactive
due to timeouts reaching the CoreOS / Flatcar NTP pool
* Fix Azure worker nodes not providing outbount UDP connectivity
Background:
Azure provides VMs outbound connectivity either by having a public
IP or via an SNAT masquerade feature bundled with their virtual
load balancing abstraction (in contrast with, say, a NAT gateway).
Azure worker nodes have only a private IP, but are associated with
the cluster load balancer's backend pool and ingress frontend IP.
Outbound traffic uses SNAT with this frontend IP. A subtle detail
with Azure SNAT seems to be that since both inbound lb_rule's are
TCP only, outbound UDP traffic isn't SNAT'd (highlights the reasons
Azure shouldn't have conflated inbound load balancing with outbound
SNAT concepts). However, adding a separate outbound rule and
disabling outbound SNAT on our ingress lb_rule's we can tell Azure
to continue load balancing as before, and support outbound SNAT for
worker traffic of both the TCP and UDP protocol.
Fixes clock synchronization timeouts:
```
systemd-timesyncd[786]: Timed out waiting for reply from
45.79.36.123:123 (3.flatcar.pool.ntp.org)
```
Azure controller nodes have their own public IP, so controllers (and
etcd) nodes have not had clock synchronization or outbound UDP issues
* Fix bootstrap error for missing `manifests-networking/crd*yaml`
when `networking = "flannel"`
* Cleanup manifest-networking directory left during bootstrap
* Regressed in v1.18.0 changes for Calico https://github.com/poseidon/typhoon/pull/675
* Change kube-proxy, flannel, and calico-node DaemonSet
tolerations to tolerate `node.kubernetes.io/not-ready`
and `node-role.kubernetes.io/master` (i.e. controllers)
explicitly, rather than tolerating all taints
* kube-system DaemonSets will no longer tolerate custom
node taints by default. Instead, custom node taints must
be enumerated to opt-in to scheduling/executing the
kube-system DaemonSets
* Consider setting the daemonset_tolerations variable
of terraform-render-bootstrap at a later date
Background: Tolerating all taints ruled out use-cases
where certain nodes might legitimately need to keep
kube-proxy or CNI networking disabled
Related: https://github.com/poseidon/terraform-render-bootstrap/pull/179
* Problem: Fedora CoreOS images are manually uploaded to GCP. When a
cluster is created with a stale image, Zincati immediately checks
for the latest stable image, fetches, and reboots. In practice,
this can unfortunately occur exactly during the initial cluster
bootstrap phase.
* Recommended: Upload the latest Fedora CoreOS image regularly
* Mitigation: Allow a failed bootstrap.service run (which won't touch
the done ConditionalPathExists) to be re-run by running `terraforma apply`
again. Add a known issue to CHANGES
* Update docs to show the current Fedora CoreOS stable version to
reduce likelihood users see this issue
Longer term ideas:
* Ideal: Fedora CoreOS publishes a stable channel. Instances will always
boot with the latest image in a channel. The problem disappears since
it works the same way AWS does
* Timer: Consider some timer-based approach to have zincati delay any
system reboots for the first ~30 min of a machine's life. Possibly just
configured on the controller node https://github.com/coreos/zincati/pull/251
* External coordination: For Container Linux, locksmith filled a similar
role and was disabled to allow CLUO to coordinate reboots. By running
atop Kubernetes, it was not possible for the reboot to occur before
cluster bootstrap
* Rely on https://github.com/coreos/zincati/issues/115 to delay the
reboot since bootstrap involves an SSH session
* Use path-based activation of zincati on controllers and set that
path at the end of the bootstrap process
Rel: https://github.com/coreos/fedora-coreos-tracker/issues/239
* Fix issue observed in us-east-1 where AMI filters chose the
latest testing channel release, rather than the stable chanel
* Fedora CoreOS AMI filter selects the latest image with a
matching name, x86_64, and hvm, excluding dev images. Add a
filter for "Fedora CoreOS stable", which seems to be the only
distinguishing metadata indicating the channel
* Recommend that users who have not yet tried Fedora CoreOS or
Flatcar Linux do so. Likely, Container Linux will reach EOL
and platform support / stability ratings will be in a mixed
state. Nevertheless, folks should migrate by September.
* Before Kubernetes v1.18.0, Kubelet only supported kubectl
`--limit-bytes` with the Docker `json-file` log driver so
the Fedora CoreOS default was overridden for conformance.
See https://github.com/poseidon/typhoon/pull/642
* Kubelet v1.18+ implemented support for other docker log
drivers, so the Fedora CoreOS default `journald` can be
used again
Rel: https://github.com/kubernetes/kubernetes/issues/86367
* Remove docs for the `asset_dir` variable and deprecate
it in CHANGES. It will be removed in an upcoming release
* Typhoon v1.17.0 introduced a new mechanism for managing
and distributing generated assets that stopped relying on
writing out to disk. `asset_dir` became optional and
defaulted to being unset / off (recommended)
* Kubernetes plans to stop releasing the hyperkube container image
* Upstream will continue to publish `kube-apiserver`, `kube-controller-manager`,
`kube-scheduler`, and `kube-proxy` container images to `k8s.gcr.io`
* Upstream will publish Kubelet only as a binary for distros to package,
either as a DEB/RPM on traditional distros or a container image on
container-optimized operating systems
* Typhoon will package the upstream Kubelet (checksummed) and its
dependencies as a container image for use on CoreOS Container Linux,
Flatcar Linux, and Fedora CoreOS
* Update the Typhoon container image security policy to list
`quay.io/poseidon/kubelet`as an official distributed artifact
Hyperkube: https://github.com/kubernetes/kubernetes/pull/88676
Kubelet Container Image: https://github.com/poseidon/kubelet
Kubelet Quay Repo: https://quay.io/repository/poseidon/kubelet
* Stop providing example manifests for the Container Linux
Update Operator (CLUO)
* CLUO requires patches to support Kubernetes v1.16+, but the
project and push access is rather unowned
* CLUO hasn't been in active use in our clusters and won't be
relevant beyond Container Linux. Not to say folks can't patch
it and run it on their own. Examples just aren't provided here
Related: https://github.com/coreos/container-linux-update-operator/pull/197
* Accept `os_image` "flatcar-stable" and "flatcar-beta" to
use Kinvolk's Flatcar Linux images from the Azure Marketplace
Note: Flatcar Linux Azure Marketplace images require terms be
accepted before use
* Add `worker_node_labels` map from node name to a list of initial
node label strings
* Add `worker_node_taints` map from node name to a list of initial
node taint strings
* Unlike cloud platforms, bare-metal node labels and taints
are defined via a map from node name to list of labels/taints.
Bare-metal clusters may have heterogeneous hardware so per node
labels and taints are accepted
* Only worker node names are allowed. Workloads are not scheduled
on controller nodes so altering their labels/taints isn't suitable
```
module "mercury" {
...
worker_node_labels = {
"node2" = ["role=special"]
}
worker_node_taints = {
"node2" = ["role=special:NoSchedule"]
}
}
```
Related: https://github.com/poseidon/typhoon/issues/429
* Add support for `terraform-provider-azurerm` v2.0+. Require
`terraform-provider-azurerm` v2.0+ and drop v1.x support since
the Azure provider major release is not backwards compatible
* Use Azure's new Linux VM and Linux VM Scale Set resources
* Change controller's Azure disk caching to None
* Associate subnets (in addition to NICs) with security groups
(aesthetic)
* If set, change `worker_priority` from `Low` to `Spot` (action required)
Related:
* https://www.terraform.io/docs/providers/azurerm/guides/2.0-upgrade-guide.html
* Add 2 min wait before KubeNodeUnreachable to be less
noisy on premeptible clusters
* Add a BlackboxProbeFailure alert for any failing probes
for services annotated `prometheus.io/probe: true`
* Quay has historically generated ACI signatures for images to
facilitate rkt's notions of verification (it allowed authors to
actually sign images, though `--trust-keys-from-https` is in use
since etcd and most authors don't sign images). OCI standardization
didn't adopt verification ideas and checking signatures has fallen
out of favor.
* Fix an issue where Quay no longer seems to be generating ACI
signatures for new images (e.g. quay.io/coreos/etcd:v.3.4.4)
* Don't be alarmed by rkt `--insecure-options=image`. It refers
to disabling image signature checking (i.e. docker pull doesn't
check signatures either)
* System containers for Kubelet and bootstrap have transitioned
to the docker:// transport, so there is precedent and this brings
all the system containers on Container Linux controllers into
alignment
* On clouds where workers can scale down or be preempted
(AWS, GCP, Azure), shutdown runs delete-node.service to
remove a node a prevent NotReady nodes from lingering
* Add the delete-node.service that wasn't carried over
from Container Linux and port it to use podman
* Allow users to extend the route table using a data reference
and adding route resources (e.g. unusual peering setups)
* Note: Internally connecting AWS clusters can reduce cross-cloud
flexibility and inhibits blue-green cluster patterns. It is not
recommended
* Add Terraform strip markers to consume beginning and
trailing whitespace in templated Kubelet arguments for
podman (Fedora CoreOS only)
* Fix initial `worker_node_labels` being quietly ignored
on Fedora CoreOS cloud platforms that offer the feature
* Close https://github.com/poseidon/typhoon/issues/650
* Add docs on manually uploading a Flatcar Linux DigitalOcean
bin image as a custom image and using a data reference
* Set status of Flatcar Linux on DigitalOcean to alpha
* IPv6 is not supported for DigitalOcean custom images
* Fix the last minor issue for Fedora CoreOS clusters to pass CNCF's
Kubernetes conformance tests
* Kubelet supports a seldom used feature `kubectl logs --limit-bytes=N`
to trim a log stream to a desired length. Kubelet handles this in the
CRI driver. The Kubelet docker shim only supports the limit bytes
feature when Docker is configured with the default `json-file` logging
driver
* CNCF conformance tests started requiring limit-bytes be supported,
indirectly forcing the log driver choice until either the Kubelet or
the conformance tests are fixed
* Fedora CoreOS defaults Docker to use `journald` (desired). For now,
as a workaround to offer conformant clusters, the log driver can
be set back to `json-file`. RHEL CoreOS likely won't have noticed the
non-conformance since its using crio runtime
* https://github.com/kubernetes/kubernetes/issues/86367
Note: When upstream has a fix, the aim is to drop the docker config
override and use the journald default
* Add docs on manually uploading a Flatcar Linux GCE/GCP gzipped
tarball image as a Compute Engine image for use with the Typhoon
container-linux module
* Set status of Flatcar Linux on Google Cloud to alpha
* Add Typhoon Fedora CoreOS on Google Cloud as alpha
* Add docs on uploading the Fedora CoreOS GCP gzipped tarball to
Google Cloud storage to create a boot disk image
* Use Fedora CoreOS production download streams (change)
* Use live PXE kernel and initramfs images
* https://getfedora.org/coreos/download/
* Update docs example to use public images (cache is still
recommended at large scale) and stable stream
* Typhoon Google Cloud is compatible with `terraform-provider-google`
v3.x releases
* No v3.x specific features are used, so v2.19+ provider versions are
still allowed, to ease migrations
* Change node-exporter DaemonSet tolerations from tolerating
all possible NoSchedule taints to tolerating the master taint
and the not ready taint (we'd like metrics regardless)
* Users who add custom node taints must add their custom taints
to the addon node-exporter DaemonSet. As an addon, its expected
users copy and manipulate manifests out-of-band in their own
systems
* Configure kube-proxy --metrics-bind-address=0.0.0.0 (default
127.0.0.1) to serve metrics on 0.0.0.0:10249
* Add firewall rules to allow Prometheus (resides on a worker) to
scrape kube-proxy service endpoints on controllers or workers
* Add a clusterIP: None service for kube-proxy endpoint discovery
* Change kubelet.service on Container Linux nodes to ExecStart Kubelet
inline to replace the use of the host OS kubelet-wrapper script
* Express rkt run flags and volume mounts in a clear, uniform way to
make the Kubelet service easier to audit, manage, and understand
* Eliminate reliance on a Container Linux kubelet-wrapper script
* Typhoon for Fedora CoreOS developed a kubelet.service that similarly
uses an inline ExecStart (except with podman instead of rkt) and a
more minimal set of volume mounts. Adopt the volume improvements:
* Change Kubelet /etc/kubernetes volume to read-only
* Change Kubelet /etc/resolv.conf volume to read-only
* Remove unneeded /var/lib/cni volume mount
Background:
* kubelet-wrapper was added in CoreOS around the time of Kubernetes v1.0
to simplify running a CoreOS-built hyperkube ACI image via rkt-fly. The
script defaults are no longer ideal (e.g. rkt's notion of trust dates
back to quay.io ACI image serving and signing, which informed the OCI
standard images we use today, though they still lack rkt's signing ideas).
* Shipping kubelet-wrapper was regretted at CoreOS, but remains in the
distro for compatibility. The script is not updated to track hyperkube
changes, but it is stable and kubelet.env overrides bridge most gaps
* Typhoon Container Linux nodes have used kubelet-wrapper to rkt/rkt-fly
run the Kubelet via the official k8s.gcr.io hyperkube image using overrides
(new image registry, new image format, restart handling, new mounts, new
entrypoint in v1.17).
* Observation: Most of what it takes to run a Kubelet container is defined
in Typhoon, not in kubelet-wrapper. The wrapper's value is now undermined
by having to workaround its dated defaults. Typhoon may be better served
defining Kubelet.service explicitly
* Typhoon for Fedora CoreOS developed a kubelet.service without the use
of a host OS kubelet-wrapper which is both clearer and eliminated some
volume mounts
* Update recommended Terraform and provider plugin versions
* Update the rough count of resources created per cluster
since its not been refreshed in a while (will vary based
on cluster options)
* Stop mapping node labels to targets discovered via Kubernetes
nodes (e.g. etcd, kubelet, cadvisor). It is rarely useful to
store node labels (e.g. kubernetes.io/os=linux) on these metrics
* kube-apiserver's apiserver_request_duration_seconds_bucket metric
has a high cardinality that includes labels for the API group, verb,
scope, resource, and component for each object type, including for
each CRD. This one metric has ~10k time series in a typical cluster
(btw 10-40% of total)
* Removing the apiserver request duration outright would make latency
alerts a NoOp and break a Grafana apiserver panel. Instead, drop series
that have a "group" label. Effectively, only request durations for
core Kubernetes APIs will be kept (e.g. cardinality won't grow with
each CRD added). This reduces the metric to ~2k unique series
* Original tutorials favored including the platform (e.g.
google-cloud) in modules (e.g. google-cloud-yavin). Prefer
naming conventions where each module / cluster has a simple
name (e.g. yavin) since the platform is usually redundant
* Retain the example cluster naming themes per platform
* kubernetes-mixin pod resource dashboards were split into
two ConfigMap parts because they provide richer networking
details
* New dashboards have been used by the author at the global
level, but were missing in the per-cluster Grafana tracked
here
* Add Prometheus alerts from node-exporter
* Add Grafana dashboard nodes.json, from node-exporter
* Not adding recording rules, since those are only used
by some node-exporter USE dashboards not being included
* terraform-provider-google v2.19.0 deprecates `instance_template`
within `google_compute_region_instance_group_manager` in order to
support a scheme with multiple version blocks. Adapt our single
version to the new format to resolve deprecation warnings.
* Fixes: Warning: "instance_template": [DEPRECATED] This field
will be replaced by `version.instance_template` in 3.0.0
* Require terraform-provider-google v2.19.0+ (action required)
* Set small CPU requests on static pods kube-apiserver,
kube-controller-manager, and kube-scheduler to align with
upstream tooling and for edge cases
* Effectively, a practical case for these requests hasn't been
observed. However, a small static pod CPU request may offer
a slight benefit if a controller became overloaded and the
below mechanisms were insufficient
Existing safeguards:
* Control plane nodes are tainted to isolate them from
ordinary workloads. Even dense workloads can only compress
CPU resources on worker nodes.
* Control plane static pods use the highest priority class, so
contention favors control plane pods (over say node-exporter)
and CPU is compressible too.
See: https://github.com/poseidon/terraform-render-bootstrap/pull/161
* Update terraform-render-bootstrap module to adopt the
Terrform v0.12 templatefile function feature to replace
the use of terraform-provider-template's `template_dir`
* Require Terraform v0.12.6+ which adds `for_each`
Background:
* `template_dir` was added to `terraform-provider-template`
to add support for template directory rendering in CoreOS
Tectonic Kubernetes distribution (~2017)
* Terraform v0.12 introduced a native `templatefile` function
and v0.12.6 introduced native `for_each` support (July 2019)
that makes it possible to replace `template_dir` usage
* addon-resizer is outdated and has been dropped from
kube-state-metrics examples. Those using it should look
to the cluster-proportional-vertical-autoscaler.
* Eliminate addon-resizer log spew
* Remove associated Role and RoleBinding
* Also fix kube-state-metrics readinessProbe port
* Update Prometheus rules/alerts and Grafana dashboards
* Remove dashboards that were moved to node-exporter, they
may be added back later if valuable
* Remove kube-prometheus based rules/alerts (ClockSkew alert)
* Change `networking` default from flannel to calico on
Azure and DigitalOcean
* AWS, bare-metal, and Google Cloud continue to default
to Calico (as they have since v1.7.5)
* Typhoon now defaults to using Calico and supporting
NetworkPolicy on all platforms
* Define bare-metal `controllers` and `workers` as a complex type
list(object{name=string, mac=string, domain=string}) to allow
clusters with many machines to be defined more cleanly
* Remove `controller_names` list variable
* Remove `controller_macs` list variable
* Remove `controller_domains` list variable
* Remove `worker_names` list variable
* Remove `worker_macs` list variable
* Remove `worker_domains` list variable
* Detect the most recent Fedora CoreOS AMI to allow usage
of Fedora CoreOS in supported regions (previously just
us-east-1)
* Unpin the Fedora CoreOS AMI image which was pinned to
images that had been checked. This does mean if Fedora
publishes a broken image, it will be selected
* Filter out "dev" images which have similar naming
* Review variables available in bare-metal kubernetes modules
for Container Linux and Fedora CoreOS
* Deprecate cluster_domain_suffix variable
* Remove deprecated container_linux_oem variable
* Prometheus was configured to use kubernetes discovery
of etcd targets based on nodes matching the node label
node-role.kubernetes.io/controller=true
* Kubernetes v1.16 stopped permitting node role labels
node-role.kubernetes.io/* so Typhoon renamed these labels
(no longer any association with roles) to
node.kubermetes.io/controller=true
* As a result, Prometheus didn't discover etcd targets,
etcd metrics were missing, etcd alerts were ineffective,
and the etcd Grafana dashboard was empty
* Introduced: https://github.com/poseidon/typhoon/pull/543
* Review variables available in DigitalOcean kubernetes
module and sync with documentation
* Promote Calico for DigitalOcean and Azure beyond experimental
(its the primary mode I've used since it was introduced)
* Review variables available in Azure kubernetes and workers
modules and sync with documentation
* Fix internal workers module default type to Standard_DS1_v2
* Review variables available in AWS kubernetes and workers
modules and documentation
* Switching between spot and on-demand has worked since
Terraform v0.12
* Generally, there are too many knobs. Less useful ones
should be de-emphasized or removed
* Remove `cluster_domain_suffix` documentation
* Document worker pools `node_labels` variable to set the
initial node labels for a homogeneous set of workers
* Document `worker_node_labels` convenience variable to
set the initial node labels for default worker nodes
* Drop `node-role.kubernetes.io/master` and
`node-role.kubernetes.io/node` node labels
* Kubelet (v1.16) now rejects the node labels used
in the kubectl get nodes ROLES output
* https://github.com/kubernetes/kubernetes/issues/75457
* Change default controller_type to Standard_B2s. A B2s is cheaper
by $17/month and provides 2 vCPU, 4GB RAM (vs 1 vCPU, 3.5GB RAM)
* Change default worker_type to Standard_DS1_v2. F1 was the previous
generation. The DS1_v2 is newer, similar cost, more memory, and still
supports Low Priority mode, if desired
* Use unique dashboard ID so that multiple replicas of Grafana
serve dashboards with uniform paths
* Fix issue where refreshing a dashboard served by one replica
could show a 404 unless the request went to the same replica
* terraform-provider-aws v2.23.0 allows AWS root block devices
to enable encryption by default.
* Require updating terraform-provider-aws to v2.23.0 or higher
* Enable root EBS device encryption by default for controller
instances and worker instances in auto-scaling groups
For comparison:
* Google Cloud persistent disks have been encrypted by
default for years
* Azure managed disk encryption is not ready yet (#486)
* Add a CoreDNS dashboard originally based on an upstream dashboard,
but now customized according to preferences
* Add an Nginx Ingress Controller based on an upstream dashboard,
but customized according to preferences
* Add all Azure controllers to the apiserver load balancer
backend address pool
* Previously, kube-apiserver availability relied on the 0th
controller being up. Multi-controller was just providing etcd
data redundancy
* Allow updating terraform-provider-ct to any release
beyond v0.3.2, but below v1.0. This relaxes the prior
constraint that allowed only v0.3.y provider versions
* Run kube-apiserver as a non-root user (nobody). User
no longer needs to bind low number ports.
* On most platforms, the kube-apiserver load balancer listens
on 6443 and fronts controllers with kube-apiserver pods using
port 6443. Google Cloud TCP proxy load balancers cannot listen
on 6443. However, GCP's load balancer can be made to listen on
443, while kube-apiserver uses 6443 across all platforms.
* Fix a GCP errata item https://github.com/poseidon/typhoon/wiki/Errata
* Removal of a Google Cloud cluster often required 2 runs of
`terraform apply` because network resource deletes timeout
after 4m. Raise the network deletion timeout to 6m to
ensure apply only needs to be run once to remove a cluster
* google_compute_backend_services use nested blocks to define
backends (instance groups heterogeneous controllers)
* Use Terraform v0.12.x dynamic blocks so the apiserver backend
service can refer to (up to zone-many) controller instance groups
* Previously, with Terraform v0.11.x, the apiserver backend service
had to list a fixed set of backends to span controller nodes across
zones in multi-controller setups. 3 backends were used because each
GCP region offered at least 3 zones. Single-controller clusters had
the cosmetic ugliness of unused instance groups
* Allow controllers to span more than 3 zones if avilable in a
region (e.g. currently only us-central1, with 4 zones)
Related:
* https://www.terraform.io/docs/providers/google/r/compute_backend_service.html
* https://www.terraform.io/docs/configuration/expressions.html#dynamic-blocks
* Delay changing README example. Its prominent display
on github.com may lead to new users copying it, even
though it corresponds to an "in between releases" state
and v1.14.4 doesn't exist yet
* Leave docs tutorials the same, they can reflect master
* Provide Terraform v0.11 to v0.12 migration guide. Show an
in-place strategy and a move resources strategy
* Describe in-place modifying an existing cluster and providers,
using the Terraform helper to edit syntax, and checking the
plan produces a zero diff
* Describe replacing existing clusters by creating a new config
directory for use with Terraform v0.12 only and moving resources
one by one
* Provide some limited advise on migrating non-Typhoon resources
* Replace v0.11 bracket type hints with Terraform v0.12 list expressions
* Use expression syntax instead of interpolated strings, where suggested
* Update bare-metal tutorial
* Define `clc_snippets` type constraint map(list(string))
* Define Terraform and plugin version requirements in versions.tf
* Require matchbox ~> 0.3.0 to support Terraform v0.12
* Require ct ~> 0.3.2 to support Terraform v0.12
* Azure only allows `eviction_policy` to be set for Low priority VMs.
Supporting Low priority VMs meant when Regular VMs were used, each
`terraform apply` rolled workers, to set eviction_policy to null.
* Terraform v0.12 nullable variables fix the issue and plan does not
produce a diff
* Replace v0.11 bracket type hints with Terraform v0.12 list expressions
* Use expression syntax instead of interpolated strings, where suggested
* Update Azure tutorial and worker pools documentation
* Define Terraform and plugin version requirements in versions.tf
* Require azurerm ~> 1.27 to support Terraform v0.12
* Require ct ~> 0.3.2 to support Terraform v0.12
* Replace v0.11 bracket type hints with Terraform v0.12 list expressions
* Use expression syntax instead of interpolated strings, where suggested
* Update Google Cloud tutorial and worker pools documentation
* Define Terraform and plugin version requirements in versions.tf
* Require google ~> 2.5 to support Terraform v0.12
* Require ct ~> 0.3.2 to support Terraform v0.12
* Replace v0.11 bracket type hints with Terraform v0.12 list expressions
* Use expression syntax instead of interpolated strings, where suggested
* Update AWS tutorial and worker pools documentation
* Define Terraform and plugin version requirements in versions.tf
* Require aws ~> 2.7 to support Terraform v0.12
* Require ct ~> 0.3.2 to support Terraform v0.12
* Replace v0.11 bracket type hints with Terraform v0.12 list expressions
* Use expression syntax instead of interpolated strings, where suggested
* Update DigitalOcean tutorial documentation
* Define Terraform and plugin version requirements in versions.tf
* Require digitalocean ~> v1.3 to support Terraform v0.12
* Require ct ~> v0.3.2 to support Terraform v0.12
* This change affects users who use worker pools on AWS, GCP, or
Azure with a Container Linux derivative
* Rename worker pool modules' `count` variable to `worker_count`,
because `count` will be a reserved variable name in Terraform v0.12
* Fix to remove a trailing slash that was erroneously introduced
in the scripting that updated from v1.14.1 to v1.14.2
* Workaround before this fix was to re-run `terraform init`
* Introduce "calico" as a `networking` option on Azure and DigitalOcean
using Calico's new VXLAN support (similar to flannel). Flannel remains
the default on these platforms for now.
* Historically, DigitalOcean and Azure only allowed Flannel as the
CNI provider, since those platforms don't support IPIP traffic that
was previously required for Calico.
* Looking forward, its desireable for Calico to become the default
across Typhoon clusters, since it provides NetworkPolicy and a
consistent experience
* No changes to AWS, GCP, or bare-metal where Calico remains the
default CNI provider. On these platforms, IPIP mode will always
be used, since its available and more performant than vxlan
* DigitalOcean clusters must secure copy a kubeconfig to
worker nodes, but Terraform could decide to try copying
before firewall rules have been added to allow SSH access.
* Add an explicit dependency on adding firewall rules first
* Change flannel port from the kernel default 8472 to the
IANA assigned VXLAN port 4789
* Update firewall rules or security groups for VXLAN
* Why now? Calico now offers its own VXLAN backend so
standardizing on the IANA port will simplify config
* https://github.com/coreos/flannel/blob/master/Documentation/backends.md#vxlan
* This reverts commit 6e5d66cf66
* kube-state-metrics v1.6.0-rc.0 fires KubeDeploymentReplicasMismatch
alerts where its own Deployment doesn't have replicas available,
(kube_deployment_status_replicas_available) even though all replicas
are available according to kubectl inspection
* This problem was present even with the CSR ClusterRole fix
(https://github.com/kubernetes/kube-state-metrics/pull/717)
* Add ability to load balance TCP/UDP applications (e.g. NodePort)
* Output the load balancer ID as `loadbalancer_id`
* Output `worker_security_group_name` and `worker_address_prefix`
for extending firewall rules
* Add an `enable_aggregation` variable to enable the kube-apiserver
aggregation layer for adding extension apiservers to clusters
* Aggregation is **disabled** by default. Typhoon recommends you not
enable aggregation. Consider whether less invasive ways to achieve your
goals are possible and whether those goals are well-founded
* Enabling aggregation and extension apiservers increases the attack
surface of a cluster and makes extensions a part of the control plane.
Admins must scrutinize and trust any extension apiserver used.
* Passing a v1.14 CNCF conformance test requires aggregation be enabled.
Having an option for aggregation keeps compliance, but retains the
stricter security posture on default clusters
* Refresh rules and dashboards from upstreams
* Add new Kubernetes "workload" dashboards
* View pods in a workload (deployment/daemonset/statefulset)
* View workloads in a namespace
* Define firewall rules on DigitialOcean to match rules used on AWS,
GCP, and Azure
* Output `controller_tag` and `worker_tag` to simplify custom firewall
rule creation
* Add ability to load balance TCP applications (e.g. NodePort)
* Output the network load balancer ARN as `nlb_id`
* Accept a `worker_target_groups` (ARN) list to which worker
instances should be added
* AWS NLBs and target groups don't support UDP
* Background: A managed instance group of workers is used in backend
services for global load balancing (HTTP/HTTPS Ingress) and output
for custom global load balancing use cases
* Add worker instances to a target pool load balancing TCP/UDP
applications (NodePort or proxied). Output as `worker_target_pool`
* Health check for workers with a healthy Ingress controller. Forward
rules (regional) to target pools don't support different external and
internal ports so choosing nodes with Ingress allows proxying as a
workaround
* A target pool is a logical grouping only. It doesn't add costs to
clusters or worker pools
* Allow traffic from Grafana to Prometheus in monitoring
* Allow traffic from Prometheus to Prometheus in monitoring
* NetworkPolicy denies non-whitelisted traffic. Define policy
to allow other access
* Add calico-ipam CRDs and RBAC permissions
* Switch IPAM from host-local to calico-ipam
* `calico-ipam` subnets `ippools` (defaults to pod CIDR) into
`ipamblocks` (defaults to /26, but set to /24 in Typhoon)
* `host-local` subnets the pod CIDR based on the node PodCIDR
field (set via kube-controller-manager as /24's)
* Create a custom default IPv4 IPPool to ensure the block size
is kept at /24 to allow 110 pods per node (Kubernetes default)
* Retaining host-local was slightly preferred, but Calico v3.6
is migrating all usage to calico-ipam. The codepath that skipped
calico-ipam for KDD was removed
* https://docs.projectcalico.org/v3.6/release-notes/
* Heapster addon powers `kubectl top`
* In early Kubernetes, people legitimately used and expected
`kubectl top` to work, so the optional addon was provided
* Today the standards are different. Many better monitoring
tools exist, that are also less coupled to Kubernetes "kubectl
top" reliance on a non-core extensions means its not in-scope
for minimal Kubernetes clusters. No more exceptionalism
* Finally, Heapster isn't that useful anymore. Its manifests
have no need for Typhoon-specific modification
* Look to prior releases if you still wish to apply heapster
* Restore the original special-casing of DigitalOcean Kubelets
* Fix node metadata InternalIP being set to the IP of the default
gateway on DigitalOcean nodes (regressed in v1.12.3)
* Reverts the "pretty" node names on DigitalOcean (worker-2 vs IP)
* Closes#424 (full details)
* Require an iPXE-enabled network boot environment with support for
TLS downloads. PXE clients must chainload to iPXE firmware compiled
with `DOWNLOAD_PROTO_HTTPS` enabled ([crypto](https://ipxe.org/crypto))
* iPXE's pre-compiled firmware binaries do _not_ enable HTTPS. Admins
should build iPXE from source with support enabled
* Affects the Container Linux and Flatcar Linux install profiles that
pull from public downloads. No effect when cached_install=true
or using Fedora Atomic, as those download from Matchbox
* Add `download_protocol` variable. Recognizing boot firmware TLS
support is difficult in some environments, set the protocol to "http"
for the old behavior (discouraged)
* Allow terraform-provider-aws >= v1.13, but < 3.0. No change
to the minimum version, but allow using v2.x.y releases
* Verify compatability with terraform-provider-aws v2.1.0
* Resolve in-addr.arpa and ip6.arpa DNS PTR requests for Kubernetes
service IPs and pod IPs
* Previously, CoreDNS was configured to resolve in-addr.arpa PTR
records for service IPs (but not pod IPs)
* Support terraform-provider-google v1.19.0, v1.19.1, v1.20.0
and v2.0+ (and allow for future 2.x.y releases)
* Require terraform-provider-google v1.19.0 or newer. v1.19.0
introduced `network_interface` fields `network_ip` and `nat_ip`
to deprecate `address` and `assigned_nat_ip`. Those deprecated
fields are removed in terraform-provider-google v2.0
* https://github.com/terraform-providers/terraform-provider-google/releases/tag/v2.0.0
* Assign pod priorityClassNames to critical cluster and node
components (higher is higher priority) to inform node out-of-resource
eviction order and scheduler preemption and scheduling order
* Priority Admission Controller has been enabled since Typhoon
v1.11.1
* Intel Haswell or better is available in every zone around the world
* Neither Kubernetes nor Typhoon have a particular minimum processor
family. However, a few Google Cloud zones still default to Sandy/Ivy
bridge (scheduled to shift April 2019). Price is only based on machine
type so it is beneficial to opt for the next processor family
* Intel Haswell is a suitable minimum since it still allows plenty of
liberty in choosing any region or machine type
* Likely a slight increase to preemption probability in a few zones,
but any lower probability on Sandy/Ivy bridge is due to lower
desirability as they're phased out
* https://cloud.google.com/compute/docs/regions-zones/
* Collate upstream rules, alerts, and dashboards and tune for use
in Typhoon
* Previously, a well-chosen (but older) set of rules, alerts, and
dashboards were maintained to reflect metric name changes
* Prometheus queries from some upstreams use joins of node-exporter
and kube-state-metrics metrics by (namespace,pod). Add the Kubernetes
pod name to service endpoint metrics
* Rename the kubernetes_namespace field to namespace
* Honor labels since kube-state-metrics already include a `pod` field
that should not be overridden
* Fix a regression caused by lowering the Kubelet TLS client
certificate to system:nodes group (#100) since dropping
cluster-admin dropped the Kubelet's ability to delete nodes.
* On clouds where workers can scale down (manual terraform apply,
AWS spot termination, Azure low priority deletion), worker shutdown
runs the delete-node.service to remove a node to prevent NotReady
nodes from accumulating
* Allow Kubelets to delete cluster nodes via system:nodes group. Kubelets
acting with system:node and kubelet-delete ClusterRoles is still an
improvement over acting as cluster-admin
* DNS zones served by AWS Route53 may use AWS's special alias records
(other DNS providers would use a CNAME) to resolve the ingress NLB.
Alias records require the NLB DNS name's DNS zone id (not the cluster
`dns_zone_id`)
* System components that require certificates signed by the cluster
CA can submit a CSR to the apiserver, have an administrator inspect
and approve it, and be issued a certificate
* Configure kube-controller-manager to sign Approved CSR's using the
cluster CA private key
* Admins are responsible for approving or denying CSRs, otherwise,
no certificate is issued. Read the Kubernetes docs carefully and
verify the entity making the request and the authorization level
* https://kubernetes.io/docs/tasks/tls/managing-tls-in-a-cluster
* Kubelets can use a lower-privilege TLS client certificate with
Org system:nodes and a binding to the system:node ClusterRole
* Admin kubeconfig's continue to belong to Org system:masters to
provide cluster-admin (available in assets/auth/kubeconfig or as
a Terraform output kubeconfig-admin)
* Remove bare-metal output variable kubeconfig
* Add ServiceAccounts and ClusterRoleBindings for kube-apiserver
and kube-scheduler
* Remove the ClusterRoleBinding for the kube-system default ServiceAccount
* Rename the CA certificate CommonName for consistency with upstream
* T3 is the next generation general purpose burstable
instance type. Compared with t2.small, the t3.small is
cheaper, has 2 vCPU (instead of 1) and provides 5 Gbps
of pod-to-pod bandwidth (instead of 1 Gbps)
* Provide migration instructions for upgrading terraform-provider-ct
in-place for v1.12.2+ clusters
* Require switching from ~/.terraformrc to the Terraform third-party
plugins directory ~/.terraform.d/plugins/
* Require Container Linux 1688.5.3 or newer
* On GCP, kubectl port-forward connections to pods are closed
after a timeout (unlike AWS NLB's or Azure load balancers)
* Increase the GCP apiserver backend service timeout from 1 minute
to 5 minutes to be more similar to AWS/Azure LB behavior
* Example manifests aim to provide a read-only dashboard visible
to any users with network access (i.e. kubectl port-forward, LAN)
* Problem: Grafana always has an admin user, even with the user
management system disabled
* Disable the login form to prevent admin login
* Switch tutorials from using ~/.terraformrc to using the 3rd-party
plugin directory so 3rd-party plugins can be pinned
* Continue to show using terraform-provider-ct v0.2.2. Updating to
a newer version is only safe once all managed clusters are v1.12.2
or higher
