-
Notifications
You must be signed in to change notification settings - Fork 34
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
agent: delay reboot if ongoing interactive sessions #115
Comments
This requires #114 to be implemented first. |
* 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 coreos/zincati#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 coreos/zincati#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
* 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 coreos/zincati#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 coreos/zincati#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: coreos/fedora-coreos-tracker#239
@lucab Now we(OpenStack Magnum) are using Fedora CoreOS as nodes for k8s cluster. But because of the auto update, we have to totally disable it now to avoid interrupting the k8s bootstrapping. However, I would like to see a delayed reboot or manual reboot to leave the option to the owner of the server. Looking forward to seeing your thoughts. Cheers. |
Hi @openstacker! This specific ticket here is not what you are looking for. The topic however has been previously discussed at coreos/fedora-coreos-tracker#147 (comment). The approaches from there are (in order of personal preference):
|
@lucab Re. #115 (comment), can we document 1 and 2 in https://github.com/coreos/zincati/blob/master/docs/usage/auto-updates.md? This topic came up in the Nest with Fedora Hands-On Lab. |
* Update Grafana from v6.4.4 to v6.5.0 * https://grafana.com/docs/guides/whats-new-in-v6-5/ * Update Grafana from v6.5.0 to v6.5.1 * https://github.com/grafana/grafana/releases/tag/v6.5.1 * Fix DigitalOcean controller and worker ipv4/ipv6 outputs (#594) * Fix controller and worker ipv4/ipv4 outputs to be lists of strings * With Terraform v0.11 syntax, an enclosing list was required to coerce the output to be a list of strings * With Terraform v0.12 syntax, the enclosing list shouldn't be needed * Update mkdocs-material from v4.5.0 to v4.5.1 * Introduce cluster creation without local writes to asset_dir * Allow generated assets (TLS materials, manifests) to be securely distributed to controller node(s) via file provisioner (i.e. ssh-agent) as an assets bundle file, rather than relying on assets being locally rendered to disk in an asset_dir and then securely distributed * Change `asset_dir` from required to optional. Left unset, asset_dir defaults to "" and no assets will be written to files on the machine that runs terraform apply * Enhancement: Managed cluster assets are kept only in Terraform state, which supports different backends (GCS, S3, etcd, etc) and optional encryption. terraform apply accesses state, runs in-memory, and distributes sensitive materials to controllers without making use of local disk (simplifies use in CI systems) * Enhancement: Improve asset unpack and layout process to position etcd certificates and control plane certificates more cleanly, without unneeded secret materials Details: * Terraform file provisioner support for distributing directories of contents (with unknown structure) has been limited to reading from a local directory, meaning local writes to asset_dir were required. https://github.com/poseidon/typhoon/issues/585 discusses the problem and newer or upcoming Terraform features that might help. * Observation: Terraform provisioner support for single files works well, but iteration isn't viable. We're also constrained to Terraform language features on the apply side (no extra plugins, no shelling out) and CoreOS / Fedora tools on the receive side. * Take a map representation of the contents that would have been splayed out in asset_dir and pack/encode them into a single file format devised for easy unpacking. Use an awk one-liner on the receive side to unpack. In pratice, this has worked well and its rather nice that a single assets file is transferred by file provisioner (all or none) Rel: https://github.com/poseidon/terraform-render-bootstrap/pull/162 * Add/update docs for asset_dir and kubeconfig usage * 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 * Reduce apiserver metrics cardinality and extraneous labels * 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 * Reduce kube-controller-manager pod eviction timeout from 5m to 1m * Reduce time to delete pods on unready nodes from 5m to 1m * Present since v1.13.3, but mistakenly removed in v1.16.0 static pod control plane migration Related: * https://github.com/poseidon/terraform-render-bootstrap/pull/148 * https://github.com/poseidon/terraform-render-bootstrap/pull/164 * Update Kubernetes from v1.16.3 to v1.17.0 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG-1.17.md/#v1170 * Update systemd services for the v0.17.x hyperkube * Binary asset locations within the upstream hyperkube image changed https://github.com/kubernetes/kubernetes/pull/84662 * Fix Container Linux and Flatcar Linux kubelet.service (rkt-fly with fairly dated CoreOS kubelet-wrapper) * Fix Fedora CoreOS kubelet.service (podman) * Fix Fedora CoreOS bootstrap.service * Fix delete-node kubectl usage for workers where nodes may delete themselves on shutdown (e.g. preemptible instances) * Update Calico from v3.10.1 to v3.10.2 * https://docs.projectcalico.org/v3.10/release-notes/ * Update CHANGES and tutorial notes for release * 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) * Fix minor example typo in README * Update mkdocs-material from v4.5.1 to v4.6.0 * Update Grafana from v6.5.1 to v6.5.2 * https://github.com/grafana/grafana/releases/tag/v6.5.2 * Update kube-state-metrics from v1.8.0 to v1.9.0-rc.1 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.0-rc.1 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.0-rc.0 * Add Kubelet kubeconfig output for DigitalOcean * Allow the raw kubelet kubeconfig to be consumed via Terraform output * Update kube-state-metrics from v1.9.0-rc.1 to v1.9.0 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.0 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.0-rc.1 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.0-rc.0 * Update CoreDNS from v1.6.5 to v1.6.6 * https://coredns.io/2019/12/11/coredns-1.6.6-release/ * Update Prometheus from v2.14.0 to v2.15.0 * https://github.com/prometheus/prometheus/releases/tag/v2.15.0 * Update Prometheus from v2.15.0 to v2.15.1 * https://github.com/prometheus/prometheus/releases/tag/v2.15.1 * Update Calico from v3.10.2 to v3.11.1 * https://docs.projectcalico.org/v3.11/release-notes/ * Rename CLC files and favor Terraform list index syntax * Rename Container Linux Config (CLC) files to *.yaml to align with Fedora CoreOS Config (FCC) files and for syntax highlighting * Replace common uses of Terraform `element` (which wraps around) with `list[index]` syntax to surface index errors * Inline Container Linux kubelet.service, deprecate kubelet-wrapper * 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 * Disable Kubelet 127.0.0.1.10248 healthz endpoint * Kubelet runs a healthz server listening on 127.0.0.1:10248 by default. Its unused by Typhoon and can be disabled * https://kubernetes.io/docs/reference/command-line-tools-reference/kubelet/ * Enable kube-proxy metrics and allow Prometheus scrapes * 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 * Reduce Prometheus addon's node-exporter tolerations * 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 * Ensure /etc/kubernetes exists following Kubelet inlining * Inlining the Kubelet service removed the need for the kubelet.env file declared in Ignition. However, on some platforms, this removed the guarantee that /etc/kubernetes exists. Bare-Metal and DigitalOcean distribute the kubelet kubeconfig through Terraform file provisioner (scp) and place it in (now missing) /etc/kubernetes * https://github.com/poseidon/typhoon/pull/606 * Fix bare-metal and DigitalOcean Ignition to ensure the desired directory exists following first boot from disk * Cloud platforms with worker pools distribute the kubeconfig through Ignition user data (no impact or need) * Update Prometheus from v2.15.1 to v2.15.2 * https://github.com/prometheus/prometheus/releases/tag/v2.15.2 * Allow terraform-provider-google v3.x plugin versions * 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 * Update kube-state-metrics from v1.9.0 to v1.9.1 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.1 * Remove unneeded Kubelet /var/run mount on Fedora CoreOS * /var/run symlinks to /run (already mounted) * Fix bare-metal instruction for watching install to disk * Original instructions were to watch install to disk by SSH'ing via port 2222 following Typhoon v1.10.1. Restore that message, since the version number in the instruction was incorrectly bumped on each release * Update AWS Fedora CoreOS AMI filter for fedora-coreos-31 * Select the most recent fedora-coreos-31 AMI on AWS, instead of the most recent fedora-coreos-30 AMI (Nov 27, 2019) * Evaluated with fedora-coreos-31.20200108.2.0-hvm * Update Kubernetes from v1.17.0 to v1.17.1 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG-1.17.md#v1171 * Update Calico from v3.11.1 to v3.11.2 * https://docs.projectcalico.org/v3.11/release-notes/ * Fix link in maintenance docs * Also a fix version mention, since Terraform v0.12 was added in Typhoon v1.15.0 * Update Grafana from v6.5.2 to v6.5.3 * https://github.com/grafana/grafana/releases/tag/v6.5.3 * Update kube-state-metrics from v1.9.1 to v1.9.2 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.2 * Update bare-metal Fedora CoreOS image location * 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 * Update nginx-ingress from v0.26.1 to v0.27.1 * Change runAsUser from 33 to 101 for new alpine-based image * https://github.com/kubernetes/ingress-nginx/releases/tag/nginx-0.27.0 * https://github.com/kubernetes/ingress-nginx/releases/tag/nginx-0.27.1 * Update Kubernetes from v1.17.1 to v1.17.2 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG-1.17.md#v1172 * Update kube-state-metrics from v1.9.2 to v1.9.3 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.3 * Promote Fedora CoreOS from preview to alpha in docs * Add an announcement to the website as well * Fix minor typo in announcement date * Update Grafana from v6.5.3 to v6.6.0 * https://github.com/grafana/grafana/releases/tag/v6.6.0 * Update nginx-ingress from v0.27.1 to v0.28.0 * https://github.com/kubernetes/ingress-nginx/releases/tag/nginx-0.28.0 * Add module for Fedora CoreOS on Google Cloud * 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 * Update kube-state-metrics from v1.9.3 to v1.9.4 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.4 * Update Calico from v3.11.2 to v3.12.0 * https://docs.projectcalico.org/release-notes/#v3120 * Remove reverse packet filter override, since Calico no longer relies on the setting * https://github.com/coreos/fedora-coreos-tracker/issues/219 * https://github.com/projectcalico/felix/pull/2189 * Update Grafana from v6.6.0 to v6.6.1 * https://github.com/grafana/grafana/releases/tag/v6.6.1 * Update docs generation packages * Update mkdocs-material from v4.6.0 to v4.6.2 * Add guide for Typhoon with Flatcar Linux on Google Cloud * 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 * Update Fedora CoreOS kernel arguments to align with upstream * Align bare-metal kernel arguments with upstream docs * Add missing initrd argument which can cause issues if not present. Fix #638 * Add tty0 and ttyS0 consoles (matches Container Linux) * Remove unused coreos.inst=yes Related: https://docs.fedoraproject.org/en-US/fedora-coreos/bare-metal/ * Update Kubernetes from v1.17.2 to v1.17.3 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.17.md#v1173 * Set docker log driver to json-file on Fedora CoreOS * 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 * Promote Fedora CoreOS AWS/bare-metal to beta * Remove alpha warnings from docs headers * Update recommended Terraform versions and providers * Sync the documented Terraform versions and provider plugin versions to those that are actively used/tested by the author * Update CHANGELOG sections and links * Add guide for Typhoon with Flatcar Linux on DigitalOcean * 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 * Update Prometheus from v1.15.2 to v1.16.0 * https://github.com/prometheus/prometheus/releases/tag/v2.16.0 * Change Kubelet /var/lib/calico mount to read-only (#643) * Kubelet only requires read access to /var/lib/calico Signed-off-by: Suraj Deshmukh <[email protected]> * Update CoreDNS from v1.6.6 to v1.6.7 * https://coredns.io/2020/01/28/coredns-1.6.7-release/ * Update mkdocs-material from v4.6.2 to v4.6.3 * Fix worker_node_labels for initial Fedora CoreOS * 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 * Update Grafana from v6.6.1 to v6.6.2 * https://github.com/grafana/grafana/releases/tag/v6.6.2 * Update kube-state-metrics from v1.9.4 to v1.9.5 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.5 * Update nginx-ingress from v0.28.0 to v0.29.0 * https://github.com/kubernetes/ingress-nginx/releases/tag/nginx-0.29.0 * Update nginx-ingress from v0.29.0 to v0.30.0 * https://github.com/kubernetes/ingress-nginx/releases/tag/nginx-0.30.0 * Update node-exporter from v0.18.1 to v1.0.0-rc.0 * Update mdadm alert rule; node-exporter adds `state` label to `node_md_disks` and removes `node_md_disks_active` * https://github.com/prometheus/node_exporter/releases/tag/v1.0.0-rc.0 * Use a route table with separate (rather than inline) routes * 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 * Update etcd from v3.4.3 to v3.4.4 * https://github.com/etcd-io/etcd/releases/tag/v3.4.4 * Add automatic worker deletion on Fedora CoreOS clouds * 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 * Change Container Linux etcd-member to fetch with docker:// * 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 * Refresh Prometheus alerts and Grafana dashboards * 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` * Upgrade terraform-provider-azurerm to v2.0+ * 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 * Accept initial worker node labels and taints map on bare-metal * 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 Flatcar Linux on Azure * 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 * Update Calico from v3.12.0 to v3.13.1 * https://docs.projectcalico.org/v3.13/release-notes/ * Update Kubernetes from v1.17.3 to v1.17.4 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.17.md#v1174 * Update recommended Terraform versions and providers * Sync the documented Terraform versions and provider plugin versions to those that are actively used/tested by the author * Remove Container Linux Update Operator (CLUO) addon * 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 * Promote Fedora CoreOS AWS and Google Cloud * Promote Fedora CoreOS AWS to stable * Promote Fedora CoreOS GCP to beta * Update etcd from v3.4.4 to v3.4.5 * https://github.com/etcd-io/etcd/releases/tag/v3.4.5 * Update Prometheus from v2.16.0 to v2.17.0-rc.3 * https://github.com/prometheus/prometheus/releases/tag/v2.17.0-rc.3 * Update Grafana from v6.6.2 to v6.7.1 * https://github.com/grafana/grafana/releases/tag/v6.7.1 * Switch from upstream hyperkube image to individual images * 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 * Fix image tag for Container Linux AWS workers * #669 left one reference to the original SHA tagged image before the v1.17.4 image tag was applied * Update Prometheus from v2.17.0-rc.3 to v2.17.0 * https://github.com/prometheus/prometheus/releases/tag/v2.17.0 * Rename DigitalOcean image variable to os_image * Rename variable `image` to `os_image` to match the naming used for the same purpose on other supported platforms (e.g. AWS, Azure, GCP) * Deprecate asset_dir variable and remove docs * 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) * Add Fedora CoreOS to issue template and docs * Update several Container Linux references to start referring to Flatcar Linux * Update docs and mentions of Fedora CoreOS * Update Kubernetes from v1.17.4 to v1.18.0 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md * Update docs from Kubernetes v1.17.4 to v1.18.0 * Set docker log driver to journald on Fedora CoreOS * 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 * Update Prometheus from v2.17.0 to v2.17.1 * https://github.com/prometheus/prometheus/releases/tag/v2.17.1 * Add CoreOS Container Linux EOL recommendation to CHANGES * 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. * Fix delete-node.service kubectl service exec's * Fix delete-node service that runs on worker (cloud-only) shutdown to delete a Kubernetes node. Regressed in #669 (unreleased) * Use rkt `--exec` to invoke kubectl binary in the kubelet image * Use podman `--entrypoint` to invoke the kubectl binary in the kubelet image * Fix Fedora CoreOS AMI to filter for stable images * 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 * Add support for Fedora CoreOS snippets * Refresh snippets customization docs * Requires terraform-provider-ct v0.5+ * Allow bootstrap re-apply for Fedora CoreOS GCP * 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 * Change default kube-system DaemonSet tolerations * 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 * Fix bootstrap regression when networking="flannel" * 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 * Rename Container Linux snippets variable for consistency * Rename controller_clc_snippets to controller_snippets (cloud platforms) * Rename worker_clc_snippets to worker_snippets (cloud platforms) * Rename clc_snippets to snippets (bare-metal) * Update flannel from v0.11.0 to v0.12.0 * https://github.com/coreos/flannel/releases/tag/v0.12.0 * Fix UDP outbound and clock sync timeouts on Azure workers * 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 terraform fmt * Refresh Prometheus rules/alerts and Grafana dashboards * Refresh upstream Prometheus rules and alerts and Grafana dashboards * All Loki recording rules for convenience * Update Grafana from v6.7.1 to v6.7.2 * https://github.com/grafana/grafana/releases/tag/v6.7.2 * Update etcd from v3.4.5 to v3.4.7 * https://github.com/etcd-io/etcd/releases/tag/v3.4.7 * https://github.com/etcd-io/etcd/releases/tag/v3.4.6 * Update Kubernetes from v1.18.0 to v1.18.1 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md#v1181 * Add support for Fedora CoreOS on DigitalOcean * Add `digital-ocean/fedora-coreos/kubernetes` module * DigitalOcean custom uploaded images do not permit droplet IPv6 networking * Update CHANGES for v1.18.1 release * Change order of modules in the README * Fix docs TOC to include Fedora CoreOS DigitalOcean * Change `container-linux` module preference to Flatcar Linux * 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 support for Fedora CoreOS on Azure * Add `azure/fedora-coreos/kubernetes` module * Fix Fedora CoreOS Azure MTU with Calico * With Calico VXLAN on Fedora CoreOS the 1450 MTU should be used * Update Kubernetes from v1.18.1 to v1.18.2 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md#changelog-since-v1181 * Remove temporary workaround for v1.18.0 apply issue * In v1.18.0, kubectl apply would fail to apply manifests if any single manifest was unable to validate. For example, if a CRD and CR were defined in the same directory, apply would fail since the CR would be invalid as the CRD wouldn't exist * Typhoon temporary workaround was to separate CNI CRD manifests and explicitly apply them first. No longer needed in v1.18.1+ * Kubernetes v1.18.1 restored the prior behavior where kubectl apply applies as many valid manifests as it can. In the example above, the CRD would be applied and the CR could be applied if the kubectl apply was re-run (allowing for apply loops). * Upstream fix: https://github.com/kubernetes/kubernetes/pull/89864 * Revert Flatcar Linux Azure to manual upload images * 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) * Fix bootstrap mount to use shared volume SELinux label * 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 * Fix race condition creating DigitalOcean firewall rules * 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 * Update Prometheus from v2.17.1 to v2.17.2 * https://github.com/prometheus/prometheus/releases/tag/v2.17.2 * Remove extraneous sudo from layout asset unpacking * Update Calico from v3.13.1 to v3.13.3 * https://docs.projectcalico.org/v3.13/release-notes/ * Enable Kubelet TLS bootstrap and NodeRestriction * 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 * Add Fedora CoreOS Azure docs to site navigation * Fix missing Fedora CoreOS Azure docs * Update recommended Terraform provider versions * Sync the Terraform provider plugin versions to those actively used and tested by the author * Fix terraform fmt * Use Terraform element wrap-around for AWS controllers subnet_id (#714) * Fix Terraform plan error when controller_count exceeds available AWS zones (e.g. 5 controllers) * Update Grafana from v6.7.2 to v7.0.0-beta1 * https://github.com/grafana/grafana/releases/tag/v7.0.0-beta1 * Update Prometheus from v2.17.2 to v2.18.0-rc.1 * https://github.com/prometheus/prometheus/releases/tag/v2.18.0-rc.1 * Update nginx-ingress from v0.30.0 to v0.32.0 * Add support for IngressClass and RBAC authorization * Since our nginx ingress controller example uses the flag `--ingress-class=public`, add an IngressClass to go along with it Rel: https://kubernetes.io/docs/concepts/services-networking/ingress/#ingress-class * Update Prometheus from v2.18.0-rc.1 to v2.18.0 * https://github.com/prometheus/prometheus/releases/tag/v2.18.0 * Update Prometheus from v2.18.0 to v2.18.1 * https://github.com/prometheus/prometheus/releases/tag/v2.18.1 * Update Grafana from v7.0.0-beta1 to v7.0.0-beta2 * https://github.com/grafana/grafana/releases/tag/v7.0.0-beta2 * Use Fedora CoreOS image streams on Google Cloud * Add `os_stream` variable to set a Fedora CoreOS stream to `stable` (default), `testing`, or `next` * Deprecate `os_image` variable. Remove docs about uploading Fedora CoreOS images manually, this is no longer needed * https://docs.fedoraproject.org/en-US/fedora-coreos/update-streams/ Rel: https://github.com/coreos/fedora-coreos-docs/pull/70 * Fix Calico install-cni crash loop on Pod restarts * 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 * Update Calico from v3.13.3 to v3.14.0 * https://docs.projectcalico.org/v3.14/release-notes/ * Update Grafana from v7.0.0-beta2 to v7.0.0-beta.3 * https://github.com/grafana/grafana/releases/tag/v7.0.0-beta3 * Support Fedora CoreOS OS image streams on AWS * Add `os_stream` variable to set the stream to stable (default), testing, or next * Remove unused os_image variable on Fedora CoreOS AWS * Highlight SELinux enforcing mode in features * Restore use of Flatcar Linux Azure Marketplace image * Switch Flatcar Linux Azure to use the Marketplace image from Kinvolk (offer `flatcar-container-linux-free`) * Accepting Azure Marketplace terms is still neccessary, update docs to show accepting the free offer rather than BYOL * Upstream Flatcar: https://github.com/flatcar-linux/Flatcar/issues/82 * Typhoon: https://github.com/poseidon/typhoon/issues/703 * Update Grafana from v7.0.0-beta3 to v7.0.0 * https://github.com/grafana/grafana/releases/tag/7.0.0 * Update kube-state-metrics from v1.9.5 to v1.9.6 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.6 * Update node-exporter from v1.0.0-rc.0 to v1.0.0-rc.1 * https://github.com/prometheus/node_exporter/releases/tag/v1.0.0-rc.1 * Rollback Grafana to v7.0.0-beta3, v7.0.0 image is missing * Grafana hasn't published the v7.0.0 image yet * Use new Azure subnet to set address_prefixes list * Update Azure subnet `address_prefix` to `azure_prefixes` list * Fix warning that `address_prefix` is deprecated * Require `terraform-provider-azurerm` v2.8.0+ (action required) Rel: https://github.com/terraform-providers/terraform-provider-azurerm/pull/6493 * Update Grafana from v7.0.0-beta2 to v7.0.0 * https://grafana.com/docs/grafana/latest/guides/whats-new-in-v7-0/ * Update etcd from v3.4.7 to v3.4.8 * https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.4.md#v348-2020-05-18 * Fix Fedora CoreOS on GCP proposing controller recreate * With Fedora CoreOS image stream support (#727), the latest resolved image will change over the lifecycle of a cluster. * Fix issue where an image diff proposed replacing a Fedora CoreOS controller on GCP, introduced in #727 (unreleased) * Also ignore image diffs to the GCP managed instance group of workers. This aligns with worker AMI diffs being ignored on AWS and similar on Azure, since workers update themselves. Background: * Controller nodes should strictly not be recreated by Terraform, they are stateful (etcd) and should not be replaced * Across cloud platforms, OS image diffs are ignored since both Flatcar Linux and Fedora CoreOS nodes update themselves. For workers, user-data or disk size diffs (where relevant) are allowed to recreate workers templates/configs since these are considered to be user-initiated declarations that a reprovision should be done * Set Kubelet image via kubelet.service KUBELET_IMAGE * Write the systemd kubelet.service to use `KUBELET_IMAGE` as the Kubelet. This provides a nice way to use systemd dropins to temporarily override the image (e.g. during a registry outage) Note: Only Typhoon Kubelet images and registries are supported. * Update Kubernetes from v1.18.2 to v1.18.3 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md * Upgrade docs packages and refresh content * 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 * Update etcd from v3.4.8 to v3.4.9 * https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.4.md#v349-2020-05-20 * Update recommended Terraform provider versions * Sync Terraform provider plugin versions to those actively used internally * Fix terraform fmt * Update node-exporter from v1.0.0-rc.1 to v1.0.0 * https://github.com/prometheus/node_exporter/releases/tag/v1.0.0 * Update Grafana from v7.0.0 to v7.0.1 * https://github.com/grafana/grafana/releases/tag/v7.0.1 * Update mkdocs-material from v5.2.0 to v5.2.2 * https://github.com/squidfunk/mkdocs-material/releases/tag/5.2.2 * Update Github issue template to use drop-downs (#747) * Create a stricter bug report template * Highlight topics that are not accepted in issues: operation, support, debugging, advice, or Kubernetes concepts * Add a section to strongly suggest bug reports link a PR or describe a solution. This may be able to weed out topics that aren't focused bug reports * Update the fallback issue template * Even "blank" issues need to fill out the fallback template * Update Calico from v3.14.0 to v3.14.1 * https://docs.projectcalico.org/v3.14/release-notes/ * Change Kubelet container image publishing * 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 * Tweak minor style elements of issue templates * Update kube-state-metrics from v1.9.6 to v1.9.7 * https://github.com/kubernetes/kube-state-metrics/releases/tag/v1.9.7 * Update Grafana from v7.0.1 to v7.0.3 * https://github.com/grafana/grafana/releases/tag/v7.0.2 * https://github.com/grafana/grafana/releases/tag/v7.0.3 * Update Prometheus from v2.18.1 to v2.19.0-rc.0 * https://github.com/prometheus/prometheus/releases/tag/v2.19.0-rc.0 * Fix Fedora CoreOS docs for selecting a stream * Fedora CoreOS image `os_stream` stable, testing, and next have been configurable since v1.18.3 * Remove mention of outdated `os_image` variable * Update security disclosure contact email * Use [email protected] across github.com/poseidon projects * Use strict mode for Container Linux Configs * 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 * Update Prometheus from v2.19.0-rc.0 to v2.19.0 * https://github.com/prometheus/prometheus/releases/tag/v2.19.0 * Remove unused Kubelet cert / key Terraform state * Generated Kubelet TLS certificate and key are not longer used or distributed to machines since Kubelet TLS bootstrap is used instead. Remove the certificate and key from state * Remove unused Kubelet lock-file and exit-on-lock-contention * 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 * Update node-exporter from v1.0.0 to v1.0.1 * https://github.com/prometheus/node_exporter/releases/tag/v1.0.1 * Update mkdocs packages for website * Fix typo in DigitalOcean docs title * Update nginx-ingress from v0.32.0 to v0.33.0 * https://github.com/kubernetes/ingress-nginx/releases/tag/controller-0.33.0 * Update Kubernetes from v1.18.3 to v1.18.4 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md#v1184 * Update recommended Terraform provider versions * Sync Terraform provider plugin versions with those used internally * Rename controller node label and NoSchedule taint * 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 * Fix Kubelet starting before hostname set on FCOS AWS * 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 * Reduce Calcio MTU on Fedora CoreOS Azure * 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 * Add experimental Cilium CNI provider * Accept experimental CNI `networking` mode "cilium" * Run Cilium v1.8.0-rc4 with overlay vxlan tunnels and a minimal set of features. We're interested in: * IPAM: Divide pod_cidr into /24 subnets per node * CNI networking pod-to-pod, pod-to-external * BPF masquerade * NetworkPolicy as defined by Kubernetes (no L7 Policy) * Continue using kube-proxy with Cilium probe mode * Firewall changes: * Require UDP 8472 for vxlan (Linux kernel default) between nodes * Optional ICMP echo(8) between nodes for host reachability (health) * Optional TCP 4240 between nodes for endpoint reachability (health) Known Issues: * Containers with `hostPort` don't listen on all host addresses, these workloads must use `hostNetwork` for now https://github.com/cilium/cilium/issues/12116 * Erroneous warning on Fedora CoreOS https://github.com/cilium/cilium/issues/10256 Note: This is experimental. It is not listed in docs and may be changed or removed without a deprecation notice Related: * https://github.com/poseidon/terraform-render-bootstrap/pull/192 * https://github.com/cilium/cilium/issues/12217 * Update Cilium from v1.8.0-rc4 to v1.8.0 * https://github.com/cilium/cilium/releases/tag/v1.8.0 * Update Prometheus from v2.19.0 to v2.19.1 * https://github.com/prometheus/prometheus/releases/tag/v2.19.1 * Update Grafana from v7.0.3 to v7.0.4 * https://github.com/grafana/grafana/releases/tag/v7.0.4 * Update mkdocs-material from v5.3.0 to v5.3.3 * Update Calico from v3.14.1 to v3.15.0 * https://docs.projectcalico.org/v3.15/release-notes/ * Update Kubernetes from v1.18.4 to v1.18.5 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md#v1185 * Update Prometheus from v2.19.1 to v2.19.2 * https://github.com/prometheus/prometheus/releases/tag/v2.19.2 * Update recommended Terraform provider versions * Sync Terraform provider plugin versions with those used internally * Revert "Update Prometheus from v2.19.1 to v2.19.2" * Prometheus has not published the v1.19.2 * This reverts commit 81b6f54169119702c3cc6a3ecabca77f8646b444. * Isolate each DigitalOcean cluster in its own VPC * 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 * Remove os_image variable on Google Cloud Fedora CoreOS * 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 * Fix terraform fmt in firewall rules * Promote Fedora CoreOS on Google Cloud to stable status * Allow using Flatcar Linux edge on Azure * 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 * Remove CoreOS Container Linux image names from docs * Remove coreos-stable, coreos-beta, and coreos-alpha channel references from docs * CoreOS Container Linux is end of life (see changelog) * Update Grafana from v7.0.4 to v7.0.5 * https://github.com/grafana/grafana/releases/tag/v7.0.5 * Update Cilium from v1.8.0 to v1.8.1 * https://github.com/cilium/cilium/releases/tag/v1.8.1 * Update Prometheus from v2.19.1 to v2.19.2 * https://github.com/prometheus/prometheus/releases/tag/v2.19.2 * Update Grafana from v7.0.5 to v7.0.6 * https://github.com/grafana/grafana/releases/tag/v7.0.6 * Update mkdocs-material from v5.3.3 to v5.4.0 * Update Kubernetes from v1.18.5 to v1.18.6 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md#v1186 * https://github.com/poseidon/terraform-render-bootstrap/pull/201 * Update ingress-nginx from v0.33.0 to v0.34.1 * Switch to ingress-nginx controller images from us.grc.io (eu, asia can also be used if desired) * https://github.com/kubernetes/ingress-nginx/releases/tag/controller-v0.34.1 * https://github.com/kubernetes/ingress-nginx/releases/tag/controller-v0.34.0 * Update recommended Terraform provider versions * Sync Terraform provider plugin versions with those used internally * Show Cilium as a CNI provider option in docs * Start to show Cilium as a CNI option * https://github.com/cilium/cilium * Update Grafana from v7.0.6 to v7.1.0 * https://github.com/grafana/grafana/releases/tag/v7.1.0 * Update etcd from v3.4.9 to v3.4.10 * https://github.com/etcd-io/etcd/releases/tag/v3.4.10 * Declare etcd data directory permissions * Set etcd data directory /var/lib/etcd permissions to 700 * On Flatcar Linux, /var/lib/etcd is pre-existing and Ignition v2 doesn't overwrite the directory. Update the Container Linux config, but add the manual chmod workaround to bootstrap for Flatcar Linux users * https://github.com/etcd-io/etcd/blob/master/CHANGELOG-3.4.md#v3410-2020-07-16 * https://github.com/etcd-io/etcd/pull/11798 * Update CoreDNS from v1.6.7 to v1.7.0 * https://coredns.io/2020/06/15/coredns-1.7.0-release/ * Update Grafana dashboard with revised metrics names * Update Cilium from v1.8.1 to v1.8.2 * https://github.com/cilium/cilium/releases/tag/v1.8.2 * Fix some links in docs (#788) * Update Grafana from v7.1.0 to v7.1.1 * https://github.com/grafana/grafana/releases/tag/v7.1.1 * Update Prometheus from v2.19.2 to v2.20.0 * https://github.com/prometheus/prometheus/releases/tag/v2.20.0 * Migrate to Fedora CoreOS * bastion only needs base fcos config * Revert "bastion only needs base fcos config" This reverts commit 2984f3e70be6f8ecdc6a96f70cf809b260c919d6. * support custom bastion snippet as a variable * Fix flannel support on Fedora CoreOS * 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. * Relex terraform-provider-matchbox version constraint * Allow use of terraform-provider-matchbox v0.3+ (which allows v0.3.0 <= version < v1.0) for any pre 1.0 release * Before, the requirement was v0.3.0 <= version < v0.4.0 * Update from coreos/flannel-cni to poseidon/flannel-cni * Update CNI plugins from v0.6.0 to v0.8.6 to fix several CVEs * Update the base image to alpine:3.12 * Use `flannel-cni` as an init container and remove sleep * https://github.com/poseidon/terraform-render-bootstrap/pull/205 * https://github.com/poseidon/flannel-cni * https://quay.io/repository/poseidon/flannel-cni Background * Switch from github.com/coreos/flannel-cni v0.3.0 which was last published by me in 2017 and is no longer accessible to me to maintain or patch * Port to the poseidon/flannel-cni rewrite, which releases v0.4.0 to continue the prior release numbering * Update mkdocs-material from v5.4.0 to v5.5.1 * use scoop's fork of terraform-render-bootstrap * policy arn * Revert "policy arn" This reverts commit 4579af8bda4a5720e791c56fa02c14ecb767a537. * workers and controllers need to stay private * fedora coreos 32 * fedora coreos 32 * Support Fedora CoreOS OS image streams on AWS * fix mistakes in resolving merging conflicts * add new security components * fix json format * Update Grafana from v7.1.1 to v7.1.3 * https://github.com/grafana/grafana/releases/tag/v7.1.3 * https://github.com/grafana/grafana/releases/tag/v7.1.2 * Allow terraform-provider-aws v3.0+ plugin * Typhoon AWS is compatible with terraform-provider-aws v3.x releases * Continue to allow v2.23+, no v3.x specific features are used * Set required provider versions in the worker module, since it can be used independently Related: * https://github.com/terraform-providers/terraform-provider-aws/releases/tag/v3.0.0 * Update recommended Terraform provider versions * Sync Terraform provider plugin versions to those used internally * fix ssl cert mounts * Migrate from Terraform v0.12.x to v0.13.x * Recommend Terraform v0.13.x * Support automatic install of poseidon's provider plugins * Update tutorial docs for Terraform v0.13.x * Add migration guide for Terraform v0.13.x (best-effort) * Require Terraform v0.12.26+ (migration compatibility) * Require `terraform-provider-ct` v0.6.1 * Require `terraform-provider-matchbox` v0.4.1 * Require `terraform-provider-digitalocean` v1.20+ Related: * https://www.hashicorp.com/blog/announcing-hashicorp-terraform-0-13/ * https://www.terraform.io/upgrade-guides/0-13.html * https://registry.terraform.io/providers/poseidon/ct/latest * https://registry.terraform.io/providers/poseidon/matchbox/latest * apiserver nlb should be internal * update terraform-render-bootstrap with latest upstream * Update Terraform migration guide SHA * 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 * Update Kubernetes from v1.18.6 to v1.18.8 * https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.18.md#v1188 * Update recommended Terraform provider versions * Sync Terraform provider plugin versions to those used internally * Update mkdocs-material from v5.5.1 to v5.5.6 * Fix minor details in docs * try relabeling /etc/kubernetes/bootstrap-secrets by explicitly mounting to kubelet * relabeling does not need explicitly mounting to kubelet * need to update the type label of bootstrap-secret in the newest typhoon * update terraform-render-bootstrap with latest upstream * rm unnecessary volume mounts on etcd * rm output/ Co-authored-by: Dalton Hubble <[email protected]> Co-authored-by: Arve Knudsen <[email protected]> Co-authored-by: Suraj Deshmukh <[email protected]> Co-authored-by: Ben Drucker <[email protected]> Co-authored-by: Eldon <[email protected]>
WRT the inhibitor-lock protocol, because there's currently no convenient verb/command for a user to lift the systemd-inhibit lock, and it doesn't seem intuitive for an interactive user to need to "cancel the inhibit and finalize now" in order to reboot into an update, perhaps it would be simpler to match Locksmith behaviour and introduce a delay if there are logged in interactive users. To allow users to be able to manually reboot into an update after they are done with their work, could we include instructions to use |
That's a bit of a complex UX, let's avoid going that route for now and re-evaluate later on.
👍 I like this approach of splitting the delay into smaller intervals. If we go granular enough it is maybe enough to avoid the need for a manual finalization procedure. |
This is a followup to coreos/fedora-coreos-tracker#239.
Right now Zincati triggers a reboot in a silent way, which could surprise any logged users while they are in middle of some interactive maintenance.
For better user experience, it should somehow delay the reboot if it notices any active sessions. Current proposal is to count the number of active consoles before finalizing, and introduce a delay if the result is non-zero. This would match locksmith behavior.
Locksmith used to just sleep for a fixed amount of time. A better option to explore here would be implementing the inhibitor-lock protocol.
The text was updated successfully, but these errors were encountered: