Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog.

0.14.0 - 2019-03-05

Changed

A new, long-lived, self-signed certificate authority has been added to sign kubelet certificate-signing requests. This works around the current lack of certificate rotation in the machine-config operator.
Machine(Set) labels have been migrated from sigs.k8s.io/cluster-api-... to machine.openshift.io, continuing the transition begun in 0.13.0.
On AWS, control-plane nodes are now based on encrypted AMIs. These AMIs are copied into the target account from unencrypted, public AMIs provided by Red Hat. To support the copy and post-cluster cleanup, the installer requires the following additional AWS credentials: ec2:CopyImage, ec2:DeregisterImage, and ec2:DeleteSnapshot. 0.14.0 doesn't actually clean up the snapshots associated with the copied AMIs yet, but we have a fix for that landed for the next release. In the meantime, you should manually prune your snapshots after destroying a cluster.
On AWS, the security-group simplification from 0.13.1 accidentially removed global SSH access to the bootstrap machine. We've fixed that with this release. Unfortunately, this release also moves the bootstrap machine into the same subnet as the first control-plane node, and since 0.13.0, control-plane nodes are in private subnets. So SSH access to the bootstrap machine from outside the cluster is still broken, but we've landed a fix to get it working again in the next release. In the meantime, you can set up a SSH bastion or debug pod if you need SSH access to cluster machines.
On OpenStack, the Machine(Set)s have been updated to track provider changes. For example, the SecurityGroups schema has changed, as has the schema for selecting subnets.
Several doc and internal cleanups.

Fixed

On AWS, we now respect the availability zones configured in the control-plane Machine manifests, which are in turn fed by the install-config (previously control-plane nodes were always striped over zones regardless of the configuration).
On AWS, the credentials-checking logic now uses the standard logger instead of creating its own custom logger.

0.13.1 - 2019-02-28

Changed

The aggregator and etcd-client certificate authorities are now self-signed authorities decoupled from the root certificate authority, continuing the transition begun in 0.13.0.
On AWS, Route 53 A records for the API load balancer no longer use health checks.
On AWS, the security group configuration has been simplified, with several stale rules being removed.

Fixed

When rendering manifests before pushing them to the cluster, the bootstrap machine now correctly cleans up broken renders before re-rendering.
The bootstrap machine now uses an etcdctl referenced from the release image, instead of hard-coding its own version.

Removed

The nominal install-config compatibility with v1beta1 and v1beta2 has been removed, so the installer will error out if provided with an older install-config.yaml. v1beta1 was deprecated in 0.12.0 and v1beta2 was deprecated in 0.13.0. In both cases, the installer would ignore removed properties but not error out.

0.13.0 - 2019-02-26

Added

When cluster-creation times out waiting for cluster-version completion, the installer now logs the last failing-operator message (if any).
The installer now invokes the cluster-config operator on the bootstrap machine to generate config.openshift.io custom resource definitions.

Changed

The install-config version has been bumped from v1beta2 to v1beta3. All users will need to update any saved install-config.yaml to use the new schema.
- machines has been split into controlPlane and compute. Multiple compute pools are now supported (previously, only a single worker pool was supported). Every compute pool will use the same Ignition configuration. The installer will warn about but allow configurations where there are zero compute replicas.
- On libvirt, the masterIPs property has been removed, since you cannot configure master IPs via the libvirt machine API provider.
- On OpenStack, there is also a new lbFloatingIP property, which allows you to provide an IP address to be used by the load balancer. This allows you to create local DNS entries ahead of time before calling create cluster.
Cluster domain names have been adjusted so that the cluster lives entirely within a per-cluster subdomain. This keeps split-horizon DNS from masking other clusters with the same base domain.
The cluster-version update URL has been changed from the dummy http://localhost:8080/graph to the functioning https://api.openshift.com/api/upgrades_info/v1/graph and the channel has been changed from fast to stable-4.0, to opt clusters in to 4.0 upgrades.
Machine-API resources have been moved from cluster.k8s.io to machine.openshift.io to clarify our divergence from the upstream types while they are unstable. The openshift-cluster-api namespace has been replaced with openshift-machine-api as well.
The installer now uses etcd and OS images referenced by the update payload when configuring the machine-config operator.
The etcd, aggregator, and other certificate authorities are now self-signed, decoupling their chains of trust from the root certificate authority.
The installer no longer creates a service-serving certificate authority. The certificate authority is now created by the service-CA operator.
On AWS, the worker IAM role permissions were reduced to a smaller set required for kubelet initialization.
On AWS, the worker security group has been expanded to allow ports 9000-9999 for for host network services. This matches the approach we have been using for masters since 0.4.0. The master security group has also been adjusted to fix a 9990 -> 9999 typo from 0.4.0.
On libvirt, the default compute nodes have been bumped from 2 to 4 GiB of memory and the control-plane nodes have been bumped from 4 to 6 GiB of memory and 2 to 4 vCPUs.
Several doc and internal cleanups and minor fixes.

Fixed

The router certificate authority is appended to the admin kubeconfig to fix the OAuth flow behind oc login.
The install-config.yaml validation is now more robust, with the installer:
- Validating cluster names (it previously only validated cluster names provided via the install-config wizard).
- Validating networking.clusterNetworks[].cidr and explicitly checking for nil machineCIDR and serviceCIDR.
Terraform variables are now generated from master machine configurations instead of from the install configuration. This allows them to reflect changes made by editing master machine configurations during staged installs.
metadata.json is generated before the Terraform invocation, fixing a bug introduced in 0.12.0 which made it hard to clean up after failed Terraform creation.
The machine-config server has moved its Ignition-config service from port 49500 to 22623 to avoid the dynamic-port range starting at 49152.
When the installer prompts for AWS credentials, it now respects AWS_PROFILE and will update an existing credentials file instead of erroring out.
On AWS, the default instance types now depend on the selected region, with regions that do not support m4 types falling back to m5.
On AWS, the installer now verifies that the user-supplied credentials have sufficient permissions for creating a cluster. Previously, permissions issues would surface as Terraform errors or broken cluster functionality after a nominally successful install.
On AWS, the destroy cluster implementation is now more robust, fixing several bugs from 0.10.1:
- The destroy code now checks for nil before dereferencing, avoiding panics when removing internet gateways which had not yet been associated with a VPC, and in other similar cases.
- The destoy code now treats already-deleted instances as successfully deleted, instead of looping forever while trying to delete them.
- The destroy code now treats a non-existant public DNS zone as success, instead of looping forever while trying to delete records from it.
On AWS and OpenStack, there is a new infra ID that is a uniqified, possibly-abbreviated form of the cluster name. The infra ID is used to name and tag cluster resources, allowing for multiple clusters that share the same cluster name in a single account without naming conflicts (beyond DNS conflicts if both clusters also share the same base domain).
On OpenStack, the HAProxy configuration on the service VM now only balances ports 80 and 443 across compute nodes (it used to also balance them across control-plane nodes).
On OpenStack, the service VM now uses CoreDNS instead of dnsmasq. And it now includes records for *.apps.{cluster-domain} and the Kubernetes API.
On OpenStack, the service VM has been moved to its own subnet.

Removed

On AWS, control-plane nodes have been moved to private subnets and no longer have public IPs. Use a VPN or bastion host if you need SSH access to them.

0.12.0 - 2019-02-05

Changed

We now wait for ClusterVersion to report all operators as available before returning from create cluster.
We now configure the network operator via networks.config.openshift.io and reserve networkconfigs.networkoperator.openshift.io for lower-level configuration (although we still generate it as well).
We now set apiServerURL and etcdDiscoveryDomain in infrastructures.config.openshift.io.
Release binaries are now stripped, which dramatically reduces their size. Builds with MODE=dev remain unstripped if you want to attach a debugger.
On AWS, destroy cluster no longer depends directly on the cluster name (although it still depends on the cluster name indirectly via the kubernetes.io/cluster/{name} tag). This makes it easier to reconstruct metadata.json for destroy cluster if you accidentally removed the file before destroying your cluster.
On AWS, the default worker MachineSets have been bumped to 120 GiB volumes to increase our baseline performance from on gp2's sliding IOPS scale. The new default worker volumes match our master bump from 0.5.0.
On OpenStack, the HAProxy configuration on the service VM is dynamically updated as masters and workers are added and removed. This supports console access, among other things.
Several doc and internal cleanups.

Fixed

We no longer write distracting ERROR: logging before flag.Parse... messages from our underlying Kubernetes libraries.
On loading install-config.yaml, we now error on CIDRs whose IP is not at the beginning of the masked subnet. For example, we now error for 192.168.126.10/24, since the beginning of that subnet is 192.168.126.0.
On loading install-config.yaml, we now fill in defaults for replicas when it is unset or explicitly null.
We have fixed some issues with round-tripping assets between the installer and the asset directory which lead to the reloaded assets being falsely identified as dirty and rebuilt.
On OpenStack, a new security rule exposes port 443 to allow OpenShift web-console access.
On OpenStack, credentials secret generation now respects the install configuration's cloud value, and the secret name has been updated from openstack-creds to openstack-credentials.
On OpenStack, the local-dns service will now restart on failure (e.g. when the initial image pull fails) and it no longer sets the name of the container (so we can always re-run it without running into duplicate name issues).

Removed

On loading install-config.yaml, the installer no longer restricts networking.type to a known value. If the network operator sees an unrecognized type, it assumes the user is configurating networking and doesn't react.
We no longer seed ~core/.bash_history on the bootstrap node, as part of becoming less opinionated about which users are present on the underlying operating system.
On AWS, the iamRoleName machine-pool property is gone, and the podCIDR networking property (deprecated in 0.4.0) is gone. The install-config version has been bumped from v1beta1 to v1beta2. All users, regardless of platform, will need to update any saved install-config.yaml to use the new version. IAM roles are being replaced by the credentials operator, and while we still create IAM roles for our master, worker, and bootstrap machines, we're removing the user-facing property now to avoid making this breaking change later.
On AWS, the bootstrap machine security group allowing kubelet access (added in 0.10.1) has been removed. Static pod logs should soon be available from journald (although they aren't yet).

0.11.0 - 2019-01-27

Added

On AWS, the installer creates DHCP options for the VPC to support internal unqualified-hostname resolution. This works around some limitations with oc rsh and Kubernetes node registration in the face of inappropriate default DHCP options. And because the AWS domain-name logic is region-specific, there is no single DHCP options configuration that provides internal unqualified-hostname resolution for multiple regions.

Changed

On AWS, the installer now prompts for missing credentials even if you supplied an install-config.yaml. Previously, only the install-config wizard would prompt.
On OpenStack, the developer-only internal DNS server which was removed in 0.10.0 has been restored, because the approach taken in 0.10.0 broke etcd cluster formation for some users.
Several doc and internal cleanups.

Fixed

openshift-install has improved error handling for various invalid command lines. It now errors when additional positional arguments are passed to commands that do not take positional arguments (previously those commands silently ignored the presence of positional arguments). And it logs an error and exits 1 when an invalid value is provided to --log-level (previously it exited 1 but did not write to the standard error stream).
The slow-input issues for the install-config wizard have been fixed.
On AWS, destroy cluster fixed a bug in the 0.10.1 refactor which could lead to leaked resources and a claim of successful deletion if a call to get tagged resources failed (for example, because the caller lacked the tag:GetResources permission).

On AWS, a new explicit dependency in the Terraform modules prevents errors like:

  * module.vpc.aws_lb.api_external: 1 error occurred:
  * aws_lb.api_external: Error creating Application Load Balancer: InvalidSubnet: VPC vpc-0765c67bbc82a1b7d has no internet gateway
  status code: 400, request id: 5a...d5

On libvirt, the installer no longer holds the OS image in memory after it has been written to disk. Ideally it would stream the OS image to disk instead of ever holding it in memory, but this fix mitigates our current in-memory buffering.

0.10.1 - 2019-01-22

Changed

create ignition-configs now also writes metadata.json to the asset directory, which allows Hive to more reliably destroy clusters.
destroy cluster now removes .openshift_install_state.json on success, clearing the way for future create cluster runs in the same asset directory.
On AWS, we now default to m4.xlarge masters. The increased CPU reduces etcd latencies, which in turn helps with cluster stability.
On AWS, the bootstrap machine has a new security-group allowing journald-gateway and kubelet access, for easier debugging when bootstrapping fails.
Several doc and internal cleanups.

Removed

The SSH public key is no longer inserted in the pointer Ignition configurations, now that authorized public keys are managed by the machine-config daemon.

Fixed

On AWS, the cluster-API provider now supports configuring machine volumes, so rootVolume settings in install-config.yaml will be respected.

On AWS, the generated Terraform variables no longer clobber master instance type and root volume configuration set via install-config.yaml. You can now use:

machines:
- name: master
  platform:
    aws:
      type: m5.large
      rootVolume:
        iops: 3000
        size: 220
        type: io1
  replicas: 3
- name: worker
  ...

and similar to successfully customize your master machines.

On AWS, destroy cluster has been adjusted to use more efficient tag-based lookup and fix several bugs due to previously-missing pagination. This should address some issues we had been seeing with leaking AWS resources despite destroy cluster claiming success.

0.10.0 - 2019-01-15

Added

The installer pushes an Infrastructure object to infrastructures.config.openshift.io with platform information. Cluster components should use this instead of the deprecated cluster-config-v1 resource.
openshift-install has a new completion subcommand, to generation shell-completion code (currently only for Bash).
On AWS, destroy cluster now also removed IAM users with the usual tags. We don't create these users yet, but the removal sets the stage for the coming credentials operator.

Changed

Install configuration now includes a new apiVersion property which must be set to v1beta1. Future changes to the install-config schema will result in new versions, allowing new installers to continue to support older install-config schema (and older installers to error out when presented with newer install-config schema). Changes to the schema since 0.9.0:
- clusterID has been removed. This should be a new UUID for every cluster, so there is no longer an easy way for users to configure it.
- Image configuration has been removed. Almost all users should be fine with the installer-chosen RHCOS. Users who need to override the RHCOS build (because they're testing new RHCOS releases) can set a new OPENSHIFT_INSTALL_OS_IMAGE_OVERRIDE environment variable.
- Per-platform network properties have been consolidated into the platform-agnostic machineCIDR from which node IP addresses are assigned.
- On libvirt, all machine-pool configuration has been removed, as no remaining properties were supported by the libvirt cluster-API provider.
install-config.yaml read during staged installs will now have installer-defaults applied for missing properties. This allows you to set only the properties you are interested in overriding, and allow the installer to manage the remaining properties.
create ignition-configs now also writes the admin kubeconfig to the asset directory, to support bring-your-own-infrastructure use cases.
The bootstrap node now serves journals for easier troubleshooting.
The validity for the initial kubelet TLS certificate has been increased from one hour to 24 hours, to give bring-your-own-infrastructure users longer to manually distribute the certificate before it expires.
The key for the root certificate authority is no longer pushed into the cluster (not even to the bootstrap node).
Machine(set)s generated by the installer now use providerSpec instead of the deprecated providerConfig.
On AWS, the load balancers now use HTTPS health checks to reduce log noise like:
```
  http: TLS handshake error from 10.0.20.86:28372: EOF
```
On AWS, IAM roles are now tagged with the usual resource tags (openshiftClusterID, etc.). Some other resources have had their tags updated to match those conventions (e.g. the internal Route 53 hosted zone was updated from KubernetesCluster to kubernetes.io/cluster/{name}: owned).
The OpenStack platform has been removed from the install-config wizard while it remains experimental. It is still available for users who supply their own install-config.yaml.
On OpenStack, the service VP now respects any SSH key specified in the install configuration.
On OpenStack, a developer-only internal DNS server has been removed, so users need to configure additional records for the existing external DNS zone.
On OpenStack, Neutron trunk ports are now used for VM network interfaces if Neutron supports them to support future Kuryr integration.
On OpenStack, masters and workers have been consolidated in a single subnet to simplify the deployment.
On OpenStack, the Ignition security group now only allows internal connections, and no longer allows connections from outside the cluster network.
On OpenStack, the machine(set) templates have been updated to set cloudName and some other properties.
On libvirt, destroy cluster is now more robust in the face of domains which were already shutdown.
Lots of doc and internal cleanup and minor fixes.

Removed

Support for install-config.yml (deprecated in 0.8.0) has been removed.

Fixed

On AWS, domain pagination for the wizard's base-domain select widget has been fixed. Previously, it would continuously fetch the first page of hosted zones (for accounts with multiple pages of zones) until it hit an error like:
```
ERROR list hosted zones: Throttling: Rate exceeded
        status code: 400, request id: ...
```
before falling back to a free-form base-domain input.

0.9.0 - 2019-01-05

Added

There is a new none platform for bring-your-own infrastructure users who want to generate Ignition configurations. The new platform is mostly undocumented; users will usually interact with it via OpenShift Ansible.

Changed

On OpenStack, there's no longer a default flavor, because flavor names are not standardized. Instead, there's a new prompt to choose among flavors supported by the target OpenStack implementation.
On libvirt, we now use the host-passthrough CPU type, which should improve performance for some use-cases.
Some doc and internal cleanup and minor fixes.

0.8.0 - 2018-12-23

Added

The installer binary now includes all required Terraform plugins, so there is no longer a need to separately download and install them. This will be most noticeable on libvirt, where users used to be required to install the libvirt plugin manually. This avoids issues with mismatched plugin versions (which we saw sometimes on libvirt) and network-connectivity issues during create cluster invocations (which we saw sometimes on all platforms).
The configured base domain is now pushed into the cluster's config.openshift.io as a DNS custom resource.

Changed

install-config.yml is now install-config.yaml to align with our usual YAML extension. install-config.yml is deprecated, and support for it will be removed completely in the next release.
On AWS, we now use a select widget for the base-domain wizard prompt, making it easier to choose an existing public zone.
On AWS, Route 53 rate limits during cluster destroy are now less disruptive, reducing the AWS request load in busy accounts.
On OpenStack, the HAProxy configuration no longer hard-codes the cluster name and base domain.

On OpenStack, the 0.7.0 fix for:

  FATAL Expected HTTP response code [202 204] when accessing [DELETE https://osp-xxxxx:13696/v2.0/routers/52093478-dcf1-4bcc-9a2c-dbb1e42da880], but got 409 instead
  {"NeutronError": {"message": "Router 52093478-dcf1-4bcc-9a2c-dbb1e42da880 still has ports", "type": "RouterInUse", "detail": ""}}

was incorrect and has been reverted. We'll land a real fix for this issue in future work.

On OpenStack, the service VM from 0.7.0 now has a floating IP address.
All libvirt functionality is behind TAGS=libvirt now. Previously installer builds with TAGS=libvirt_destroy included all libvirt functionality, while builds without that tag would include create cluster but not destroy cluster functionality. With the change, all users using the installer with libvirt clusters will need to set the new build tag.
Lots of doc and internal cleanup and minor fixes.

Removed

On AWS and OpenStack, the tectonicClusterID tag which was deprecated in 0.7.0 has been removed.

0.7.0 - 2018-12-14

Added

We now validate install-config when loading it during staged installs. Previously we only validated that input when it was entered into the wizard or via environment variables. This also leads to some changes in which values are considered valid:
- Base domains may no longer contain uppercase letters.
- Cluster names may now be longer than 63 characters, although as specified in RFC 1123 host software may not support names longer than 63 characters.
- Pull secrets require more content (e.g. it must contain an auths property). Previously we only required pull secrets to be valid JSON.
- SSH public keys must be parsable with ParseAuthorizedKey. Previously we had our own logic that was not as well developed.
We've added images/installer/Dockerfile.ci.rhel7 for building installer images on a RHEL base.
On AWS, we now create an S3 endpoint for the VPC.
We've added OpenStack documentation.

Changed

The pull-secret prompt now masks the input to avoid leaking it to shoulder surfers and similar.
The pull-secret prompt's help now points to cloud.openshift.com instead of try.openshift.com. This should make it easier to find the pull secret without digging through a page of introductory content.
The initial kubeconfig inserted on master nodes used to have admin privileges, but only for 30 minutes. Now it has role bindings that allow it to create and receive automatic approval for certificate signing requests, but it does not have additional privileges beyond that.
On AWS and OpenStack, master ports 10251 (scheduler) and 10252 (controller manager) have been opened to access from all machines. This allows Prometheus (which runs on the worker nodes) to scrape all machines for metrics.
On AWS and OpenStack, the installer and subsequent cluster will now tag resources it creates with openshiftClusterID. tectonicClusterID is deprecated.
On OpenStack, only the OpenStack clouds entry is marshalled into the openstack-creds secret. Previously we had injected the host's entire cloud configuration.
On OpenStack, there is now a service VM to provide DNS and load balancing for the OpenShift cluster. The service VM will eventually be removed, but for now its a convenient hack to get usable clusters on OpenStack.
On libvirt, we now document host DNS configuration as required, because too many users were skipping that step and then reporting errors with Kubernetes API detection when the install-host failed to resolve the cluster domain name while waiting for the bootstrap-complete event.
Lots of doc and internal cleanup and minor fixes.

Fixed

Fixed OpenShift manifest loading during staged installs. The installer had been ignoring changes to those files since 0.4.0.
Fixed you must pass a pointer as the target of a Write operation errors introduced in 0.6.0 for the AWS access key ID prompt.
When create cluster times out waiting for the Kubernetes API, we now exit immediately. Previously we'd wait through another 30 minutes of failed event-listener connections before failing this case. We've also fixed similar timeout detection for the code that waits for the OpenShift console route.

On OpenStack, we've fixed a bug in router deletion:

  FATAL Expected HTTP response code [202 204] when accessing [DELETE https://osp-xxxxx:13696/v2.0/routers/52093478-dcf1-4bcc-9a2c-dbb1e42da880], but got 409 instead
  {"NeutronError": {"message": "Router 52093478-dcf1-4bcc-9a2c-dbb1e42da880 still has ports", "type": "RouterInUse", "detail": ""}}

On libvirt, we've fixed a bug introduced in 0.6.0 and are now back to removing the bootstrap node from round-robin DNS when we destroy the bootstrap resources.

Removed

The user-facing OPENSHIFT_INSTALL_* environment variables are gone. Instead, users who want to skip the wizard are encouraged to provide their own install-config.
On AWS, the option to install a cluster into an existing VPC is gone. Users who would have previously done this can use VPC peering.

0.6.0 - 2018-12-09

Added

We now push a kubeadmin user (with an internally-generated password) into the cluster for the new bootstrap identity provider. This gives users a way to access the web console, Prometheus, etc. without needing to configure a full-fledged identity provider or install oc. The create cluster subcommand now blocks until the web-console route is available and then exits after printing instructions for using the new credentials.
The installer binary now includes Terraform so there is no longer a need to separately download and install it.

Changed

The SSH public key configuration has moved a level up in the install config, now that the admin structure has been removed.
build.sh now checks to make sure you have a new enough go, instead of erroring out partway through the build.
We now resolve the update payload to a digest on the bootstrap node, so the cluster-version-operator can figure out exactly which image we used.
Creation logging has been overhauled to increase it's signal-to-noise while waiting for the Kubernetes API to come up.
On AWS, the installer will now prompt you for an access key and secret if it cannot find your AWS credentials in the usual places.
On AWS, the installer will look at AWS_DEFAULT_REGION and in other usual places when picking a default for the region prompt. You still have to set OPENSHIFT_INSTALL_AWS_REGION if you want to skip the prompt entirely.
On libvirt, we've bumped masters from 3 GiB of memory to 4 GiB to address out-of-memory issues we had been seeing at 3 GiB.
Lots of doc and internal cleanup and minor fixes.

Removed

The old admin username and password inputs have been removed. They weren't being used anyway, and their intended role has been replaced by the newly-added kubeadmin user and bootstrap identity provider.
The old openshift-web-console namespace is gone. The new console is in the openshift-console namespace.

0.5.0 - 2018-12-03

Added

We now push the ingress custom resource definition and initial configuration, allowing the ingress operator to configure itself without referencing the deprecated cluster-config-v1 resource.

Changed

Pull secret documentation now points to try.openshift.com for pull-secret acquisition, instead of pointing at account.coreos.com. Users will need to update their pull secrets.
If the automatic bootstrap teardown (which landed in 0.4.0) times out waiting for the bootstrap-complete event, the installer exits with a non-zero exit code. We had ignored watcher timeouts in 0.4.0 due to concerns about watcher robustness, but the current watcher code has been reliable in our continuous integration testing.
The hard-coded quay.io/coreos/bootkube dependency has been replaced by the new cluster-bootstrap image, which is referenced from the release image.
The etcd service now uses selectors to determine the pods it exposes, and the explict etcd endpoints object is gone (replaced by the one Kubernetes maintains based on the selector).
On AWS, both masters and worker have moved from t2.medium nodes m4.large nodes (more on AWS instance types here) to address CPU and memory constraints.
On AWS, master volume size has been bumped from 30 GiB to 120 GiB to increase our baseline performance from on gp2's sliding IOPS scale from the 100 IOPS floor up to 360 IOPS. Volume information is not currently supported by the cluster-API AWS provider's AWSMachineProviderConfig, so this change is currently limited to masters created by the installer.
On Openstack, we now validate cloud, region, and image-name user input instead of blindly accepting entries.
On libvirt, we pass Ignition information for masters and workers via secrets instead of passing a libvirt volume path. This makes the libvirt approach consistent with how we already handle AWS and OpenStack.
Lots of internal cleanup, especially around trimming dead code.

Fixed

The .openshift_install.log addition from 0.4.0 removed Terraform output from --log-level=debug. We've fixed that in 0.5.0; now .openshift_install.log will always contain the full Terraform output, while standard error returns to containing the Terraform output if and only if --log-level=debug or higher.
On AWS teardown, errors retrieving tags for S3 buckets and Route 53 zones are no longer fatal. This allows the teardown code to continue it's exponential backoff and try to remove the bucket or zone later. It avoids some resource leaks we were seeing due to AWS rate limiting on those tag lookups as many simultaneous CI jobs searched for Route 53 zones with their cluster's tags. We'll still hit those rate limits, but they no longer cause us to give up on reaping resources.
On AWS, we've removed some unused data blocks, fixing occasional errors like:
```
  data.aws_route_table.worker.1: Your query returned no results.
```
On OpenStack, similar retry-during-teardown changes were made for removing ports and for removing subnets from routers.
On libvirt, Terraform no longer errors out when launching clusters configured for more than one master, fixing a bug from 0.4.0.

0.4.0 - 2018-11-22

Added

The creation targets have been moved below a new create subcommand (e.g. openshift-install create cluster instead of the old openshift-install cluster). This makes them easier to distinguish from other openshift-install subcommands and also mirrors the approach taken by destroy in 0.3.0.
A new manifest-templates target has been added to create, allowing users to edit templates and have descendant assets generated from their altered templates during a staged install.
The ingress operator is no longer masked. The old Tectonic ingress operator has been removed.
The the registry operator has been added, and the kube-addon operator which used to provide a registry (among other things) has been removed.
The checkpointer operator is no longer masked. It runs on the production cluster, but not on the bootstrap node.
Cloud credentials are now pushed into a secret where they can be consumed by cluster-API operators and other tools.
OpenStack now has destroy support.
We log verbosely to ${INSTALL_DIR}/.openshift_install.log for most operations, giving access to the logs for troubleshooting even if you neglected to run with --log-level=debug.
We've grown troubleshooting documentation.

Changed

The create cluster subcommand now waits for the bootstrap-complete event and automatically removes the bootstrap assets after receiving it. This means that after create cluster returns successfully, the cluster has its production control plane and topology (although there may still be operators working through their initialization). The bootstrap-complete event was new in 0.3.0, and it is now pushed at the appropriate time (it was too early in 0.3.0). The destroy bootstrap subcommand is still available, to allow users to manually trigger bootstrap deletion if the automatic removal fails for whatever reason.
On AWS, bootstrap deletion now also removes the S3 bucket used for the bootstrap node's Ignition configuration.
Asset state is preserved even while moving backwards through a staged install. For example:
```
openshift-install --dir=example create ignition-configs
openshift-install --dir=example create install-config
```
now preserves the full state including the generated Ignition configuration. In 0.3.0, the install-config call would have removed the Ignition configuration and other downstream assets from the stored state.
Some asset state is removed by successful destroy cluster runs. This reduces the change of contaminating future cluster creation with assets left over from a previous cluster, but users are still encouraged to remove state between clusters to avoid accidentally contaminating the subsequent cluster's state.
etcd discovery now happens via SRV records. On libvirt, this requires a new Terraform provider, so users with older providers should install a newer version. This also allows all masters to use a single Ignition file.
On AWS, the API and service load balancers have been changed from classic load balancers to network load balancers. This should avoid some latency issues we were seeing with classic load balancers, and network load balancers are cheaper.
On AWS, master Machine entries now include load balancer references, ensuring that new masters created by the AWS cluster-API provider will be attached to the load balancers.
On AWS and OpenStack, the default network CIDRs have changed to 172.30.0.0/16 for services and 10.128.0.0/14 for the cluster, to be consistent with previous versions of OpenStack.
The bootstrap kubelet is no longer part of the production cluster. This reduces complexity and keeps production pods off of the temporary bootstrap node.
The cluster-version operator now runs in a static pod on the bootstrap node until the production control plane comes up. This breaks a cyclic dependency between the production API server and operators.
The bootstrap control plane now waits for some core pods to come up before exiting.
The machine-API operator now reads the install-config from the cluster-config-v1 config-map, instead of from an operator-specific configuration.
AWS AMIs and libvirt images are now pulled from the new RHCOS pipeline.
Updated the security contact information for CoreOS -> Red Hat.
We push a ClusterVersion custom resource. The old CVOConfig is still being pushed, but it is deprecated.
OpenStack credentials are loaded from standard system paths.
On AWS and OpenStack, ports 9000-9999 are now open for host network services.
Lots of doc and internal cleanup and minor fixes.

Fixed

On AWS, destroy cluster is now more robust, removing resources with either the tectonicClusterID or kubernetes.io/cluster/<name>: owned tags. It also removes pending instances as well (it used to only remove running instances).
On libvirt, destroy cluster is now more precise, only removing resources which are prefixed by the cluster name.
Bootstrap Ignition edits (via create ignition-configs) no longer suffer from a worker.ign dependency cycle, which had been clobbering manual bootstrap.ign changes.
The state-purging implementation respects --dir, avoiding remove ...: no such file or directory errors during staged installs.
Cross-filesystem Terraform state recovery during destroy bootstrap no longer raises invalid cross-device link.
Bootstrap binaries are now located under /usr/local/bin, avoiding SELinux violations on RHEL 8.

Removed

All the old Tectonic operators and the tectonic-system namespace have been removed.
On libvirt, the image URI prompt has been removed. You can still control this via the OPENSHIFT_INSTALL_LIBVIRT_IMAGE environment variable, but too many users were breaking their cluster by pointing the installer at an outdated RHCOS, so we removed the prompt to make that knob less obvious.
On libvirt, we've removed .gz suffix handling for images. The new RHCOS pipeline supports Content-Encoding: gzip, so the suffix-based hack is no longer necessary.
The destroy-cluster command, which was deprecated in favor of destroy cluster in 0.3.0, has been removed.
The creation target subcommands of openshift-install have been removed. Use the target subcommands of create instead (e.g. openshift-install create cluster instead of openshift-install cluster).

0.3.0 - 2018-10-22

Added

Asset state is loaded from the install directory, allowing for a staged install.
A new openshift-install destroy bootstrap command destroys the bootstrap resources. Ideally, this would be safe to run after the new bootstrap-complete event is pushed to the kube-system namespace, but there is currently a bug causing that event to be pushed too early. For now, you're on your own figuring out when to call this command.

For consistency, the old destroy-cluster has been deprecated in favor of openshift-install destroy cluster.
The installer creates worker MachineSets, instead of leaving that to the machine-API operator.
The installer creates master Machines and tags masters to be picked up by the AWS cluster-API provider.

Changed

The installer now respects the AWS_PROFILE environment variable when launching AWS clusters.
Worker subnets are now created in the appropriate availability zone for AWS clusters.
Use the released hyperkube and hypershift instead of hard-coded images.
Lots of changes to keep up with the advancing release image, as OpenShift operators are added to control various cluster components.
Lots of internal cleanup and minor fixes.

Removed

The Tectonic kube-core operator, which has been replaced by OpenShift operators.

0.2.0 - 2018-10-12

Added

Asset state is preserved between invocations, allowing for a staged install like:
```
$ openshift-install --dir=example install-config
$ openshift-install --dir=example cluster
```
which creates a cluster using the same data given in the install-config (including the same random cluster ID, etc.).
The kube-apiserver and kube-controller-manager operators are called to render additional cluster manifests.
etcd is now available as a service in the kube-system namespace, and the new service is labeled so Prometheus will scrape it.
The service-serving-cert-signer-signing-key secret is now available in the openshift-service-cert-signer namespace, which gives the service-serving cert signer the keys it needs to mint and manage certificates for Kubernetes services.
The etcd-serving certificate is now passed through to the kube-controller-manager operator.
We disable some components which the cluster-version operator would otherwise install but which conflict with the legacy tectonic-operators.
The new openshift-install graph outputs the asset graph in the DOT language.
openshift-install version now outputs the Terraform version as well as the installer version.

Changed

The cluster-version operator is no longer run as a static pod. Instead, we just wait until the control plane comes up and run it them.
Terraform errors are logged to standard error even when --log-level is less than debug.
Terraform is now invoked with -no-color and -input=false.
The cluster target now includes both launching the cluster and populating metadata.json, regardless of whether the terraform invocation succeeds. This allows destroy-cluster to cleanup cluster resources even when the terraform invocation fails.
Reported errors now include more context, making them less enigmatic.
Libvirt image caching is more efficient, caching unzipped images with a cache that grows by one unzipped image per RHCOS release in $XDG_CACHE_HOME/openshift-install/libvirt/image. The previous implementation unzipped, when necessary, for every launched cluster, which was slow. And the previous implementation added one unzipped image to /tmp per cluster launch, which consumed more disk space.
Work continues on the OpenStack platform.
Lots of internal cleanup, especially around asset generation.

Removed

The operatorstatus CRD. Now the cluster-version operator creates this on its own.
The machine-config-operator-images config-map. Now the cluster-version operator pulls these from the machine-config images.
The machine-api app-version from the tectonic-system namespace.

0.1.0 - 2018-10-02

Added

The openshift-install command. This moves us to the new install-config approach with asset generation in Go instead of in Terraform. Terraform is still used to push the assets out to resources on the backing platform (AWS, libvirt, or OpenStack), but that push happens in a single Terraform invocation instead of in multiple steps. This makes installation faster, because more resources can be created in parallel. openshift-install also dispenses with the distribution tarball; all required assets except for a terraform binary are distributed in the openshift-install binary.

The configuration and command-line interface are quite different, so previous tectonic users are encouraged to start from scratch when getting acquainted with openshift-install. AWS users should look here. Libvirt users should look here. The new openshift-install also includes an interactive configuration generator, so you can launch the installer and follow along as it guides you through the process.

Removed

The tectonic command and tarball distribution are gone. Please use the new openshift-install command instead.

Files

CHANGELOG.md

Latest commit

History

CHANGELOG.md

File metadata and controls

Changelog

0.14.0 - 2019-03-05

Changed

Fixed

0.13.1 - 2019-02-28

Changed

Fixed

Removed

0.13.0 - 2019-02-26

Added

Changed

Fixed

Removed

0.12.0 - 2019-02-05

Changed

Fixed

Removed

0.11.0 - 2019-01-27

Added

Changed

Fixed

0.10.1 - 2019-01-22

Changed

Removed

Fixed

0.10.0 - 2019-01-15

Added

Changed

Removed

Fixed

0.9.0 - 2019-01-05

Added

Changed

0.8.0 - 2018-12-23

Added

Changed

Removed

0.7.0 - 2018-12-14

Added

Changed

Fixed

Removed

0.6.0 - 2018-12-09

Added

Changed

Removed

0.5.0 - 2018-12-03

Added

Changed

Fixed

0.4.0 - 2018-11-22

Added

Changed

Fixed

Removed

0.3.0 - 2018-10-22

Added

Changed

Removed

0.2.0 - 2018-10-12

Added

Changed

Removed

0.1.0 - 2018-10-02

Added

Removed