Commit Graph

1280 Commits

Author SHA1 Message Date
Dalton Hubble 805dd772a8 Run etcd cluster on-host, across controllers on AWS
* Change controllers ASG to heterogeneous EC2 instances
* Create DNS records for each controller's private IP for etcd
* Change etcd to run on-host, across controllers (etcd-member.service)
* Reduce time to bootstrap a cluster
* Deprecate self-hosted-etcd on the AWS platform
2017-11-06 01:03:53 -08:00
Dalton Hubble c6ec6596d8 Minor cleanup for zones, docs, and outputs
* Spread across all zones, regardless of UP/DOWN state
* Remove unused outputs of private IPs
2017-11-06 00:56:26 -08:00
Dalton Hubble 47a9989927 Fix null_resource ordering constraints
* Ensure etcd TLS assets and kubeconfig are copied before
any attempt is made to run bootkube start
2017-11-06 00:55:44 -08:00
Dalton Hubble 10b977d54a addons: Set kube-state-metrics to have clusterIP None
* kube-state-metrics service exists to facilitate prometheus discovery
2017-11-05 17:54:09 -08:00
Dalton Hubble b7a268fc45 addons: Add prometheus alertmanager flag
* Pass -alertmanager.url to work with a user's in-cluster
alertmanager deployment, if any
2017-11-05 15:50:46 -08:00
Dalton Hubble 279f36effd addons: Add grafana 4.6.1 and extend prometheus docs 2017-11-05 15:23:56 -08:00
Dalton Hubble 77fc14db71 Workaround target pool issue by listing instances as zone/name
* Instances can be listed by zone/name or self_link URL, but the
provider desires they be in zone/name form, which causes a diff
* https://github.com/terraform-providers/terraform-provider-google/issues/46
2017-11-05 14:07:05 -08:00
Dalton Hubble 2b0296d671 Create controller instances across zones in the region
* Change controller instances to automatically span zones in a region
* Remove the `zone` required variable
2017-11-05 13:24:32 -08:00
Dalton Hubble 7b38271212 Run etcd cluster on-host, across controllers on Google Cloud
* Change controllers from a managed group to individual instances
* Create discrete DNS records to each controller's private IP for etcd
* Change etcd to run on-host, across controllers (etcd-member.service)
* Reduce time to bootstrap a cluster
* Deprecate self-hosted-etcd on the Google Cloud platform
2017-11-05 11:03:35 -08:00
Dalton Hubble ae07a21e3d addons: Omit static resource requests/limits for kube-state-metrics
* Allow the addon-resizer to dynamically set resource values
* https://github.com/kubernetes/kube-state-metrics/pull/285
2017-11-04 14:41:04 -07:00
Dalton Hubble 0ab1ae3210 addons: Fix typo in kube-state-metrics strategy 2017-11-04 14:39:56 -07:00
Dalton Hubble 67e3d2b86e docs: GCE network bandwidth is excellent, even btw zones
* Remove performance note that the GCE vs AWS network performance
is not an equal comparison. On both platforms, workers now span the
(availability) zones of a region.
* Testing host-to-host and pod-to-pod network bandwidth between nodes
(now located in different zones) showed no reduction in bandwidth
2017-11-04 14:08:20 -07:00
Dalton Hubble a48dd9ebd8 Require google provider version ~> 1.1
* Require google provider plugin 1.1 or higher which includes fix:
https://github.com/terraform-providers/terraform-provider-google/issues/574
* Remove workaround which statically set the persistent disk name
* Original reasons for workaround in a97df839 or GH #34
2017-11-04 12:59:19 -07:00
Dalton Hubble 26a291aef4 Remove controller_preemptible option on Google Cloud
* Controller preemption is not safe or covered in documentation. Delete
the option, the variable is a holdover from old experiments
* Note, worker_preemeptible is still a great feature that's supported
2017-11-04 12:59:19 -07:00
Dalton Hubble 251a14519f Fix typo in internal template variable name
* ssh_authorized_keys should be ssh_authorized_key to match the user
facing variable which only allows a single SSH authorized key
2017-11-04 12:59:19 -07:00
Dalton Hubble 6300383b43 Change worker managed instance group to span zones in region
* Change Google Cloud module to require the `region` variable
* Workers are created in random zones within the given region
* Tolerate Google Cloud zone failures or capacity issues
* If workers are preempted (if enabled), replacement instances can
be drawn from any zone in the region, which should avoid scheduling
issues that were possible before if a single zone aggressively
preempts instances (presumably due to Google Cloud capacity)
2017-11-04 12:59:19 -07:00
Dalton Hubble e32885c9cd addons: Update prometheus from v1.8.0 to v1.8.2
* https://github.com/prometheus/prometheus/releases/tag/v1.8.2
2017-11-04 11:00:39 -07:00
Dalton Hubble fe8afdbee9 Update Typhoon logo and favicon 2017-11-04 01:20:17 -07:00
Dalton Hubble 878f5a3647 Bump bootkube and terraform-render-bootkube to v0.8.1
* Use the v0.8.1 tagged terraform-render-bootkube module
* Use the v0.8.1 quay.io/coreos/bootkube image to bootstrap
2017-10-28 12:50:37 -07:00
Dalton Hubble 34ec7e9862 Relax pessimistic constraints on 1.0+ providers
* Constrains ~> 1.0 means users can use 1.0.1, 1.1, but not 2.0
* https://www.terraform.io/docs/configuration/terraform.html
2017-10-25 23:27:28 -07:00
Dalton Hubble f6c6e85f84 Require minimum Terraform and plugin versions
* Bump minimum Terraform version to v0.10.4
* Allow minor version updates for 1.0+ plugins
* Fix versions for plugins which are pre-1.0
2017-10-25 23:00:31 -07:00
Dalton Hubble 8582e19077 Expand Nginx Ingress liveness and readiness probes
* Remove dnsPolicy: ClusterFirst
* https://github.com/kubernetes/ingress-nginx/pull/1584
2017-10-25 22:29:20 -07:00
Dalton Hubble 3727c40c6c Update Nginx Ingress defaultbackend from 1.0 to 1.4
* https://github.com/kubernetes/ingress-nginx/pull/1568
2017-10-25 22:16:23 -07:00
Dalton Hubble b608f9c615 addons: Use service endpoints to scrape node-exporter 2017-10-24 22:59:00 -07:00
Dalton Hubble ec1dbb853c addons: Include kube-state-metrics exporter manifests 2017-10-24 22:59:00 -07:00
Dalton Hubble d046d45769 addons: Include Prometheus and node-exporter manifests 2017-10-24 22:58:59 -07:00
Dalton Hubble a73f57fe4e Update CLUO from v0.4.0 to v0.4.1 2017-10-24 22:14:03 -07:00
Dalton Hubble 60bc8957c9 Update Kubernetes from v1.8.1 to v1.8.2
* Kubernetes v1.8.2 fixes a memory leak in the v1.8.1 apiserver
* Switch to using the `gcr.io/google_containers/hyperkube` for the
on-host kubelet and shutdown drains
* Update terraform-render-bootkube manifests generation
  * Update flannel from v0.8.0 to v0.9.0
  * Add `hairpinMode` to flannel CNI config
  * Add `--no-negcache` to kube-dns dnsmasq
2017-10-24 21:44:26 -07:00
Dalton Hubble 8b78c65483 Update Google Cloud Kubernetes from v1.7.7 to v1.8.1 2017-10-20 16:09:11 -07:00
Dalton Hubble f86c00288f Add missing update-agent RBAC role to get pods
* Drain now gets pods, deletes pods, and waits for deletion
2017-10-20 01:21:46 -07:00
Dalton Hubble a57b3cf973 Update CLUO addon to v0.4.0 and RBAC ClusterRole 2017-10-20 00:40:17 -07:00
Dalton Hubble 10c5487ad7 Add docs corrections for versions and log output 2017-10-20 00:39:17 -07:00
Dalton Hubble e4c479554c Update AWS, DO, BM Kubernetes from v1.7.7 to v1.8.1
* Update from bootkube v0.7.0 to v0.8.0
* Leave Google Cloud update to a followup commit
2017-10-19 21:10:04 -07:00
Dalton Hubble be113e77b4 Fix links and add Calico BGP peering notes 2017-10-17 19:10:18 -07:00
Dalton Hubble 911c53e4ae Add Ubiquity EdgeRouter documentation 2017-10-17 18:51:40 -07:00
Dalton Hubble bfa8dfc75d Conditionally set networkd content on bare-metal
* Without this change, if a cluster doesn't set the controller
or worker networkd lists, an err "element() may not be used
with an empty list" occurs.
* controller_networkds and worker_networks are intended to be
optional and temporary, not required at all
2017-10-17 18:47:12 -07:00
Dalton Hubble 43dc44623f Fix the terraform fmt of configs 2017-10-16 01:32:25 -07:00
Dalton Hubble 734bc1d32a Add performance benchmark for flannel with bonded NICs 2017-10-16 01:12:13 -07:00
Dalton Hubble 41e632280f Remove unused storage section ala PXE-only Matchbox templating 2017-10-16 00:42:20 -07:00
Dalton Hubble fc22f04dd6 Add temporary variables for multi-nic testing
* Accept ordered lists of controller and worker networkd configs
* Do not rely on these variables. They will be replaced with a
cleaner mechanism at a future date
2017-10-16 00:39:58 -07:00
Dalton Hubble 377e14c80b Fix ingress addon docs recursive apply command 2017-10-16 00:29:04 -07:00
Dalton Hubble 9ec8ec4afc Secure copy etcd TLS credentials to controllers only
* Controllers receive etcd TLS credentials
* Controllers and workers receive a kubeconfig
2017-10-14 20:48:02 -07:00
Dalton Hubble 5c1ed37ff5 Add SSH key to user "debug" during disk-install phase
* Avoid adding SSH authorized key for user "core" during the disk
install, so that terraform apply cannot SSH until post-install
2017-10-14 20:37:42 -07:00
bzub e765fb310d Allow setting custom PXE boot kernel_args on bare-metal 2017-10-14 19:39:10 -07:00
Dalton Hubble 7b5ffd0085 Add Container Linux reboot-coordinator RBAC
* Add a reboot-coordinator namespace for CLUO components
* Define an RBAC ClusterRole for update-operator and update-agent
* Replace the older-style where CLUO ran in kube-system, with
admin privilege
2017-10-14 19:35:06 -07:00
Dalton Hubble 123439c2a4 Remove or compress docs image assets 2017-10-14 19:12:22 -07:00
Dalton Hubble 11453bac91 Update heapster addon from v1.4.0 to v1.4.3
* Use normal name and phase labels
2017-10-14 19:07:37 -07:00
Dalton Hubble dd0c61d1d9 Update Nginx Ingress controller addon to 0.9.0-beta.15 2017-10-14 18:30:58 -07:00
Dalton Hubble 5c87529011 Demote Google Cloud from stable to beta
* See #34 postmortem and action items for context on
when stable status will be restored
2017-10-11 19:32:04 -07:00
Dalton Hubble a97df839ea google-cloud: Set disk.device_name to match API default
* Terraform provider "google" plugin releases leave the disk
device_name as "" by default. Recently the API has started to
set a default name "persistent-disk-0". Plan and apply show
all instance groups need to be recreated to "fix" the name
* Impact: Controller and worker instance groups are deleted
and recreated, deleting data on controllers and bringing
down clusters
* Fix: Explicitly set the disk_name to persistent-disk-0 so
that terraform finds no diff needs to be applied.
* https://github.com/poseidon/typhoon/issues/34
* https://github.com/terraform-providers/terraform-provider-google/issues/574
2017-10-11 18:04:39 -07:00