Commit graph

238 commits

Author SHA1 Message Date
Aleksey Kasatkin
14749df6f3 Fix "netchecker-server" ClusterRole (#4730)
* Add sha256 hashes for calicoctl v3.6.1

Hashes are added to calicoctl_binary_checksums for both adm and arm platforms.

* Add rules for "network-checker.ext" resource to "netchecker-server" ClusterRole

So that it could access the resource after it is created.

Corresponding issues:
https://github.com/Mirantis/k8s-netchecker-server/issues/125
https://github.com/kubernetes-sigs/kubespray/issues/3281
2019-05-09 01:30:49 -07:00
Andreas Krüger
8a5eae94ea Minor cleanups of CoreDNS issues and CI job (#4719)
* Minor cleanups

* Add comment in docs that nodelocaldns cache is enabled by default
2019-05-07 13:20:36 -07:00
MarkusTeufelberger
e67f848abc ansible-lint: add spaces around variables [E206] (#4699) 2019-05-02 14:24:21 -07:00
Matthew Mosesohn
fc072300ea Purge legacy cleanup tasks from older than 1 year (#4450)
We don't need to support upgrades from 2 year old installs,
just from the last major version.

Also changed most retried tasks to 1s delay instead of longer.
2019-04-24 00:08:05 -07:00
Andreas Krüger
d588532c9b Update probe timeouts, delays etc. (#4612)
* Fix merge conflict

* Add check delay

* Add more liveness and readiness options to metrics-server
2019-04-23 14:46:02 -07:00
Maxime Guyot
50751bb610 Revert "Optimize kube resources creation (#4572)" (#4621)
This reverts commit f8fdc0cd93.
2019-04-23 20:37:23 +03:00
andreyshestakov
f8fdc0cd93 Optimize kube resources creation (#4572) 2019-04-22 23:34:10 -07:00
Maxime Guyot
37eac010c8 ansible-lint: Don’t compare to literal True/False (#4499) 2019-04-17 08:42:03 -07:00
Maxime Guyot
ec3daedf9e Revert "Fix for unknown 'kubernetes.io' or 'k8s.io' labels specified with --node-labels (#4320)" (#4553)
This reverts commit 586ad89d50.
2019-04-17 07:58:06 -07:00
Andreas Krüger
15597aa493 Do not force TCP connections to upstreams. (#4492) 2019-04-10 12:40:09 -07:00
Robert Neumann
586ad89d50 Fix for unknown 'kubernetes.io' or 'k8s.io' labels specified with --node-labels (#4320)
* Fix the file path for all.yml and k8s-cluster.yml

* Fix --node-labels namespace error "unknown labels specified"

* Update templates and configs kubelet node-labels
2019-04-10 04:14:12 -07:00
Sidharth Anupkrishnan
6caa639243 Update CoreDNS label as specified in the kubernetes coredns repository (#3920) 2019-04-10 04:12:13 -07:00
Andreas Krüger
4d39c1856e Fix jinja filters (#4470) 2019-04-09 02:19:06 -07:00
Xinghong Fang
d711a0c83f [nodelocaldns] expand tolerations on the daemonset (#4451) 2019-04-08 02:24:26 -07:00
Andreas Krüger
7a72e567d5 Update CoreDNS to 1.4.0 (#4422)
* Update CoreDNS to 1.4.0

* Update readme to reflect CoreDNS update
2019-04-04 00:40:50 -07:00
Andreas Krüger
3c050be0b0 Update nodelocaldns cache settings (#4423) 2019-04-04 00:38:51 -07:00
Andreas Krüger
41e684eb5a Update DNS Autoscaler to 1.4.0 (#4425)
* Update DNS Autoscaler

* Update downloads too

* Fix yamllint

* Fix yamllint
2019-04-04 00:36:51 -07:00
Matthew Mosesohn
5f12b7aedf Remove kubedns and dnsmasq. Move dns_late phase after apps (#4406)
Both kubedns and dnsmasq modes are long not maintained.
We should run dns_late steps at the end because sshd
makes DNS lookups during Ansible run and has 2s timeouts
for each failed lookup trying to connect to coredns before
it is ready.
2019-04-01 12:32:34 -07:00
Stefan Prietl
2fb27c8521 Use static files in KubeDNS templating task (#4379)
This commit adapts the "Lay Down KubeDNS Template" task to use the static
files moved by pull request [1]

[1] https://github.com/kubernetes-sigs/kubespray/pull/4341
2019-03-28 06:26:43 -07:00
Matthew Mosesohn
6d7f3c4405 Reduce jinja2 filters in coredns templates (#4390) 2019-03-26 11:09:17 -07:00
Manuel Cintron
3c4cbf133e Adding ability to override dashboard replica count (#4344) 2019-03-13 13:58:25 -07:00
Matthew Mosesohn
fd2c47b56a Move most coredns templates to static files (#4341)
* Move most coredns templates to static files

This should speed up the task slightly

* yaml lint fixes
2019-03-12 21:17:31 -07:00
Vasilis Remmas
81801ce23b Add master toleration flag in dashboard deployment (#4290) 2019-02-25 19:34:47 -08:00
Maxime Guyot
323d788f48 Add support for --enable-skip-login in Dashboard (#4265) 2019-02-19 23:24:29 -08:00
Thomas Nys
68fd7e39da Set cluster DNS correctly in case of nodelocal dns cache (#3879)
* Set cluster DNS correctly in case of nodelocal dns cache

* Pass in cluster_ip based on dns mode

* Disable nodelocaldns by default

* Fix syntax error

* Fix syntax issue

* Add nodelocadns ip to vars of node installation

* Change location of nodelocaldns_ip

* Try to remove newlines from jinja template

* Add debug for config file

* Move parameter logic outside of template

* Adapt templates after feedback

* Remove debugging
2019-01-28 23:39:27 -08:00
Danny Kulchinsky
595d6427ac [Nodelocal DNS cache] Mount host /run/xtables.lock in nodelocaldns container (#4074)
* Mount host /run/xtables.lock in nodelocaldns container

* fix typo in nodelocaldns daemonset manifest yml

* Add prometheus scrape annotation, updateStrategy and reduce termination grace period

* fix indentation

* actually fix it..

* Bump k8s-dns-node-cache tag to 1.15.1 (fixes https://github.com/kubernetes/dns/issues/282)
2019-01-28 01:57:40 -08:00
Chad Swenson
5c52a830d2 Update kubernetes dashboard to latest patch (#3995) 2019-01-08 09:46:20 -08:00
Ryler Hockenbury
3e8f4c1545 Use recommended defaults for dns autoscale (#3884) 2018-12-12 05:05:46 -08:00
Thomas Nys
3e3ee0aeb1 Add support for running a nodelocal dns cache (#3861)
* Add support for running a nodelocal dns cache

After encountering dns issues in a cluster I was recently working on I
noticed Kubernetes 1.13 introduced support for running a nodelocal dns
cache.

I believe this can usefull for more people.

73b548db06
https://github.com/kubernetes/enhancements/blob/master/keps/sig-network/0030-nodelocal-dns-cache.md

* Add requested changes

* Add additional requested changes + documentation

* Add requested changes after review

* Replace incorrect variable
2018-12-10 17:28:03 -08:00
Julien C
593a9a262d Add metrics service to kube-dns (#3852)
Metrics port is exposed through a service for CoreDNS but not for kube-dns.
2018-12-10 01:45:00 -08:00
Andreas Krüger
d5ce5874e8 Streamline path to certs dir (#3836)
* Streamline path to certs dir

* More fixes

* Set path to etcd certs in kubernetes defaults instead
2018-12-06 23:11:53 -08:00
Rong Zhang
225f765b56 Upgrade kubernetes to v1.13.0 (#3810)
* Upgrade kubernetes to v1.13.0

* Remove all precense of scheduler.alpha.kubernetes.io/critical-pod in templates

* Fix cert dir

* Use kubespray v2.8 as baseline for gitlab
2018-12-06 12:11:48 -08:00
Andreas Krüger
ddffdb63bf Remove non-kubeadm deployment (#3811)
* Remove non-kubeadm deployment

* More cleanup

* More cleanup

* More cleanup

* More cleanup

* Fix gitlab

* Try stop gce first before absent to make the delete process work

* More cleanup

* Fix bug with checking if kubeadm has already run

* Fix bug with checking if kubeadm has already run

* More fixes

* Fix test

* fix

* Fix gitlab checkout untill kubespray 2.8 is on quay

* Fixed

* Add upgrade path from non-kubeadm to kubeadm. Revert ssl path

* Readd secret checking

* Do gitlab checks from v2.7.0 test upgrade path to 2.8.0

* fix typo

* Fix CI jobs to kubeadm again. Fix broken hyperkube path

* Fix gitlab

* Fix rotate tokens

* More fixes

* More fixes

* Fix tokens
2018-12-06 02:33:38 -08:00
Chad Swenson
145687a48e Reduce log spam of verbose tasks (#3806)
Added a loop_control label to a few tasks that flood our logs.
2018-12-04 10:35:44 -08:00
Andreas Krüger
9463b70edd Cleanup defaults file from kubernetes-apps and add dashboard to download role (#3800)
* Remove variables defined in download role. Fixes #3799

* Cleanup some more variables

* Fix bad templating

* Minor fix

* Add dashboard to download role. Fixes #3736
2018-12-03 10:29:42 -08:00
Egor
3fa81bb86e Fix dns-autoscaler nodeAffinity: set to empty (#3747) 2018-11-22 05:29:09 -08:00
Egor
5daadc022d Fix: nodeAffinity for coredns-deployment and kubedns-deployment (#3746) 2018-11-22 05:27:25 -08:00
Andreas Krüger
8c18f053aa Fix DNS Autoscaler for coredns_dual deployment (#3726)
* Fix DNS Autoscaler for coredns_dual deployment

* Fix templating

* Fix templating again
2018-11-19 00:35:53 -08:00
marcstreeter
c83bfc9df6 fix dns_prevent_single_point_failure variable (#3728)
comparison that happens during `TASK [kubernetes-apps/ansible : Kubernetes Apps | Lay Down CoreDNS Template]` where the `dns-autoscaler` template is deployed causes coredns to fail deployment.  The error is caused by the variable `dns_prevent_single_point_failure` where an integer is being compared with a string. The resulting error:

```bash
'>' not supported between instances of 'int' and 'str'
```

prevents successful deployment of CoreDNS.  

The change makes the comparison happen between integers and allows CoreDNS to succeed.
2018-11-16 23:57:47 -08:00
Andreas Krüger
6f6274d0d9 Update CoreDNS, KubeDNS and Autoscaler to newest templates (#3711)
* Update DNS Autoscaler to latest

* Update CoreDNS to latest

* Update KubeDNS to latest

* Add KubeDNS config map

* Fix filename

* Add missing selector to DNS Autoscaler

* Add missing tolerations
2018-11-15 09:52:12 -08:00
Andreas Krüger
17f07e2613 Enable DNS AutoScaler for CoreDNS (#3707)
* Enable AutoScaler for CoreDNS

* Only use one template for dns autoscaler

* Rename a few variables for replicas and minimum pods

* Rename a few variables for replicas and minimum pods

* Remove replicas to make autoscale work

* Cleanup kubedns-autoscaler as it has been renamed
2018-11-15 01:28:03 -08:00
Ryler Hockenbury
d8e9b0f675 Netchecker version and namespace (#3705)
* Revert netchecker image and version

* Create namespace for netchecker

* Remove extra slashes
2018-11-14 09:27:45 -08:00
Ryler Hockenbury
e8901a2422 Apply linux node selector to coreDNS deployment (#3688)
* Apply linux node selector to coreDNS deployment

* Remove comment before linux node selector on manifests

* mend
2018-11-13 04:54:15 -08:00
Erwan Miran
7bec169d58 Fix ansible syntax to avoid ansible deprecation warnings (#3512)
* failed

* version_compare

* succeeded

* skipped

* success

* version_compare becomes version since ansible 2.5

* ansible minimal version updated in doc and spec

* last version_compare
2018-10-16 15:33:30 -07:00
Andreas Krüger
cdce8c81da Update CoreDNS templates to newest version and fix kubedns-autoscaler (#3483)
* Update CoreDNS templates to newest version

* Add watch to ClusterRole. Fixes #3460
2018-10-11 00:12:58 -07:00
Andreas Krüger
7e195b06a6 Fix DNS loop when resolvconf_mode is set to host_resolvconf (#3390)
* Fix DNS loop when resolvconf_mode is set to host_resolvconf

* Make sure upstream_dns_servers is defined when using resolvconf_mode == 'host_resolvconf'

* Only set upstream dns servers on KubeDNS and CoreDNS if they are defined

* Only set upstream dns servers on KubeDNS and CoreDNS if they are defined
2018-10-08 07:08:51 -07:00
Matthew Mosesohn
4b7d59224d Fix tag based deploy of apps by skipping kubeadm dns tasks (#3462) 2018-10-08 01:22:57 -07:00
Mikael Berthe
b4989b5a2a Fix netcheck agent/server image variable names (#3417)
According to the documentation, container images are described
by vars like `foo_image_repo` and `foo_image_tag`.
The variables netcheck_{agent,server}_{img_repo,tag} do not
follow that convention.
2018-09-29 20:44:01 -07:00
Rong Zhang
0232e755f3 Upgrade kubedns and kubednsautoscaler (#3407) 2018-09-28 01:20:08 -07:00
Kuldip Madnani
36898a2c39 Adding pod priority for all the components. (#3361)
* Changes to assign pod priority to kube components.

* Removed the boolean flag pod_priority_assignment

* Created new priorityclass k8s-cluster-critical

* Created new priorityclass k8s-cluster-critical

* Fixed the trailing spaces

* Fixed the trailing spaces

* Added kube version check while creating Priority Class k8s-cluster-critical

* Moved k8s-cluster-critical.yml

* Moved k8s-cluster-critical.yml to kube_config_dir
2018-09-25 07:50:22 -07:00
k8s-ci-robot
7f91f6e034
Merge pull request #3287 from Kami-no/coredns_metrics
Monitor CoreDNS over svc
2018-09-16 23:39:59 -07:00
k8s-ci-robot
e6a2e34dd1
Merge pull request #3315 from riverzhang/upgrade-kubedns
Upgrade kubedns to 1.14.11
2018-09-15 02:08:20 -07:00
rongzhang
934d92f09c Upgrade kubedns to 1.14.11 2018-09-15 15:22:38 +08:00
Zinin D.A
29c7775ea1 Monitor CoreDNS over svc 2018-09-12 10:24:15 +03:00
Erwan Miran
e24b1220a0 Fix wrong sa name in crb when psp is enabled 2018-09-11 15:04:55 +02:00
k8s-ci-robot
db11394711
Merge pull request #3200 from pablodav/feature/k8s_win_v1.11
Required support to start working on windows node support
2018-09-03 04:51:23 -07:00
Pablo Estigarribia
7cbe3c2171 ensure there is pin priority for docker package to avoid upgrade of docker to incompatible version
ensure there is pin priority for docker package to avoid upgrade of docker to incompatible version

remove empty when line

ensure there is pin priority for docker package to avoid upgrade of docker to incompatible version

force kubeadm upgrade due to failure without --force flag

ensure there is pin priority for docker package to avoid upgrade of docker to incompatible version

added nodeSelector to have compatibility with hybrid cluster with win nodes, also fix for download with missing container type

fixes in syntax and LF for newline in files

fix on yamllint check

ensure there is pin priority for docker package to avoid upgrade of docker to incompatible version

some cleanup for innecesary lines

remove conditions for nodeselector
2018-09-02 12:47:06 -03:00
Antoine Legrand
2f1fe44762 update images to use arch 2018-08-31 13:45:08 +02:00
Arslanbekov Denis
fe1e758856 Up dashboard version to 1.10.0 2018-08-28 14:10:19 +03:00
Erwan Miran
a6a14e7f77 create the service account and roles even if the rbac is not enabled. it will just be ignored 2018-08-22 18:17:11 +02:00
Erwan Miran
80cfeea957 psp, roles and rbs for PodSecurityPolicy when podsecuritypolicy_enabled is true 2018-08-22 18:16:13 +02:00
Wong Hoi Sing Edison
c3b3572025 Always create service account even rbac_enabled = false 2018-08-22 11:41:29 +08:00
Andreas Krüger
b3e32c1393
Merge pull request #3094 from hedayat/master
Add --dns-loop-detect to dnsmasq used in kube-dns
2018-08-20 09:27:15 +02:00
rongzhang
35efc387c4 Fix pull dns image error 2018-08-19 22:47:17 +08:00
Antoine Legrand
26bf719a02
Merge branch 'master' into multi-arch-support 2018-08-17 16:35:50 +02:00
Chad Swenson
3a85a2f81c
Merge pull request #3080 from mirwan/netchecker_template_rendering_filename
Netchecker manifests should not have j2 extension
2018-08-14 13:24:16 -05:00
Hedayat Vatankhah
c0221c2e72 Add --dns-loop-detect to dnsmasq used in kube-dns
It prevents DNS loops when host's DNS server is a localhost DNS server,
or when DNS server of cluster is also added as an upstream DNS server
2018-08-12 20:36:33 +04:30
Cédric de Saint Martin
e3dcd96301 kubedns & kubedns-autoscaler: Stick to master nodes. (#2909)
* kubedns & kubedns-autoscaler: Stick to master nodes.

 - Tolerate only master nodes and not any NoSchedule taint
 - Pods are on different nodes
 - Pods are required to be on a master node.

* kubedns: use soft nodeAffinity.

Prefer to be on a master node, don't require.

* coredns: Stick to (different) master nodes.

     - Pods are on different nodes
     - Pods are preferred to be on a master node.
2018-08-09 10:42:53 -05:00
Erwan Miran
494ff9522b j2 extension should only be used for template filename, not target file on remote host 2018-08-09 11:29:45 +02:00
Mathieu Herbert
d285565475 Add tags for coredns and kubedns 2018-08-07 20:55:38 +02:00
DBLaci
b61c64a8ea token-ttl default value is int in seconds 2018-07-19 12:15:47 +02:00
DBLaci
cb91003cea dashboard_token_ttl option override possibility with default 2018-07-13 15:26:18 +02:00
Derek Lemon
1e98e8444e Using dns domain instead of cluster name for coredns, incase they differ 2018-06-13 18:52:35 +00:00
Di Xu
1081f620d2 add support for non-amd64 arch gcr.io images
Currently all the gcr.io images used in kubespray can only run on x86.
Also gcr.io has not fully support multi-arch docker images.

Add extra var "image_arch" (default is amd64) to support running other
platforms, like arm64.

Change-Id: I8e1c9af533c021cb96ade291a1ce58773b40e271
2018-06-05 17:29:02 +08:00
Julien Girardin
f88cd27686 Add dashboard url as part of kubectl cluster-info output 2018-05-28 11:46:11 +02:00
rongzhang
742a8782dd Bump kube-dns to 1.14.10
Upgrade kube-dns to 1.14.10
https://github.com/kubernetes/kubernetes/tree/master/cluster/addons/dns
2018-05-15 03:29:10 +00:00
Cédric de Saint Martin
44cb126e7d Update netchecker to v1.2.2.
Using official image from mirantis at dockerhub.
2018-04-24 09:13:56 +02:00
Matthew Mosesohn
03bcfa7ff5
Stop templating kube-system namespace and creating it (#2545)
Kubernetes makes this namespace automatically, so there is
no need for kubespray to manage it.
2018-03-30 14:29:13 +03:00
Kuldip Madnani
daeeae1a91 Added retries in pre-upgrade.yml and retries while applying kube-dns.yml (#2553)
* Added retries in pre-upgrade.yml and retries while applying kube-dns.yml

* Removed trailing spaces
2018-03-29 11:37:32 -05:00
woopstar
e40368ae2b Add CoreDNS support with various fixes
Added CoreDNS to downloads

Updated with labels. Should now work without RBAC too

Fix DNS settings on hosts

Rename CoreDNS service from kube-dns to coredns

Add rotate based on http://edgeofsanity.net/rant/2017/12/20/systemd-resolved-is-broken.html

Updated docs with CoreDNS info

Added labels and fixed minor settings from official yaml file: https://github.com/kubernetes/kubernetes/blob/release-1.9/cluster/addons/dns/coredns.yaml.sed

Added a secondary deployment and secondary service ip. This is to mitigate dns timeouts and create high resitency for failures. See discussion at 'https://github.com/coreos/coreos-kubernetes/issues/641#issuecomment-281174806'

Set dns list correct. Thanks to @whereismyjetpack

Only download KubeDNS or CoreDNS if selected

Move dns cleanup to its own file and import tasks based on dns mode

Fix install of KubeDNS when dnsmask_kubedns mode is selected

Add new dns option coredns_dual for dual stack deployment. Added variable to configure replicas deployed. Updated docs for dual stack deployment. Removed rotate option in resolv.conf.

Run DNS manifests for CoreDNS and KubeDNS

Set skydns servers on dual stack deployment

Use only one template for CoreDNS dual deployment

Set correct cluster ip for the dns server
2018-03-16 21:51:37 +01:00
Dmitry Vlasov
977e7ae105 remove obsolete init image, bump dashboard version 1.8.1 -> 1.8.3 2018-02-28 12:52:59 +03:00
Antoine Legrand
37cfd289d8
Merge pull request #2248 from hswong3i/dashboard.yml.j2
Dashboard template should not suffix with .yml.j2
2018-02-06 11:25:02 +01:00
Wong Hoi Sing Edison
4ad53339f6 KubeDNS template should not suffix with .yml.j2 2018-02-05 16:26:54 +08:00
Wong Hoi Sing Edison
a4d3da6a8e Dashboard template should not suffix with .yml.j2 2018-02-05 16:18:21 +08:00
RongZhang
3846384d56 Bump kube-dns to 1.14.8 (#2204)
Bump kube-dns to 1.14.8
2018-01-30 19:23:37 +03:00
Matthew Mosesohn
dc6a17e092
Use include/import tasks (#2192)
import_tasks will consume far less memory, so it should be
used whenever it is compatible.
2018-01-29 14:37:48 +03:00
Brad Beam
08fe61e058
Merge pull request #2071 from riverzhang/dashboard
Update dashboard version to v1.8.1
2018-01-24 20:10:05 -06:00
rong.zhang
df21fc8643 Remove initContainer 2018-01-10 12:17:17 +08:00
rong.zhang
6ed2a60978 fix run dashboard error 2018-01-04 13:13:36 +08:00
rong.zhang
5aef52e8c0 fix dashboard certs secret 2017-12-22 11:17:05 +08:00
rong.zhang
b974b144a8 Add RBAC to binding Dahsboard UI 2017-12-18 23:07:19 +08:00
rong.zhang
0771cd8599 Remove dashboard_tls_key and dashboard_tls_cert 2017-12-13 15:42:20 +08:00
rong.zhang
40edf8c6f5 Update dashboard version to v1.8.0
Update dependencies to be compatible with Kubernetes v1.8
2017-12-13 12:50:44 +08:00
Chad Swenson
b8788421d5 Support for disabling apiserver insecure port
This allows `kube_apiserver_insecure_port` to be set to 0 (disabled).

Rework of #1937 with kubeadm support

Also, fixed an issue in `kubeadm-migrate-certs` where the old apiserver cert was copied as the kubeadm key
2017-12-05 09:13:45 -06:00
Chad Swenson
a89ee8c406 Add ability to use custom cert secret instead of init container provisioned self-signed certs 2017-11-15 10:05:52 -06:00
Chad Swenson
0c6f172e75 Kubernetes Dashboard v1.7.1 Refactor
This version required changing the previous access model for dashboard completely but it's a change for the better. Docs were updated.

* New login/auth options that use apiserver auth proxying by default
* Requires RBAC in `authorization_modes`
* Only serves over https
* No longer available at https://first_master:6443/ui until apiserver is updated with the https proxy URL:
* Can access from https://first_master:6443/api/v1/namespaces/kube-system/services/https:kubernetes-dashboard:/proxy/#!/login you will be prompted for credentials
* Or you can run 'kubectl proxy' from your local machine to access dashboard in your browser from: http://localhost:8001/api/v1/namespaces/kube-system/services/https:kubernetes-dashboard:/proxy/
* It is recommended to access dashboard from behind a gateway that enforces an authentication token, details and other access options here: https://github.com/kubernetes/dashboard/wiki/Accessing-Dashboard---1.7.X-and-above
2017-11-15 10:05:48 -06:00
Matthew Mosesohn
f9b68a5d17
Revert "Support for disabling apiserver insecure port" (#1974) 2017-11-14 13:41:28 +00:00
Chad Swenson
0c7e1889e4 Support for disabling apiserver insecure port
This allows `kube_apiserver_insecure_port` to be set to 0 (disabled). It's working, but so far I have had to:

1. Make the `uri` module "Wait for apiserver up" checks use `kube_apiserver_port` (HTTPS)
2. Add apiserver client cert/key to the "Wait for apiserver up" checks
3. Update apiserver liveness probe to use HTTPS ports
4. Set `kube_api_anonymous_auth` to true to allow liveness probe to hit apiserver's /healthz over HTTPS (livenessProbes can't use client cert/key unfortunately)
5. RBAC has to be enabled. Anonymous requests are in the `system:unauthenticated` group which is granted access to /healthz by one of RBAC's default ClusterRoleBindings. An equivalent ABAC rule could allow this as well.

Changes 1 and 2 should work for everyone, but 3, 4, and 5 require new coupling of currently independent configuration settings. So I also added a new settings check.

Options:

1. The problem goes away if you have both anonymous-auth and RBAC enabled. This is how kubeadm does it. This may be the best way to go since RBAC is already on by default but anonymous auth is not.
2. Include conditional templates to set a different liveness probe for possible combinations of `kube_apiserver_insecure_port = 0`, RBAC, and `kube_api_anonymous_auth` (won't be possible to cover every case without a guaranteed authorizer for the secure port)
3. Use basic auth headers for the liveness probe (I really don't like this, it adds a new dependency on basic auth which I'd also like to leave independently configurable, and it requires encoded passwords in the apiserver manifest)

Option 1 seems like the clear winner to me, but is there a reason we wouldn't want anonymous-auth on by default? The apiserver binary defaults anonymous-auth to true, but kubespray's default was false.
2017-11-06 14:01:10 -06:00
Andrew Greenwood
c383c7e2c1
Update kubedns image to latest 2017-10-29 21:58:05 -04:00
Matthew Mosesohn
ec53b8b66a Move cluster roles and system namespace to new role
This should be done after kubeconfig is set for admin and
before network plugins are up.
2017-10-26 14:36:05 +01:00