Commit graph

286 commits

Author SHA1 Message Date
Matthew Mosesohn
95b460ae94 Remove etcd-proxy from all nodes and use etcd multiaccess 2016-11-09 13:31:12 +03:00
Aleksandr Didenko
60a217766f Add ConfigMap for basic configuration options
Container settings moved from deamonset yaml to a separate
configmap.
2016-11-08 12:57:34 +01:00
Aleksandr Didenko
309240cd6f Adding support for canal network plugin
This patch provides support for Canal network plugin installation
as a self-hosted app, see the following link for details:

https://github.com/tigera/canal/tree/master/k8s-install
2016-11-08 11:04:01 +01:00
Spencer Smith
8f20d90f88 update admission controllers for > 1.4 2016-11-04 12:54:35 -04:00
Smaine Kahlouch
d6f206b5fd Merge pull request #561 from kubespray/rsync_certs
Use tar+register instead of copy/slurp for distributing tokens and certs
2016-10-27 10:52:41 +02:00
Matthew Mosesohn
c7b00caeaa Use tar+register instead of copy/slurp for distributing tokens and certs
Related bug: https://github.com/ansible/ansible/issues/15405

Uses tar and register because synchronize module cannot sudo on the
remote side correctly and copy is too slow.

This patch dramatically cuts down the number of tasks to process
for cert synchronization.
2016-10-26 15:46:18 +03:00
Bogdan Dobrelya
c59c3a1bcf Fix idempotency/recurrence of download and preinstall
* Don't push containers if not changed
* Do preinstall role only once and redistribute defaults to
  corresponding roles

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-10-24 18:28:53 +02:00
Matthew Mosesohn
11f1f71b3b dynamically calculate etcd peer names 2016-10-21 16:17:50 +03:00
Matthew Mosesohn
0e9d1e09e3 Sync master tokens only with those in play_hosts 2016-10-21 14:43:41 +03:00
Matthew Mosesohn
65d2a3b0e5 Use only native cachable hostvars for etcd set_facts 2016-10-21 14:39:58 +03:00
Chad Swenson
a5137affeb Hostname alias fixes
Change the kubelet --hostname-override flag to use the ansible_hostname variable which should be more consistent with the value required by cloud providers

Add ansible_hostname alias to /etc/hosts when it is different from inventory_hostname to overcome node name limitations see https://github.com/kubernetes/kubernetes/issues/22770

Signed-off-by: Chad Swenson <chadswen@gmail.com>
2016-10-18 16:22:32 -05:00
Chad Swenson
c402feffbd Parameterize several dependency endpoints so that they can be overridden with internal mirrors.
Signed-off-by: Chad Swenson <chadswen@gmail.com>
2016-10-15 12:26:52 -05:00
Matthew Mosesohn
71347322d6 Add cluster-cidr to kube-proxy config
This option enables masquerading for traffic directed at pods
that comes frmom outside the cluster.
2016-10-12 19:13:33 +03:00
Artem Roma
3919d666c1 Add possibility to enable network policy via Calico network controller
The requirements for network policy feature are described here [1]. In
order to enable it, appropriate configuration must be provided to the CNI
plug in and Calico policy controller must be set up. Beside that
corresponding extensions needed to be enabled in k8s API.

Now to turn on the feature user can define `enable_network_policy`
customization variable for Ansible.

[1] http://kubernetes.io/docs/user-guide/networkpolicies/
2016-10-10 17:22:12 +03:00
Sergey Vasilenko
dea4210da1 Bump Calico-CNI plugin binaries versions
and correct checksums
2016-10-07 13:14:46 +03:00
Sergey Vasilenko
a6344f7561 Changes in Kubernetes and Calico-CNI plugin config files
required for usage of Calico CNI plugin version 1.4.2
2016-10-06 19:33:16 +03:00
Smaine Kahlouch
c490e5c8a1 Merge pull request #528 from kubespray/proxy-nginx
Use nginx proxy on non-master nodes to proxy apiserver traffic
2016-10-05 19:19:32 +02:00
Matthew Mosesohn
84052ff0b6 use nginx proxy on non-master nodes to proxy apiserver traffic
Also adds all masters by hostname and localhost/127.0.0.1 to
apiserver SSL certificate.

Includes documentation update on how localhost loadbalancer works.
2016-10-05 20:09:10 +03:00
Matthew Mosesohn
f4e6fdc193 Enable quorum read for apiserver
This reduces the likelihood of apiserver status updates
timing out due to etcd write conflicts.
2016-10-04 18:31:42 +03:00
Aleksandr Didenko
fb0ee9d84a Add support for --masquerade-all in kube-proxy
New boolean var `kube_proxy_masquerade_all` which enables/disables
`--masquerade-all` argument for kube-proxy.

Closes #524
2016-10-03 12:24:43 +02:00
Matthew Mosesohn
d9641771ed add kube-masters to SSL certificate 2016-09-29 15:12:30 +03:00
Bogdan Dobrelya
5fd43b7cf0 Allow subdomains of dns_domain and fix kubelet restarts
* Add a var for ndots (default 5) and put it hosts' /etc/resolv.conf.
* Poke kube dns container image to v1.7
* In order to apply changes to kubelet, notify it to
be restarted on changes made to /etc/resolv.conf. Ignore errors as the kubelet
may yet to be present up to the moment of the notification being processed.
* Remove unnecessary kubelet restart for master role as the node role ensures
it is up and running. Notify master static pods waiters for apiserver,
scheduler, controller-manager instead.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-09-27 14:32:49 +02:00
Bogdan Dobrelya
82ee60fe8b Make dnsmasq daemon set optional
Change additional dnsmasq opts:
- Adjust caching size and TTL
- Disable resolve conf to not create loops
- Change dnsPolicy to default (similarly to kubedns's dnsmasq). The
  ClusterFirst should not be used to not create loops
- Disable negative NXDOMAIN replies to be cached
- Make its very installation as optional step (enabled by default).
  If you don't want more than 3 DNS servers, including 1 for K8s, disable
  it.
- Add docs and a drawing to clarify DNS setup.
- Fix stdout logs for dnsmasq/kubedns app configs
- Add missed notifies to resolvconf -u handler
- Fix idempotency of resolvconf head file changes

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-09-23 12:59:06 +02:00
Smaine Kahlouch
0643ed968f Merge pull request #494 from kubespray/etcd_proxy_fix
always bind etcd_proxy to localhost
2016-09-19 14:19:55 +02:00
Smaine Kahlouch
1572aaf6ca Merge pull request #489 from lukaszo/patch-1
Add socat do required pkgs
2016-09-19 12:19:46 +02:00
Smaine Kahlouch
5803de1ac5 Merge pull request #486 from kubespray/etchosts
switch /etc/hosts to use blockinfile
2016-09-19 12:19:37 +02:00
Matthew Mosesohn
341ea5a6ea always bind etcd_proxy to localhost 2016-09-18 19:58:15 +04:00
Bogdan Dobrelya
390764c2b4 Add retry_stagger var for failed download/pushes.
* Add the retry_stagger var to tweak push and retry time strategies.
* Add large deployments related docs.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-09-15 16:43:58 +02:00
Bogdan Dobrelya
422428908a Download containers and save all
Move version/repo vars to download role.
Add container to download params, which overrides url/source_url,
if enabled.
Fix networking plugins download depending on kube_network_plugin.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-09-15 16:43:56 +02:00
Łukasz Oleś
0db441b28f Add socat do required pkgs
It's required for port forwarding.
2016-09-14 21:27:33 +02:00
Matthew Mosesohn
e3ebabc3b0 switch /etc/hosts to use blockinfile 2016-09-14 19:43:33 +03:00
Bogdan Dobrelya
783871a253 Add retries for packages installation
Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-09-13 18:12:07 +02:00
Bogdan Dobrelya
6fdcaa1a63 Add retries for copying binaries from containers
Closes issue: https://github.com/kubespray/kargo/issues/479

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-09-13 15:09:34 +02:00
Matthew Mosesohn
c50c6672f3 Remove SecurityContextDeny API plugin
This is no longer recommended for use since K8s 1.2:
http://kubernetes.io/docs/admin/admission-controllers/#is-there-a-recommended-set-of-plug-ins-to-use
2016-08-29 14:20:28 +03:00
Spencer Smith
82076f90a3 ensure bin dir for coreos before anything else 2016-08-26 13:24:47 -04:00
Bogdan Dobrelya
8168689caa Refactor roles and hosts
Shorten deployment time with:
- Remove redundand roles if duplicated by a dependency and vice versa
- When a member of k8s-cluster, always install docker as a dependency
  of the etcd role and drop the docker role from cluster.yaml.
- Drop etcd and node role dependencies from master role as they are
  covered by the node role in k8s-cluster group as well. Copy defaults
  for master from node role.
- Decouple master, node, secrets roles handlers and vars to be used w/o
  cross references.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-08-25 13:27:57 +02:00
Smaine Kahlouch
c71b078c8e Merge pull request #437 from kubespray/issues/429
Fix handler triggering for kubelet restart
2016-08-25 11:33:50 +02:00
Bogdan Dobrelya
caa8efbf86 Fix handler triggering for kubelet restart
Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-08-25 09:12:25 +02:00
Spencer Smith
4e76bced53 merge with current master, update typos in doc 2016-08-24 09:56:42 -04:00
Spencer Smith
60f263b629 updated to no longer handle gce as cloud-provider. provided aws setup doc 2016-08-24 09:48:32 -04:00
Smana
346eca5748 Revert "pass cloud provider flag in all cases, not just openstack"
This reverts commit f35e5e864f.
2016-08-24 14:32:54 +02:00
Smaine Kahlouch
1938c96239 Merge pull request #420 from bogdando/collect_info
Adjust collect-info playbook
2016-08-24 10:06:30 +02:00
Spencer Smith
f35e5e864f pass cloud provider flag in all cases, not just openstack 2016-08-23 13:57:32 -04:00
Bogdan Dobrelya
47b4242613 Adjust collect-info playbook
Cleanup collected artifacts,
drop unrelated files/commands.
Always install gitinfos script to binaries for external
use.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-08-23 11:28:27 +02:00
Smaine Kahlouch
92c4428cfd Merge pull request #422 from kubespray/issue-421
remove host ca-certs, as they aren't necessary
2016-08-23 10:17:38 +02:00
Bogdan Dobrelya
f61071312a Fix gen-gitinfos.sh
Fix the error gen-gitinfos.sh: 57: [: foo: unexpected operator

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-08-23 10:15:30 +02:00
Spencer Smith
234608433e remove host ca-certs, as they aren't necessary 2016-08-22 16:09:33 -04:00
Matthew Mosesohn
6f07da9f41 Restart kubelet if launcher changed
Fixes #409
2016-08-18 19:00:05 +03:00
Matthew Mosesohn
0c953101ff Fix init scripts for etcd. Fixes #383
Fixes Ubuntu 14.04 deployment of etcd.
2016-08-15 14:09:42 +03:00
Matthew Mosesohn
e727bd52f1 Add option to disable ipv6 dns lookup
New variable disable_ipv6_dns in kubernetes/preinstall.
2016-08-08 13:59:20 +03:00
Matthew Mosesohn
e38258381f Wait for static pods when setting up
Fixes #390
2016-08-02 17:56:31 +03:00
Matthew Mosesohn
e8a1c7a53f Move docker systemd unit creation to docker role
Creating the unit using default settings early on
and then changing it during network_plugin section
leads to too many docker restarts and duplicated code.

Reversed Wants= dependence on docker.service so it does not
restart docker when reloading systemd

Consolidated all docker restart handlers.
2016-08-02 17:56:24 +03:00
Bogdan Dobrelya
2af71f31b4 Rework systemd service units
* Add for docker system units:
    ExecReload=/bin/kill -s HUP $MAINPID
    Delegate=yes
    KillMode=process.
* Add missed DOCKER_OPTIONS for calico/weave docker systemd unit.
* Change Requires= to a less strict and non-faily Wants=, add missing
  Wants= for After=.
* Align wants/after in a wat if Wants=foo, After= has foo as well.
* Make wants/after docker.service to ask for the docker.socket as well.
* Move "docker rm -f" commands from ExecStartPre= to ExecStopPost=.
  hooks to ensure non-destructive start attempts issued by Wants=.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-08-02 10:55:42 +02:00
Matthew Mosesohn
5668e5f767 Fix etcd restart and handler systemd tasks
Changed Wants=docker.service to docker.socket

Renamed handlers for reloading systemd to contain role in task name.
2016-07-29 16:32:35 +03:00
Antoine Legrand
9fb391fed5 Merge pull request #381 from kubespray/fixetcdstandalone
Fix etcd standalone deployment
2016-07-26 16:04:26 -07:00
Antoine Legrand
fbc55da2bf Merge pull request #378 from bogdando/issues/26
Add HA/LB endpoints for kube-apiserver
2016-07-26 16:03:31 -07:00
Matthew Mosesohn
1b1f5f22d4 Fix etcd standalone deployment
etcd facts are generated in kubernetes/preinstall, so etcd nodes need
to be evaluated first before the rest of the deployment.

Moved several directory facts from kubernetes/node to
kubernetes/preinstall because they are not backward dependent.
2016-07-26 18:15:06 +03:00
Bogdan Dobrelya
731d32afda Add HA/LB endpoints for kube-apiserver
* Add HA docs for API server.
* Add auto-evaluated internal endpoints and clarify the loadbalancer_apiserver
vars and usecases.
* Use facts for kube_apiserver to not repeat code and enable LB endpoints use.
* Use /healthz check for the wait-for apiserver.
* Use the single endpoint for kubelet instead of the list of apiservers
* Specify kube_apiserver_count to for HA layout

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-07-25 17:25:45 +02:00
Matthew Mosesohn
b4688701ea Copy kubectl from docker container
Nearly the last stage of source all components to containers.
Kubectl will be called from hyperkube image.

Remaining tasks:
 * Move kube_version variable to kubernetes/preinstall
 * Drop placeholder download.nothing requirement
2016-07-25 18:17:59 +03:00
Matthew Mosesohn
d0a1e15ef3 Deploy kubelet and kube-apiserver as containers
kubelet via docker
kube-apiserver as a static pod

Fixed etcd service start to be more tolerant of slow start.

Workaround for kube_version to stay in download role, but not
download an files by creating a new "nothing" download entry.
2016-07-22 16:42:34 +03:00
Matthew Mosesohn
7f212ca9cb Revert "Add HA/LB endpoints for kube-apiserver"
This reverts commit a70c3b661e.
2016-07-22 13:54:38 +03:00
Bogdan Dobrelya
a70c3b661e Add HA/LB endpoints for kube-apiserver
* Add auto-evaluated internal endpoints and clarify the loadbalancer_apiserver
vars and usecases.
* Add loadbalancer_apiserver_localhost (default false). If enabled, override
the external LB and expect localhost:443/8080 to be new internal only frontends.
* Add kube_apiserver_multiaccess to ignore loadbalancers, and make clients
to access the apiservers as a comma-separated list of access_ip/ip/ansible ip
(a default mode). When disabled, allow clients to use the given loadbalancers.
* Define connections security mode for kube controllers, schedulers, proxies.
It is insecure be default, which is the current deployment choice.
* Rework the groups['kube-master'][0] hardcode defining the apiserver
endpoints.
* Improve grouping of vars and add facts for kube_apiserver.
* Define kube_apiserver_insecure_bind_address as a fact, add more
facts for ease of use.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-07-21 11:05:03 +02:00
Antoine Legrand
277c5d74cc Merge pull request #367 from bogdando/set_facts
Fix set_facts visibility
2016-07-20 18:00:15 +02:00
Bogdan Dobrelya
a76e5dbb11 Fix set_facts visibility
Move set_facts to the preinstall scope, so every role
may see it. For example, network plugins to see the etcd_endpoint.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-07-20 11:41:09 +02:00
Matthew Mosesohn
c3e5aac18e Add variable kube_resolv_conf
Allow configuration of a custom /etc/resolv.conf for kubelet.
2016-07-20 11:57:47 +03:00
Bogdan Dobrelya
32cd6e99b2 Add etcd proxy support
* Enforce a etcd-proxy role to a k8s-cluster group members. This
provides an HA layout for all of the k8s cluster internal clients.
* Proxies to be run on each node in the group as a separate etcd
instances with a readwrite proxy mode and listen the given endpoint,
which is either the access_ip:2379 or the localhost:2379.
* A notion for the 'kube_etcd_multiaccess' is: ignore endpoints and
loadbalancers and use the etcd members IPs as a comma-separated
list. Otherwise, clients shall use the local endpoint provided by a
etcd-proxy instances on each etcd node. A Netwroking plugins always
use that access mode.
* Fix apiserver's etcd servers args to use the etcd_access_endpoint.
* Fix networking plugins flannel/calico to use the etcd_endpoint.
* Fix name env var for non masters to be set as well.
* Fix etcd_client_url was not used anywhere and other etcd_* facts
evaluation was duplicated in a few places.
* Define proxy modes only in the env file, if not a master. Del
an automatic proxy mode decisions for etcd nodes in init/unit scripts.
* Use Wants= instead of Requires= as "This is the recommended way to
hook start-up of one unit to the start-up of another unit"
* Make apiserver/calico Wants= etcd-proxy to keep it always up

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
Co-authored-by: Matthew Mosesohn <mmosesohn@mirantis.com>
2016-07-19 14:09:40 +02:00
Smana
dfe7bfd127 use hyperkube coreos image 2016-07-14 21:20:41 +02:00
Smana
a709cd9aa1 use iptables as default proxy mode 2016-07-12 10:20:43 +02:00
Smaine Kahlouch
a5c21ab2e8 Merge pull request #346 from bogdando/issues/345
Add hostpath dynamic provisioner for PetSets
2016-07-09 22:43:09 +02:00
Spencer Smith
c9cff5c845 updated admission controllers for >1.2 Kubernetes 2016-07-08 10:04:14 -07:00
Bogdan Dobrelya
da20d9eda4 Add hostpath dynamic provisioner for PetSets
Defaults to false. Use with v1.3 only.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-07-08 16:52:39 +02:00
Daniel Leining
1d148e9755 fix kubelet ip with quotes 2016-07-05 17:23:08 -04:00
Daniel Leining
d84ed1b4b3 make kubelet use "ip" as bind address 2016-07-05 17:14:12 -04:00
Matthew Mosesohn
d2151500b6 Fix kube-apiserver log level syntax 2016-07-05 13:11:45 +03:00
Matthew Mosesohn
b847a43c61 Set hyperkube version to kube_version 2016-07-05 12:43:27 +03:00
Smana
f1ba247844 upgrade to k8s v1.3.0 2016-07-03 14:14:09 +02:00
Daniel Leining
72ab34f210 Add --bind-address to kube-apiserver 2016-07-01 18:33:59 -04:00
Smana
85fa3efc06 upgrade kubernetes to v1.2.5 2016-06-29 15:38:33 +02:00
Smana
247a1a6e6e change hyperkube repository 2016-06-29 14:07:05 +02:00
mattymo
708d2fbd61 Add KUBE_API_INSECURE_BIND to systemd unit file
This was missing from commit c4c312c2e6
2016-06-27 13:01:22 +04:00
Matthew Mosesohn
c4c312c2e6 Add configurable option for kube_apiserver_insecure_bind_address 2016-06-24 18:10:01 +03:00
Smana
922c6897d1 Install python-pip on first master 2016-06-12 20:44:12 +02:00
Smana
8281b98e19 install kpm in order to deploy addons 2016-06-11 21:08:05 +02:00
mattymo
68808534b3 Fix order in restart kubelet to fix systemd reload
Systemd reload before reload kubelet was failing because its definition was before "restart kubelet". Its definition should be after the notify hook.
2016-05-31 20:09:49 +04:00
Spencer Smith
a740e521d2 removed os restriction for coreos 2016-05-24 12:03:16 -04:00
Smaine Kahlouch
bdc183114a Merge pull request #261 from paulczar/meta_roles_yo
turn adduser/download roles into meta roles
2016-05-23 17:29:37 +02:00
Paul Czarkowski
7de87d958e turn adduser/download roles into meta roles
This should make things a little more composable,
by making these roles meta roles that perform no
actions by default we allow each role to own its own
resources.
2016-05-22 17:25:52 -05:00
Paul Czarkowski
c226b4e5cb fixes issue #258
Kubernetes API server has an option:

```
--advertise-address=<nil>: The IP address on which to advertise the apiserver to members of the cluster. This address must be reachable by the rest of the cluster. If blank, the --bind-address will be used. If --bind-address is unspecified, the host's default interface will be used.
```

kargo does not set --bind-address, thus it binds to eth0, in vagrant and similar
environments this causes issues because nodes cannot talk to eachother over eth0.

This sets `--advertise-address` to `ip` if its set, otherwise the default behavior
of is persisted by using `ansible_default_ipv4.address`.
2016-05-22 13:48:16 -05:00
Paul Czarkowski
d8bebcd201 Fix issue with check_certs playbook
check_certs task "Check_certs | Set 'sync_certs' to true" was failing
due to the dict not existing, this sets defaults that allows the
correct behavior of the conditionals.
2016-05-15 17:15:59 -05:00
Smana
ae5ff890d4 fix flannel deployment, remove docker bridge before restarting 2016-05-13 18:10:00 +02:00
Smana
608e7dfab2 upgrade k8s vers, and add a script for future upgrades 2016-05-12 15:56:30 +02:00
Smana
1884d89d3b fixes the certs issue when masters or not in the kube-node group 2016-05-12 10:07:34 +02:00
Smaine Kahlouch
ed95f9ab81 Merge pull request #232 from rsmitty/issue-231
Issue 231: ensure ca.pem makes it to multi-masters
2016-05-11 21:24:04 +02:00
Spencer Smith
9f8466a186 ensure ALL certs are synced between masters 2016-05-11 10:09:13 -07:00
Spencer Smith
743ad0eb5c s/sync_certs/sync_tokens 2016-05-11 09:38:26 -07:00
Spencer Smith
5253b3ec13 ensure ca.pem makes it to multi-masters 2016-05-11 09:06:08 -07:00
Smaine Kahlouch
96c63cc0b6 Merge pull request #227 from paulczar/vagrant
Add native Vagrant support
2016-05-09 11:47:07 +02:00
Paul Czarkowski
5f2fa6d76f revert .gitignore for secrets 2016-05-08 23:46:35 -05:00
Paul Czarkowski
8f4e879ca7 Add native Vagrant support
This allows you to simply run `vagrant up` to get a 3 node HA cluster.

* Creates a dynamic inventory and uses the inventory/group_vars/all.yml
* commented lines in inventory.example so that ansible doesn't try to use it.
* added requirements.txt to give easy way to install ansible/ipaddr
* added gitignore files to stop attempts to save unwated files
* changed `Check if kube-system exists` to `failed_when: false` instead of
`ignore_errors`
2016-05-08 10:17:11 -05:00
Smana
4f627baf71 generate secrets on first master 2016-05-07 21:08:29 +02:00