Commit Graph

2251 Commits (400ee2aa57e57235e8abb55f4e4854dd096238ff)
 

Author SHA1 Message Date
Matthew Mosesohn 975accbe1d just use public_ip in creating gce temporary waitfor hosts (#1646)
* just use public_ip in creating gce temporary waitfor hosts

* Update create-gce.yml
2017-09-18 13:24:57 +01:00
Brad Beam aaa27d0a34 Adding quotes around parameters in cloud_config (#1664)
This is to help support escapes and special characters
2017-09-16 08:43:47 +01:00
Kevin Lefevre 9302ce0036 Enhanced OpenStack cloud provider (#1627)
- Enable Cinder API version for block storage
- Enable floating IP for LBaaS
2017-09-16 08:43:24 +01:00
Matthew Mosesohn 0aab3c97a0 Add all-in-one CI mode and make coreos test aio (#1665) 2017-09-15 22:28:37 +01:00
Matthew Mosesohn 8e731337ba Enable HA deploy of kubeadm (#1658)
* Enable HA deploy of kubeadm

* raise delay to 60s for starting gce hosts
2017-09-15 22:28:15 +01:00
Matthew Mosesohn b294db5aed fix apply for netchecker upgrade (#1659)
* fix apply for netchecker upgrade and graceful upgrade

* Speed up daemonset upgrades. Make check wait for ds upgrades.
2017-09-15 13:19:37 +01:00
Matthew Mosesohn 8d766a2ca9 Enable ssh opts by in config, set 100 connection retries (#1662)
Also update to ansible 2.3.2
2017-09-15 10:19:36 +01:00
Brad Beam f2ae16e71d Merge pull request #1651 from bradbeam/vaultnocontent
Fixing condition where vault CA already exists
2017-09-14 17:04:15 -05:00
Brad Beam ac281476c8 Prune unnecessary certs from vault setup (#1652)
* Cleaning up cert checks for vault

* Removing all unnecessary etcd certs from each node

* Removing all unnecessary kube certs from each node
2017-09-14 12:28:11 +01:00
neith00 1b1c8d31a9 upgrading from weave version 2.0.1 to 2.0.4
This upgrade has been testing offline on a 1.7.5 cluster
2017-09-14 10:29:28 +02:00
Brad Beam 4b587aaf99 Adding ability to specify altnames for vault cert (#1640) 2017-09-14 07:19:44 +01:00
Kyle Bai 016301508e Update to Kubernetes v1.7.5 (#1649) 2017-09-14 07:18:03 +01:00
Matthew Mosesohn 6744726089 kubeadm support (#1631)
* kubeadm support

* move k8s master to a subtask
* disable k8s secrets when using kubeadm
* fix etcd cert serial var
* move simple auth users to master role
* make a kubeadm-specific env file for kubelet
* add non-ha CI job

* change ci boolean vars to json format

* fixup

* Update create-gce.yml

* Update create-gce.yml

* Update create-gce.yml
2017-09-13 19:00:51 +01:00
Brad Beam 0a89f88b89 Fixing condition where CA already exists 2017-09-13 03:40:46 +00:00
Brad Beam 69fac8ea58 Merge pull request #1634 from bradbeam/calico_cni
fix for calico cni plugin node name
2017-09-11 22:18:06 -05:00
Brad Beam a51104e844 Merge pull request #1648 from kubernetes-incubator/mattymo-patch-1
Update getting-started.md
2017-09-11 17:55:51 -05:00
Matthew Mosesohn 943aaf84e5 Update getting-started.md 2017-09-11 12:47:04 +03:00
Seungkyu Ahn e8bde03a50 Setting kubectl bin directory (#1635) 2017-09-09 23:54:13 +03:00
Matthew Mosesohn 75b13caf0b Fix kube-apiserver status checks when changing insecure bind addr (#1633) 2017-09-09 23:41:48 +03:00
Matthew Mosesohn 0f231f0e76 Improve method to create and wait for gce instances (#1645) 2017-09-09 23:41:31 +03:00
Matthew Mosesohn 5d99fa0940 Purge old upgrade hooks and unused tasks (#1641) 2017-09-09 23:41:20 +03:00
Matthew Mosesohn 649388188b Fix netchecker update side effect (#1644)
* Fix netchecker update side effect

kubectl apply should only be used on resources created
with kubectl apply. To workaround this, we should apply
the old manifest before upgrading it.

* Update 030_check-network.yml
2017-09-09 23:38:38 +03:00
Matthew Mosesohn 9fa1873a65 Add kube dashboard, enabled by default (#1643)
* Add kube dashboard, enabled by default

Also add rbac role for kube user

* Update main.yml
2017-09-09 23:38:03 +03:00
Matthew Mosesohn f2057dd43d Refactor downloads (#1642)
* Refactor downloads

Add prefixes to tasks (file vs container)
Remove some delegates
Clean up some conditions

* Update ansible.cfg
2017-09-09 23:32:12 +03:00
Brad Beam eeffbbb43c Updating calicocni.hostname to calicocni.nodename 2017-09-08 12:47:40 +00:00
Brad Beam aaa0105f75 Flexing calicocni.hostname based on cloud provider 2017-09-08 12:47:40 +00:00
Matthew Mosesohn f29a42721f Clean up debug in check apiserver test (#1638)
* Clean up debug in check apiserver test

* Change password generation for kube_user

Special characters are not allowed in known_users.csv file
2017-09-08 15:47:13 +03:00
Matthew Mosesohn 079d317ade Default is_atomic to false (#1637) 2017-09-08 15:00:57 +03:00
Matthew Mosesohn 6f1fd12265 Revert "Add option for fact cache expiry" (#1636)
* Revert "Add option for fact cache expiry (#1602)"

This reverts commit fb30f65951.
2017-09-08 10:19:58 +03:00
Maxim Krasilnikov e16b57aa05 Store vault users passwords to credentials dir. Create vault and etcd roles after start vault cluster (#1632) 2017-09-07 23:30:16 +03:00
Yorgos Saslis fb30f65951 Add option for fact cache expiry (#1602)
* Add option for fact cache expiry 

By adding the `fact_caching_timeout` we avoid having really stale/invalid data ending up in there. 
Leaving commented out by default, for backwards compatibility, but nice to have there.

* Enabled cache-expiry by default

Set to 2 hours and modified comment to reflect change
2017-09-07 23:29:27 +03:00
Tennis Smith a47aaae078 Add bastion host definitions (#1621)
* Add comment line and documentation for bastion host usage

* Take out unneeded sudo parm

* Remove blank lines

* revert changes

* take out disabling of strict host checking
2017-09-07 23:26:52 +03:00
Matthew Mosesohn 7117614ee5 Use a generated password for kube user (#1624)
Removed unnecessary root user
2017-09-06 20:20:25 +03:00
Chad Swenson e26aec96b0 Consolidate kube-proxy module and sysctl loading (#1586)
This sets br_netfilter and net.bridge.bridge-nf-call-iptables sysctl from a single play before kube-proxy is first ran instead of from the flannel and weave network_plugin roles after kube-proxy is started
2017-09-06 15:11:51 +03:00
Sam Powers c60d104056 Update checksums (etcd calico calico-cni weave) to fix uploads.yml (#1584)
the uploads.yml playbook was broken with checksum mismatch errors in
various kubespray commits, for example, 3bfad5ca73
which updated the version from 3.0.6 to 3.0.17 without updating the
corresponding checksums.
2017-09-06 15:11:13 +03:00
Oliver Moser e6ff8c92a0 Using 'hostnamectl' to set unconfigured hostname on CoreOS (#1600) 2017-09-06 15:10:52 +03:00
Maxim Krasilnikov 9bce364b3c Update auth enabled methods in group_vars example (#1625) 2017-09-06 15:10:18 +03:00
Chad Swenson cbaa2b5773 Retry Remove all Docker containers in reset (#1623)
Due to various occasional docker bugs, removing a container will sometimes fail. This can often be mitigated by trying again.
2017-09-06 14:23:16 +03:00
Matthieu 0453ed8235 Fix an error with Canal when RBAC are disabled (#1619)
* Fix an error with Canal when RBAC are disabled

* Update using same rbac strategy used elsewhere
2017-09-06 11:32:32 +03:00
Brad Beam a341adb7f3 Updating CN for node certs generated by vault (#1622)
This allows the node authorization plugin to function correctly
2017-09-06 10:55:08 +03:00
Matthew Mosesohn 4c88ac69f2 Use kubectl apply instead of create/replace (#1610)
Disable checks for existing resources to speed up execution.
2017-09-06 09:36:54 +03:00
Brad Beam 85c237bc1d Merge pull request #1607 from chapsuk/vault_roles
Vault role updates
2017-09-05 11:48:41 -05:00
Tennis Smith 35d48cc88c Point apiserver address to 0.0.0.0 (#1617)
* Point apiserver address to 0.0.0.0
Added loadbalancer api server address
* Update documentation
2017-09-05 18:41:47 +03:00
mkrasilnikov 957b7115fe Remove node name from kube-proxy and admin certificates 2017-09-05 14:40:26 +03:00
Yorgos Saslis 82eedbd622 Update ansible inventory file when template changes (#1612)
This trigger ensures the inventory file is kept up-to-date. Otherwise, if the file exists and you've made changes to your terraform-managed infra without having deleted the file, it would never get updated. 

For example, consider the case where you've destroyed and re-applied the terraform resources, none of the IPs would get updated, so ansible would be trying to connect to the old ones.
2017-09-05 14:10:53 +03:00
mkrasilnikov b930b0ef5a Place vault role credentials only to vault group hosts 2017-09-05 11:16:18 +03:00
mkrasilnikov ad313c9d49 typo fix 2017-09-05 09:07:36 +03:00
mkrasilnikov 06035c0f4e Change vault CI CLOUD_MACHINE_TYPE to n1-standard-2 2017-09-05 09:07:36 +03:00
mkrasilnikov e1384f6618 Using issue cert result var instead hostvars 2017-09-05 09:07:36 +03:00
mkrasilnikov 3acb86805b Rename vault_address to vault_bind_address 2017-09-05 09:07:35 +03:00