Commit graph

1378 commits

Author SHA1 Message Date
Matthew Mosesohn
17f3d76445 Update inventory.py 2016-12-29 10:57:58 +03:00
Bogdan Dobrelya
6e1c0cdd15 Systemd units, limits, and bin path fixes
* Add restart for weave service unit
* Reuse docker_bin_dir everythere
* Limit systemd managed docker containers by CPU/RAM. Do not configure native
  systemd limits due to the lack of consensus in the kernel community
  requires out-of-tree kernel patches.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-12-28 15:49:42 +01:00
Matthew Mosesohn
3b4d9ceeca Add yaml/json loader for inventory 2016-12-28 16:50:31 +03:00
Bogdan Dobrelya
4413c38105 Merge pull request #831 from mattymo/fix_separate_etcd
Fix creation and sync of etcd certs
2016-12-28 13:56:42 +01:00
Matthew Mosesohn
2ac2a3ed93 Fix creation and sync of etcd certs
Admin certs only go to etcd nodes
Only generate cert-data for nodes that need sync
2016-12-28 14:21:17 +04:00
Matthew Mosesohn
612c5bb5f1 Merge pull request #818 from mattymo/calico-rr-certs
Fix calico-rr to use etcd certs instead of kube certs
2016-12-28 08:47:16 +03:00
Bogdan Dobrelya
44d89322d7 Rework wildcards matching all nodes
* Re-enable ansible_ssh_pipelining as expected for the cluster.yml
* Do not use 'all' wildcasts for hosts, limit only to k8s-cluster, etcd,
  calico-rr groups instead. Other nodes in inventory are out of Kargo
  scope and it's up to users how to manage them.

Signed-off-by: Bogdan Dobrelya <bogdando@mail.ru>
2016-12-27 17:02:08 +01:00
Antoine Legrand
e4562cf84c Merge pull request #828 from bogdando/triggers
Rework CI triggers/pipeline
2016-12-27 15:20:42 +01:00
Matthew Mosesohn
716b590f3b Fix calico-rr to use etcd certs instead of kube certs 2016-12-27 17:04:50 +03:00
Bogdan Dobrelya
c454e15881 Rework CI triggers/pipeline
* Run CI triggers in one step
* Run all test matrix for triggers
* Switch back to g1-small

Signed-off-by: Bogdan Dobrelya <bogdando@mail.ru>
2016-12-27 14:39:52 +01:00
Bogdan Dobrelya
1ccd4b892f Merge pull request #827 from bogdando/noreds
Rework ignore_errors to report no reds
2016-12-27 14:37:38 +01:00
Bogdan Dobrelya
9b29df183b Rework ignore_errors to report no reds
Signed-off-by: Bogdan Dobrelya <bogdando@mail.ru>
2016-12-27 13:00:50 +01:00
Antoine Legrand
f06de012dc Merge pull request #805 from bogdando/dnsmasq_armors
Do not forward private domains for upstream resolvers
2016-12-26 14:50:39 +01:00
Bogdan Dobrelya
222859601e Do not forward bogus domains for upstream resolvers
Also fix kube log level 4 to log dnsmasq queries.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-12-23 11:53:14 +01:00
Matthew Mosesohn
4ea0cbb2d0 Merge pull request #804 from mattymo/inventory_builder
Add inventory builder python script
2016-12-23 13:22:41 +03:00
Matthew Mosesohn
69b89a9155 Add inventory builder python script
Includes tox support for running unit tests.
Small note added to getting-started guide for using
inventory_builder.py

Also adds manual-only unit test.
2016-12-23 13:00:56 +03:00
Thomas Catterall
9905d0056b Update README.md 2016-12-22 22:46:23 +00:00
Spencer Smith
460571506c Merge pull request #813 from mattymo/etcdcertflags
Adjust etcd server certificates
2016-12-22 16:37:15 -05:00
Matthew Mosesohn
a2c38f5f5f Update etcd.j2 2016-12-22 22:29:24 +03:00
Matthew Mosesohn
e5374af95c Adjust etcd server certificates
ETCD doesn't need cert/key options set. It only requires peer
cert options.
2016-12-22 23:05:17 +04:00
Spencer Smith
f3f16e3676 Workaround etcdctl not yet being installed (#797)
workaround case for etcdctl not yet being installed, only allow for return code of 0 (no error)
2016-12-22 12:41:38 -05:00
Antoine Legrand
7c9db9ad73 Merge pull request #812 from mattymo/vars_doc
Document commonly used Kargo vars
2016-12-22 18:01:23 +01:00
Matthew Mosesohn
0a6d2223a5 Document commonly used Kargo vars 2016-12-22 19:57:39 +03:00
Matthew Mosesohn
370ad3acba Merge pull request #760 from genti-t/issue-748-flannel-options
Fix Flannel network on CoreOS
2016-12-22 19:02:31 +03:00
Matthew Mosesohn
a49d061298 Merge pull request #798 from mattymo/perhostssl7
Individual etcd ssl certs
2016-12-22 19:02:15 +03:00
Genti Topija
a42b458fdf Fix Flannel network on CoreOS
Resolves: #748
2016-12-22 16:50:04 +01:00
Antoine Legrand
ef69b99acc Merge pull request #809 from bogdando/ci_forks
Raise ansible forks for CI test config
2016-12-22 15:55:47 +01:00
Bogdan Dobrelya
d823248cd7 Raise ansible forks for CI test config
As we raised the flavor from small to standard, raise the ansible
forks from default 5 to 20 to speed up deployment.

Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-12-22 15:12:47 +01:00
Bogdan Dobrelya
d5b68ae194 Merge pull request #787 from bogdando/coreos_weave_manual
Add coreos-alpha weave manual CI builds
2016-12-22 14:32:03 +01:00
Matthew Mosesohn
5457799aa3 Individual etcd ssl certs
Includes hooks for triggering calico, kubelet, and kube-apiserver restarts
if etcd certs changed.
2016-12-22 13:31:11 +03:00
Bogdan Dobrelya
85f31a369e Merge pull request #786 from mattymo/bug777
Add wait for kube-apiserver to kubernetes-apps
2016-12-22 11:02:50 +01:00
Bogdan Dobrelya
b8590a13c8 Merge pull request #801 from rsmitty/issue-800
create systemd drop-in path if not existent
2016-12-22 11:02:05 +01:00
Bogdan Dobrelya
32f0b9756f Add coreos-alpha weave manual CI builds
Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-12-22 11:00:31 +01:00
Bogdan Dobrelya
4120c77a08 Merge pull request #791 from kubernetes-incubator/premptilble_gce_vm
Use preemptible instances
2016-12-22 11:00:08 +01:00
Alexander Block
272d8e754b Fix wrong path of dhclient on CentOS+Azure
This was alredy fixed in #755 but had to be reverted. This PR should be
more intelligent about deciding which path to use.
2016-12-21 21:51:07 +01:00
Spencer Smith
3575f890fc create systemd drop-in path if not existent 2016-12-21 13:06:12 -05:00
Bogdan Dobrelya
29358c08c6 Merge pull request #794 from kubernetes-incubator/revert-722-dnsmasq_armors
Revert "Do not forward private domains for upstream resolvers"
2016-12-21 17:39:11 +01:00
Bogdan Dobrelya
b103799901 Revert "Do not forward private domains for upstream resolvers" 2016-12-21 15:24:17 +01:00
Matthew Mosesohn
b1eb852207 Add wait for kube-apiserver to kubernetes-apps
Fixes #777
2016-12-21 15:39:39 +03:00
Antoine Legrand
d892adfae1 use standard vm 2016-12-21 11:43:52 +01:00
Antoine Legrand
e7d78fc7d7 Merge pull request #782 from samos123/master
Vagrant, skip synchronize tasks (bug #697)
2016-12-21 11:18:20 +01:00
Bogdan Dobrelya
d89df3a2ce Merge pull request #780 from bogdando/downloads
Add download_always_pull check and sha256 for docker images
2016-12-21 11:02:57 +01:00
Antoine Legrand
bcad456b7e Use preemptible instances 2016-12-21 09:27:21 +01:00
Bogdan Dobrelya
f45872b558 Add download_always_pull check and sha256 for docker images
Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-12-20 17:02:09 +01:00
Bogdan Dobrelya
763c1aff72 Merge pull request #722 from bogdando/dnsmasq_armors
Do not forward private domains for upstream resolvers
2016-12-20 14:25:17 +01:00
Antoine Legrand
acc4c6a61e Merge pull request #785 from bogdando/pipelines
Add gitlab CI auto builds for triggers
2016-12-20 14:18:19 +01:00
Bogdan Dobrelya
072c4e4669 Merge pull request #775 from kubernetes-incubator/register_master
Register master node as unschedulable
2016-12-20 14:17:55 +01:00
Bogdan Dobrelya
5674da25d0 Add gitlab CI auto builds for triggers
Signed-off-by: Bogdan Dobrelya <bdobrelia@mirantis.com>
2016-12-20 11:23:13 +01:00
Sam Stoelinga
0908c964b5 Fix #781 prevent vagrant asking for password
When download_run_once=True it will try to first download images to host
running vagrant, then upload the images from vagrant host to the vagrant
vms. While trying to upload the images it will ask for a the vagrant
linux user password.

For small environments this behaviour is unnecessary hence changing the
default to download_run_once=False inside the Vagrantfile seems like a
good fix.
2016-12-19 11:52:18 -08:00
Bogdan Dobrelya
c147f710ab Merge pull request #774 from kubernetes-incubator/ant31-patch-2
check if calico_peer_rr is defined
2016-12-19 18:19:03 +01:00