Commit graph

977 commits

Author SHA1 Message Date
Matthew Mosesohn
927a95fb65 Restart kubelet when updating /etc/resolv.conf on all k8s nodes 2017-04-04 20:43:47 +03:00
Matthew Mosesohn
380ad9815f Merge pull request #1153 from mattymo/graceful_drain
Move graceful upgrade test to Ubuntu canal HA, adjust drain
2017-04-04 17:33:53 +03:00
Matthew Mosesohn
af20eb08cf Merge pull request #1200 from mattymo/issue1190
Fix multiline condition for k8s check certs
2017-04-04 15:48:05 +03:00
Matthew Mosesohn
90e8d4c4ea Add /var/lib/cni to kubelet
Necessary to persist this directory for host-local IPAM used by Canal
Add pre-upgrade task to copy /var/lib/cni out of old kubelet.
2017-04-03 19:38:24 +03:00
Matthew Mosesohn
7c27fe3fdf Merge pull request #1185 from intelsdi-x/hostname
Use hostname module to set hostname, and do it for all Os not only Co…
2017-04-03 19:01:12 +03:00
Matthew Mosesohn
620944b052 Fix multiline condition for k8s check certs
Fixes #1190
2017-04-03 17:44:55 +03:00
Matthew Mosesohn
bd7b91b07a Merge pull request #1166 from rogerwelin/master
add iptables --flush to reset role
2017-04-03 17:25:10 +03:00
Matthew Mosesohn
0b2141817a Merge pull request #1182 from artem-panchenko/bumpCalicoPolicyControllerVersion
Bump calico policy controller version
2017-04-03 17:21:52 +03:00
Paweł Skrzyński
cee89ec562 Use hostname module to set hostname, and do it for all Os not only CoreOS 2017-04-03 15:09:33 +02:00
Matthew Mosesohn
a5391ff21a use etcd2 when upgrading unless forced 2017-04-03 15:07:42 +03:00
Matthew Mosesohn
c30fcc1a5b Merge pull request #1194 from adidenko/fix-sync_certs
Fix multiline when condition in sync_certs task
2017-03-31 17:39:40 +03:00
Aleksandr Didenko
0762afd390 Fix multiline when condition in sync_certs task
Folded style in multiline 'when' condition causes error with
unexpected ident. Changing it to literal style should fix
the issue.

Closes #1190
2017-03-30 22:21:04 +02:00
Spencer Smith
c2a5453f6f Merge pull request #1170 from jlothian/atomic-docker-network
1169 - fix docker systemd unit
2017-03-30 13:13:28 -07:00
Matthew Mosesohn
423fe7e51d Update .gitlab-ci.yml 2017-03-30 12:19:15 +04:00
Matthew Mosesohn
0e648347f9 fix etcd restart 2017-03-29 23:22:49 +04:00
Matthew Mosesohn
1d4e6b2ade delete master containers forcefully 2017-03-27 19:08:22 +03:00
Matthew Mosesohn
1e1dfe2cab restart scheduler and controller-manager too 2017-03-27 13:51:35 +03:00
Matthew Mosesohn
eab686bd56 ensure post-upgrade purge ones only once 2017-03-27 13:28:37 +03:00
Matthew Mosesohn
f953303722 switch debian8-canal-ha to ubuntu 2017-03-27 13:28:37 +03:00
Matthew Mosesohn
a7bedd30ab move network plugins out of grouped upgrades 2017-03-27 13:28:37 +03:00
Matthew Mosesohn
1a96970918 Fix delegate tasks for kubectl and etcdctl 2017-03-27 13:28:37 +03:00
Matthew Mosesohn
bf3425322f Significantly reduce memory requirements
Canal runs more pods and upgrades need a bit of extra
room to load new pods in and get the old ones out.
2017-03-27 13:28:37 +03:00
Matthew Mosesohn
f1f58166e5 Only cordon Ready nodes 2017-03-27 13:28:37 +03:00
Matthew Mosesohn
68a1faaaa0 Move graceful upgrade test to debian canal HA, adjust drain
Graceful upgrades require 3 nodes
Drain now has a command timeout of 40s
2017-03-27 13:28:37 +03:00
Matthew Mosesohn
986a89be66 Merge pull request #1181 from holser/refactor_etcd
Refactor etcd role
2017-03-27 13:05:35 +03:00
Sergii Golovatiuk
77671dbd05 Refactor etcd role
- Run docker run from script rather than directly from systemd target
- Refactoring styling/templates

Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>
2017-03-24 12:34:15 +01:00
Artem Panchenko
bc0db79c86 Bump calico policy controller version
Latest released version of kube-policy-controller
contains important bug fixes and should be used
by default.
2017-03-24 12:13:09 +02:00
Matthew Mosesohn
503dc26f11 Merge pull request #1177 from rutsky/replace-nbsp
replace non-breakable space with regular space
2017-03-23 12:59:45 +03:00
Matthew Mosesohn
5879d62869 Merge pull request #1179 from kubernetes-incubator/missing_defaults
Add missing defaults
2017-03-23 12:16:13 +03:00
Antoine Legrand
9f1ca9377f Add missing defaults 2017-03-23 10:05:34 +01:00
Vladimir Rutsky
bc30ea7582 replace non-breakable space with regular space
Non-brekable space is 0xc2 0xa0 byte sequence in UTF-8.

To find one:

    $ git grep -I -P '\xc2\xa0'

To replace with regular space:

    $ git grep -l -I -P '\xc2\xa0' | xargs sed -i 's/\xc2\xa0/ /g'

This commit doesn't include changes that will overlap with commit f1c59a91a1.
2017-03-23 00:25:01 +03:00
Matthew Mosesohn
daf88282cc Merge pull request #1172 from mattymo/dnsmasq_upgrade
Use checksum of dnsmasq config to trigger updates of dnsmasq
2017-03-22 18:00:10 +03:00
Matthew Mosesohn
8098b77e41 Merge pull request #1167 from mattymo/dnsmasq_when_deploying_master
Change wait for dnsmasq to skip if there are no kube-nodes in play
2017-03-22 17:59:56 +03:00
Roger Welin
da26c560b2 add iptables --flush to reset role 2017-03-22 11:10:24 +01:00
Matthew Mosesohn
e020457d1d Use checksum of dnsmasq config to trigger updates of dnsmasq
Allows config changes made by Ansible to restart dnsmasq deployment
2017-03-22 13:03:55 +03:00
Josh Lothian
18404db076 1169 - fix docker systemd unit
The docker-network environment file masks the new values
put into /etc/systemd/system/docker.service.d/flannel-options.conf
to renumber the docker0 to work correctly with flannel.
2017-03-21 15:22:14 -05:00
Matthew Mosesohn
51985ab9de Change wait for dnsmasq to skip if there are no kube-nodes in play
Also changed unnecessary delay to a max timeout (now defaulting to 1s sleep
between tries)

Also rename play_hosts to ansible_play_hosts
2017-03-21 18:55:22 +03:00
Matthew Mosesohn
df09b8dc36 Merge pull request #1159 from holser/etcd_backup_restore
Backup etcd
2017-03-21 13:07:44 +03:00
Matthew Mosesohn
b68e545bad Merge pull request #1155 from mattymo/helm
Add helm deployment
2017-03-20 17:00:06 +03:00
Sergii Golovatiuk
e635fecc01 Backup etcd data before restarting etcd
etcd is crucial part of kubernetes cluster. Ansible restarts etcd on
reconfiguration. Backup helps operator to restore cluster manually in
case of any issues.

Signed-off-by: Sergii Golovatiuk <sgolovatiuk@mirantis.com>
2017-03-20 14:50:52 +01:00
Matthew Mosesohn
f9ef55c796 Merge pull request #1152 from mattymo/redhat_weave
Fix weave on RHEL deployment
2017-03-19 16:45:20 +03:00
Matthew Mosesohn
182a2e682f Merge pull request #1149 from mattymo/centos-retries
Retry yum/apt/rpm download commands
2017-03-18 11:12:36 +03:00
Matthew Mosesohn
579a6299c0 Add helm deployment 2017-03-17 20:24:41 +03:00
Matthew Mosesohn
dd56342361 Retry yum/apt/rpm download commands, fix succeeded filter 2017-03-17 18:56:26 +03:00
Matthew Mosesohn
b060d3279c Merge pull request #1146 from mattymo/resolvconf_optimize
Condense resolvconf sources before starting loop
2017-03-17 18:42:32 +03:00
Matthew Mosesohn
446a4bbdc8 Fix weave on RHEL deployment
Reduce retry delay checking weave
Always load br_netfilter module
2017-03-17 18:17:47 +03:00
Matthew Mosesohn
a5876a1ac8 Merge pull request #1136 from adidenko/fix-calico-policy-order
Move calico-policy-controller into separate role
2017-03-17 17:32:14 +03:00
Aleksandr Didenko
1adf231512 Move calico-policy-controller into separate role
By default Calico CNI does not create any network access policies
or profiles if 'policy' is enabled in CNI config. And without any
policies/profiles network access to/from PODs is blocked.

K8s related policies are created by calico-policy-controller in
such case. So we need to start it as soon as possible, before any
real workloads.

This patch also fixes kube-api port in calico-policy-controller
yaml template.

Closes #1132
2017-03-17 11:21:52 +01:00
Matthew Mosesohn
2586d01345 Condense resolvconf sources before starting loop 2017-03-17 13:06:56 +03:00
Matthew Mosesohn
3ee77a08cd Update calico to 1.1.0-rc8
Fixes bug in CentOS/RHEL in felix related to overlayfs driver.
2017-03-16 19:23:36 +03:00