github
/
k3s
mirror of https://github.com/k3s-io/k3s
1
0
Fork 0
Code Issues Projects Releases Wiki Activity
34,786 Commits
32 Branches
881 Tags
633 MiB
Commit Graph

1001 Commits (be889f892a36082a9b39f32fd3716101d1abcbe9)

Author SHA1 Message Date
CJ Cullen c77e800495 Add liveness probe for master's etcd pod 2015-08-25 10:37:00 -07:00
derekwaynecarr d25082250d Update salt deployment for new limit-range 2015-08-24 15:43:32 -04:00
Zach Loafman 185b5af26f Merge pull request #12833 from uluyol/insecure-reg 
Launch a cluster-local registry.
 2015-08-21 14:58:37 -07:00
Muhammed Uluyol 3dc10a209b Use gcr.io for proxy image 2015-08-21 13:32:08 -07:00
Muhammed Uluyol 2fb4e7b7a4 Use quotes more consistently 2015-08-20 18:44:12 -07:00
Muhammed Uluyol 14b554cef6 Use a persistent volume for the docker registry. 2015-08-20 18:44:11 -07:00
Muhammed Uluyol 7129d477d3 Launch a cluster-local registry. 
This registry can be accessed through proxies that run on each node
listening on port 5000. We send the proxy images to the nodes directly
to avoid requests that hit the network during cluster launch. For now,
we continue to pull the registry itself over the network, especially
given its large size (we should be able to dramatically shrink the
image). On GCE we create a PD and use that for storage, otherwise we
use an emptyDir. The registry is not enabled outside of GCE. All
communication is currently plain HTTP. In order to use SSL, we will
need to be able to request a certificate/key from the apiserver signed
by the apiserver's CA cert.
 2015-08-20 18:44:05 -07:00
Clayton Coleman 02dbb95447 Add TerminationGracePeriodSeconds to API 
Set defaulting for pod spec
 2015-08-20 11:03:38 -04:00
Saad Ali 3fce3433d9 Merge pull request #12949 from satnam6502/fluentd-es 
Update Elasticsearch cluster logging to use v1.9 image
 2015-08-19 18:48:55 -07:00
Satnam Singh d3a8e5374b Update Elasticsearch cluster logging to use v1.9 image 2015-08-19 15:55:22 -07:00
Robert Bailey 08e6a43c1d Revert "Merge pull request #9165 from smarterclayton/graceful" 
This reverts commit 4f856b595d, reversing
changes made to d78525a83b.

Conflicts:
	pkg/kubelet/status_manager.go
 2015-08-18 17:34:49 -07:00
Clayton Coleman b842a7dd15 Revert "Revert "Gracefully delete pods from the Kubelet"" 
This reverts commit 98115facfd.
 2015-08-18 08:57:02 -04:00
Eric Paris c560f41a4b Update docs which were incorrectly using _ in flag names 
Each of these was verified that the actual flag declaration correctly
used -
 2015-08-17 11:18:43 -07:00
Justin Santa Barbara 0ff8dd45ef Fix escaping problem in kube-scheduler manifest 
'{{pillar[log_level]}}' was appearing literally in the command line,
instead of being substituted.

Fixes #12787
 2015-08-17 08:16:24 -04:00
Dawn Chen 1932b48db3 Restore the logic of disable docker upgrade through salt for gce. 
Currently it is commented out.
 2015-08-11 14:55:58 -07:00
Dawn Chen 3901b225e3 Docker 1.7.1 2015-08-11 14:55:57 -07:00
Marek Grabowski 5f9cefc1d8 Merge pull request #12441 from vlajos/typofixes-vlajos-20150807 
typofix - https://github.com/vlajos/misspell_fixer
 2015-08-10 16:33:52 +02:00
Ed Costello 35a5eda585 Copy edits for typos 2015-08-09 14:18:06 -04:00
Veres Lajos 9f77e49109 typofix - https://github.com/vlajos/misspell_fixer 2015-08-08 22:31:48 +01:00
Eric Paris 86ca88be61 convert cluster/ from --flag_name= --flag-name= 2015-08-07 11:00:13 -04:00
Dawn Chen 2be69e7718 Merge pull request #12325 from brendandburns/supervisord 
Add support for supervisord as a monit alternative.
 2015-08-06 16:16:35 -07:00
Brendan Burns 15b9d98733 Add support for supervisord as a monit alternative. 2015-08-06 15:48:45 -07:00
Robert Bailey 8df33bc1a7 Register the kubelet on the master node with an apiserver. This option is 
separated from the apiserver running locally on the master node so that it
can be optionally enabled or disabled as needed.

Also, fix the healthchecking configuration for the master components, which
was previously only working by coincidence:

If a kubelet doesn't register with a master, it never bothers to figure out
what its local address is. In which case it ends up constructing a URL like
http://:8080/healthz for the http probe. This happens to work on the master
because all of the pods are using host networking and explicitly binding to
127.0.0.1. Once the kubelet is registered with the master and it determines
the local node address, it tries to healthcheck on an address where the pod
isn't listening and the kubelet periodically restarts each master component
when the liveness probe fails.
 2015-08-06 13:39:32 -07:00
Marek Grabowski 77675b052d Merge pull request #12301 from derekwaynecarr/fix_systemd_salt 
Fix salt configuration on systemd setups
 2015-08-06 11:36:27 +02:00
Mike Danese fe6b15ba2f rewrite all links to issues to k8s links 2015-08-05 21:11:11 -07:00
derekwaynecarr 0de0e3e9f1 Fix salt configuration on systemd setups 2015-08-05 17:01:43 -04:00
Alex Mohr e4fde6d2ca Merge pull request #9583 from eparis/make-cert-path 
generate-cert: allow for alternative paths
 2015-08-05 00:04:17 -07:00
Alex Robinson cc02e50fc7 Collect more of the VM's log files (monit, salt) in fluentd-es. 2015-08-05 00:22:38 +00:00
Alex Robinson 7430181d65 Collect more of the VM's log files (monit, salt) in fluentd-gcp. 2015-08-05 00:21:47 +00:00
Eric Paris 0753acf2f8 Allow make-ca-cert.sh to run on OS X 
Currently make-ca-cert.sh uses (equiv of)

mktemp -d --tmpdir kube.XXXXX

but --tmpdir is not a valid option on OS X. Switch to

mktemp -d -t kube.XXXXX

Which is valid, but subtly different between OS X and Linux. The
directory you get back will be different on each.

Linux:  ${tmpdir}/kube.y5Bsu/
OS X:   ${tmpdir}/kube.XXXXX.VQ81oOui/
 2015-08-04 19:03:01 -04:00
Eric Paris c69b6f5ba7 Respect KUBECTL_BIN in kube-addon-update.sh not just kube-addons.sh 2015-08-04 19:02:31 -04:00
Eric Paris 6b9ef5b2d8 generate-cert: allow for alternative paths 
Instead of hard coding kube-cert and /srv/kubernetes allow these to be
overwritten by environment variables.  / is immutable on some systems
and so /srv is not a possible location to store data.
 2015-08-04 19:01:16 -04:00
Alex Mohr ad4086eab1 Merge pull request #11308 from gmarek/pillar 
Auto commit by PR queue bot
 2015-08-04 12:42:57 -07:00
Alex Mohr 4c5763f28c Merge pull request #11299 from wojtek-t/increase_kubelet_start_timeout 
Auto commit by PR queue bot
 2015-08-03 16:14:02 -07:00
Justin Santa Barbara 463458bf5f AWS: Install ntp daemon on all nodes 
The AWS API requires a signature on method calls, including the
timestamp to prevent replay attacks.  A time drift of up to 5 minutes
between client and server is tolerated.

However, if the client clock drifts by >5 minutes, the server will start
to reject API calls (with the cryptic "AWS was not able to validate the
provided access credentials").

To prevent this happening, we install ntp on all nodes.

Fix #11371
 2015-08-01 14:32:08 -04:00
Jian Huang 45c37b9bd0 Update README.md 
There is a bad link in the doc. change ../../docs/salt.md to ../../docs/admin/salt.md
 2015-07-31 18:44:06 +08:00
gmarek 5bb7eec5ab Use salt pillars to pass test arguments during cluster creation. 2015-07-31 09:32:01 +02:00
Mike Danese bfaa527139 Merge pull request #12045 from a-robinson/logs 
Fix fluentd configs to collect etcd logs on the master
 2015-07-30 17:36:46 -07:00
Alex Robinson 69ab7026af Update fluentd-es to collect etcd logs and remove merge conflict cruft. 2015-07-30 22:21:01 +00:00
Alex Robinson 9e18c29e9f Make fluentd-gcp collect etcd's logs on the master. 2015-07-30 22:20:57 +00:00
Alex Robinson 94ae0a9a0d Make the kubelet on a GCE master check instance metadata for manifests. 
Primary motivation: enable GKE and other cluster-as-a-service folks to
easily run additional logic on the master without having to modify salt
or SSH to the master after it's been created.
 2015-07-30 21:04:34 +00:00
Janet Kuo 180798cfa4 Use example syncer tags instead of hard-coded examples in doc 2015-07-27 14:48:41 -07:00
Piotr Szczesniak f48543aba5 Made enabling Kube UI configurable 2015-07-27 08:23:04 +02:00
Mike Danese 01349cdbfa Merge pull request #11786 from a-robinson/fluentd 
Update fluentd-gcp images to include latest Dockerfile changes
 2015-07-24 12:40:41 -07:00
Brendan Burns 26de8cd6b2 Merge pull request #10601 from marun/docker-pkg-f22 
Auto commit by PR queue bot
 2015-07-23 22:15:50 -07:00
Alex Robinson 0889307522 Update the fluentd-gcp images to include their latest Dockerfile changes. 
Their Dockerfiles were changed in #11768.
 2015-07-24 01:27:42 +00:00
Vish Kannan 919fded0f8 Merge pull request #10401 from justinsb/workaround_salt_18089 
Work around Salt 18089
 2015-07-23 16:52:21 -07:00
Dawn Chen ebc36c0e5a timeout docker load: 30 second, and restart docker daemon to workaround docker-load-hang issue. 2015-07-17 15:16:23 -07:00
Dawn Chen 7e665230c7 Stop monit kube-addon on nodes. 2015-07-15 11:11:55 -07:00
Wojciech Tyczynski 628ceac305 Increase kubelet start timeout in Monit config 2015-07-15 13:28:07 +02:00
Victor Marmol cabecc1271 Merge pull request #10611 from marekbiskup/addon-update-kill-children 
Addon update kill children
 2015-07-09 21:09:58 -07:00
Victor Marmol ee8d2857fc Merge pull request #9381 from justinsb/aws_support_jessie 
AWS: Add support for Vivid Vervet & Debian Jessie
 2015-07-09 15:20:37 -07:00
Marek Biskup aa608258a6 kill child processes of addon-update on stop 2015-07-09 13:10:54 +02:00
Victor Marmol f1e66c718d Merge pull request #10590 from roberthbailey/kubelet-on-nodes 
Install the kubectl binary on nodes in addition to the master.
 2015-07-08 16:13:12 -07:00
Justin Santa Barbara 89900fa19b AWS: Ubuntu Vivid, set cgroup_root to docker 
Otherwise libcontainer generates systemd unit file names that start with "-",
which are not valid.
 2015-07-08 17:00:43 -04:00
Victor Marmol dd37642369 Merge pull request #10488 from justinsb/salt_support_systemd 
Salt: support systemd (don't assume Redhat <=> systemd)
 2015-07-08 13:43:08 -07:00
Victor Marmol 3be3603ef3 Merge pull request #10877 from brendandburns/aws 
Fix AWS Setup.
 2015-07-08 11:33:32 -07:00
derekwaynecarr e2ddd2dd7b Missing ca crt in vagrant controllers 2015-07-08 10:59:10 -04:00
Brendan Burns 6d08ade4ce Don't start Docker by default on AWS 2015-07-07 21:57:43 -07:00
Satnam Singh 90e105521c Wait for service accounts in the kube-system namesapce 2015-07-06 18:12:25 -07:00
Brendan Burns 988aa6fdf6 Move things into a 'kube-system' namespace. 2015-07-06 15:08:23 -07:00
Yu-Ju Hong 1e0a16392d Merge pull request #10780 from cjcullen/testfix 
Hacky workaround for #9822
 2015-07-06 15:00:35 -07:00
Yu-Ju Hong b58e7c8c2d Merge pull request #10655 from dchen1107/cadvisor 
Set minimal shares for containers with no cpu specified
 2015-07-06 11:38:54 -07:00
Tim Hockin 6c88401889 Merge pull request #10271 from erictune/how-to-write-a-gsg 
Scratch Getting-started-guide
 2015-07-06 11:00:44 -07:00
Eric Tune 682abd9dab Scratch Getting-started-guide 2015-07-06 10:38:20 -07:00
CJ Cullen edd419790e Hacky workaround for #9822 2015-07-06 10:02:53 -07:00
Yu-Ju Hong 8ce6fe855b Merge pull request #10712 from justinsb/always_install_socat 
Salt: install socat; needed for port-forward
 2015-07-06 09:58:52 -07:00
Yu-Ju Hong 83fbcba97b Merge pull request #10698 from justinsb/aws_master_sans 
AWS: Configure SSL certificate alternate-names
 2015-07-06 09:57:44 -07:00
Zach Loafman a2250d1caf Merge pull request #10597 from a-robinson/fluentd 
Run fluentd on the master to collect the core master logs
 2015-07-04 21:12:47 -07:00
Justin Santa Barbara 6504df77f1 Salt: install socat; needed for port-forward 
port-forward needs socat on the node hosts; we technically
don't need it today on the master, but this seems the right
place to put it, and socat is a small dependency.
 2015-07-03 19:12:44 -04:00
Justin Santa Barbara 25007421ba Salt: re-remove comment block 
It had been removed in upstream, but had been kept by mistake here.
 2015-07-03 12:43:36 -04:00
Justin Santa Barbara 600a0d6fe7 Salt: have kube-addons service depend on init file 
For symmetry with systemd
 2015-07-03 12:41:59 -04:00
Justin Santa Barbara a5b3c73ac4 Salt: Add service block for kube-proxy for systemd 
We have the workaround for Salt's problems here, but we should still express
our intent.
 2015-07-03 10:26:21 -04:00
Justin Santa Barbara 60d822435b Salt: clean daemon_args on RedHat, not systemd 
I don't think it actually makes a difference based on how we're using it,
but this is now more of a precise change.
 2015-07-03 10:24:32 -04:00
Justin Santa Barbara ad0293e3f3 Salt: Add more dependencies to systemd services 
This may help Salt reload services correctly, although we still
need the script until Salt's bug with reloading services on systemd
is resolved.

Salt bug: https://github.com/saltstack/salt/issues/16778
 2015-07-03 01:40:15 -04:00
Justin Santa Barbara 944fc1ed58 Set standard bash options in services helper script 2015-07-03 01:40:15 -04:00
Justin Santa Barbara bcd92c7522 Salt: support systemd (don't assume Redhat <=> systemd) 
Also work around problems with Salt & systemd services, in particular
that Salt doesn't issue a daemon-reload.
 2015-07-03 01:40:15 -04:00
Justin Santa Barbara c676c11189 AWS: Configure SSL certificate alternate-names 
GCE does this in its per-provider scripts; this does the same for AWS and lets
other providers do the same; I believe kube2sky requires 10.0.0.1 as a SAN.
 2015-07-03 01:18:07 -04:00
Dawn Chen 9ce11545c0 Explicitly configure master component containers and nginx with 200m each. 2015-07-02 10:46:39 -07:00
Zach Loafman 1d16be675c Merge pull request #10653 from dchen1107/resource_management 
Set resource limit for addon containers
 2015-07-01 19:35:28 -07:00
Dawn Chen 54531d9b55 Set memory limit (200mi) to fluentd static pod 2015-07-01 17:00:06 -07:00
Zach Loafman b60c8e567f Merge pull request #10077 from timstclair/ui-server 
Move dashboard UI out of the apiserver to a separate pod
 2015-07-01 15:56:33 -07:00
nikhiljindal 274792d7bb Stop exposing v1beta3 by default 2015-07-01 14:38:02 -07:00
Tim St. Clair b6b2d6df42 Add kube-ui cluster addon for serving k8s dashboard UI. 
Changes include:
- Add kube-ui binary for serving static dashboard UI
- Add kube-ui docker image, replication controller, and service
- Make the kube-ui a cluster-addon (enabled by default)
- Split the compiled pkg/ui/datafile.go into separate dashboard and swagger packages
- Update docs to reflect changes
 2015-07-01 14:21:47 -07:00
Alex Robinson e5ea6ae854 Run fluentd on the master node by default. 
This involves adding the logging-write scope to the master when running on GCE.
 2015-07-01 20:07:52 +00:00
Marek Biskup 8df3a9cae8 namespaces in addon update 2015-07-01 21:31:52 +02:00
Zach Loafman 3094c1624b Merge pull request #10492 from derekwaynecarr/add_watch_service_file 
Fix Vagrant kube-up - missed file watch on systemd stops kubelet starting on initial kube-up
 2015-07-01 09:49:59 -07:00
Marek Biskup 7873e252f0 addon updater should not retry too many times because specs may be invalid 2015-07-01 16:22:32 +02:00
derekwaynecarr de78898170 Missing file watch on systemd stops kubelet starting on kube-up 2015-07-01 09:59:26 -04:00
Zach Loafman b83236ee7a Merge pull request #10568 from derekwaynecarr/fix_vagrant_kube_proxy 
Missed update to kube-proxy when removing nginx on vagrant
 2015-07-01 06:42:00 -07:00
Maru Newby 2da59a2f34 Install 'docker' package for Fedora 22 
Fedora < 22 provides docker via the 'docker-io' package, but this
package was renamed to 'docker' as of Fedora 22.  Though the docker
package can be installed manually with 'dnf install docker-io' on F22,
Salt requires the explicit package name or deployment will fail.
 2015-06-30 21:36:33 -07:00
Robert Bailey fedddee157 Install the kubectl binary on nodes in addition to the master. 2015-06-30 16:52:17 -07:00
derekwaynecarr 2cc869b1c5 Missed update to kube-proxy when removing nginx on vagrant 2015-06-30 15:42:52 -04:00
Dawn Chen e55cd709a1 Removed cluster/saltbase/salt/_states/container_bridge.py since it is useless now. 2015-06-30 11:00:41 -07:00
Zach Loafman f21b6b89d1 Fix issues mentioned in #10520 2015-06-30 06:51:10 -07:00
Daniel Smith dcd91666b9 wait until a token shows up to start addons 2015-06-29 18:41:12 -07:00
Justin Santa Barbara e29e9f0033 Add explicit dependency of docker pkg on the deb download 
Salt ordering continues to surprise me.  I saw them execute
out of order, though I don't know why.  Adding an explicit
dependency to prevent out-of-order execution.
 2015-06-28 23:51:12 -04:00
Justin Santa Barbara ea74bd13b8 Work around Salt 18089 
Directly create /etc/sysctl.d/99-salt.conf
 2015-06-28 23:50:00 -04:00
Justin Santa Barbara 54d3ac92ea Don't make kubelet systemd service depend on Docker 
Fixes #10379
 2015-06-26 11:03:09 -04:00
Maxwell Forbes bc1930927d Merge pull request #10347 from justinsb/vagrant_no_cidr_on_master 
Salt: don't assume that master => cbr-cidr
 2015-06-25 10:23:59 -07:00
Maxwell Forbes 655179dcfb Merge pull request #10264 from mikedanese/ca-token 
add ca cert to token controller and all service accounts
 2015-06-25 09:56:35 -07:00
Justin Santa Barbara a0ae1d90ba Salt: don't assume that master => cbr-cidr 
In particular, this is required for vagrant
 2015-06-25 09:17:00 -04:00
Brendan Burns d2334d60c8 Merge pull request #10328 from dchen1107/cleanup 
Add dependencies on docker and kubelet service when copying master co…
 2015-06-24 21:32:35 -07:00
Dawn Chen e7d217d941 Add dependencies on docker and kubelet service when copying master components manifests 
to /etc/kubernetes/manifest
 2015-06-24 21:15:08 -07:00
Maxwell Forbes 3afda5d566 Merge pull request #10312 from dchen1107/cleanup 
Take 2: Fix the race between configuring cbr0 and restarting static pods
 2015-06-24 17:59:50 -07:00
Mike Danese 56bde3342a add ca to token controller and all service accounts 2015-06-24 15:10:20 -07:00
Dawn Chen 6ddfa512de Revert "Revert "Fix the race between configuring cbr0 and restarting static pods"" 
This reverts commit fd0a95dd12.
 2015-06-24 11:10:10 -07:00
Piotr Szczesniak fd0a95dd12 Revert "Fix the race between configuring cbr0 and restarting static pods" 2015-06-24 09:56:49 +02:00
Dawn Chen 9dbe6fe4e4 Added more comments. 2015-06-23 16:28:41 -07:00
Dawn Chen 8d76d4ee57 Make master-addon service depend on both docker and kubelet service 
in salt.
 2015-06-23 12:11:20 -07:00
Dawn Chen 23200d303f Fix several issues on running syncPods until network is configured. 
Also fixed unittests and compiling.
 2015-06-23 12:11:19 -07:00
derekwaynecarr db202d4904 Remove nginx from vagrant 2015-06-23 13:07:50 -04:00
Brendan Burns 192ffdfb25 Fix the container bridge so that it can create cbr0 
Fix the kubelet so that it tries to sync status, even if Docker is down
 2015-06-22 23:18:01 -07:00
Brendan Burns a29ade2f33 Disable creation of cbr0, the kubelet does it now. 
Conditionalize the docker bridge.
 2015-06-22 23:15:29 -07:00
Jeff Lowdermilk 85770295c2 Merge pull request #10004 from ArtfulCoder/kubelet_health 
Augmented kubelet healthz with syncLoop check.
 2015-06-22 10:53:05 -07:00
Abhishek Shah 3556993179 Augmented kubelet healthz with syncLoop check. 
Monit uses read-only port for health monitoring.
 2015-06-19 18:14:01 -07:00
Brendan Burns ec1e308ae2 rev etcd to 2.0.12 2015-06-19 15:57:32 -07:00
Saad Ali 4d25121206 Merge pull request #9769 from brendandburns/secure 
Revert the revert of https://github.com/GoogleCloudPlatform/kubernetes/pull/9761
 2015-06-16 14:21:54 -07:00
Saad Ali 068000a8ba Merge pull request #9847 from mikedanese/apiserver-liveness-probe 
add livenessProbe to kube-apiserver.manifest
 2015-06-16 14:03:13 -07:00
Vishnu Kannan 31f7ea10c6 Updating heapster version to v0.14.1. 
Adding a standalone version of heapster which exposes stats via REST API.
 2015-06-16 11:56:23 -07:00
Saad Ali 5596e6f789 Merge pull request #9493 from marekbiskup/testAddonUpdate 
e2e test for addon upgrade
 2015-06-16 10:50:15 -07:00
Mike Danese 491eb5e583 add livenessProbe to kube-apiserver.manifest 2015-06-15 23:22:56 -07:00
Fabio Yeon 9b3e9a509d Merge pull request #9616 from jeffbean/kube_addons 
kube_addons - Adding variable with default for kubectl bin.
 2015-06-15 19:31:42 -07:00
Fabio Yeon 8d2c7ff1d7 Merge pull request #9687 from satnam6502/fluentd-gcp 
Adjust Fluentd source format for Docker files to be JSON for Cloud Lo…
 2015-06-15 13:32:49 -07:00
Satnam Singh 33311b46a2 Adjust Fluentd source format for Docker files to be JSON for Cloud Logging 2015-06-15 10:22:27 -07:00
Marek Biskup dcc4034d57 e2e test for addon update 2015-06-15 17:37:24 +02:00
Brendan Burns 51b20b35e2 Revert "Revert "Optionalize (default false) --insecure-registry."" 
This reverts commit 1645c9a9b8.
 2015-06-12 21:00:16 -07:00
Dawn Chen 1645c9a9b8 Revert "Optionalize (default false) --insecure-registry." 2015-06-12 17:50:38 -07:00
Justin Santa Barbara 16d6ba8a9d AWS: Stop the master kubelet from registering as a node (like GCE does) 2015-06-12 18:33:45 -04:00
Brendan Burns 675d8378f2 Optionalize (default false) --insecure-registry. 2015-06-11 16:33:14 -07:00
Abhi Shah 59a347d119 Merge pull request #9309 from saad-ali/issue9028 
Enable InfluxDB/Grafana for GCE in addition to GCL. Disable GCM
 2015-06-11 11:04:09 -07:00
Abhi Shah f4eb2f366f Merge pull request #9568 from marekbiskup/docker_log_rotate 
rotate docker log
 2015-06-10 14:50:07 -07:00
Jeffrey Bean 99f64a8a07 kube_addons - Adding variable with default for kubectl bin. Fixes #9599 2015-06-10 14:47:31 -07:00
Abhi Shah 53fa4e1366 Merge pull request #9383 from brendandburns/liveness 
Add liveness checks to the controller-manager and scheduler.
 2015-06-10 14:44:12 -07:00
Abhi Shah c8591bd323 Merge pull request #9491 from ArtfulCoder/etcd_log 
Etcd log mount
 2015-06-10 13:17:10 -07:00
Marek Biskup 83f8cd148d rotate docker log 2015-06-10 12:19:13 +02:00
Brendan Burns ff3f463e00 Add liveness checks to the controller-manager and scheduler. 2015-06-09 13:06:04 -07:00
Robert Bailey 2feb658ed7 Distribute the cluster CA cert to cluster addon pods through 
the kubeconfig file. Use the $KUBERNETES_MASTER_NAME from the
kube-env for skydns, because it can't use the service name.
 2015-06-08 20:17:45 -07:00
Abhishek Shah fe906677d4 Etcd log mount 2015-06-08 17:21:11 -07:00
krousey 8a9e0e00bf Merge pull request #9128 from marekbiskup/addonUpgrade 
Addon upgrade (partial implementation of #8107)
 2015-06-08 15:15:27 -07:00
Marek Biskup 13ba177668 kube-addon-update.sh 2015-06-08 15:10:40 +02:00
CJ Cullen 04cd9b3c75 Make sshproxy use a hostmount on master PD (don't spam sshKeys on upgrade/reboot). 
Add comment describing what SSHTunnelList.Close() does.
Simplify util.FileExists.
 2015-06-05 15:03:03 -07:00
CJ Cullen cb317604ab Some refactoring. Only selectively use ssh proxy. 
Add NetworkName to gce.Config.
Add locking to uses of master.tunnels.
 2015-06-05 14:55:16 -07:00
CJ Cullen de9a5f43bc Specify sshUser, sshKeyfile in kube-apiserver manifest. 
Trim space on ssh key so GCE doesn't treat it as 2 lines.
A couple other minor fixes.
 2015-06-05 14:55:15 -07:00
CJ Cullen 82afaaf31d Mount cloud-config files for cloudproviders in kube-apiserver & kube-controllermanager. 2015-06-05 14:54:20 -07:00
saadali bc53533c83 Enable InfluxDB/Grafana for GCE in addition to GCM/GCL 2015-06-05 01:17:45 -07:00
derekwaynecarr 2168cee414 Upgrade to Fedora 21, Docker 1.6, clean-up SDN 2015-06-04 10:59:23 -04:00
Daniel Smith 1690617ee6 remove ro service 2015-06-03 16:45:54 -07:00
Brian Grant 984b779b05 Merge pull request #9192 from dchen1107/clean 
Watch kubelet flag file, so restart kubelet upon changes.
 2015-06-03 14:45:07 -07:00
Brian Grant d37264edde Merge pull request #9158 from dchen1107/cleanup 
stop monit kube-proxy on master node.
 2015-06-03 12:13:17 -07:00
Dawn Chen 1d94348023 Watch kubelet flag file, so restart kubelet upon changes. 2015-06-03 09:13:28 -07:00
Prashanth Balasubramanian f7c0f1c1e3 Set min-request-timeout in test clusters 2015-06-03 08:46:28 -07:00
Brian Grant d861631f08 Merge pull request #9142 from roberthbailey/kubelet-on-master 
Prevent access to the /run and /exec endpoints on the master's kubelet
 2015-06-02 22:55:06 -07:00
Dawn Chen 301bed535e stop monit kube-proxy on master node. 
Fixed #8783
 2015-06-02 17:48:24 -07:00
CJ Cullen 934c553c04 Clarify description/usage of --advertise-address, Master.PublicAddress 2015-06-02 15:23:32 -07:00
Robert Bailey 1dad552edc Prevent access to the /run and /exec endpoints on the master's kubelet 
so that users can't run arbitrary code on the master.
 2015-06-02 14:42:11 -07:00
CJ Cullen 085a48a70e Add an advertise-address flag. This allows the address that the apiserver binds 
to (possibly 0.0.0.0) to be different than the address on which members of the cluster
can reach the apiserver (possibly not a local interface).
 2015-06-02 14:33:15 -07:00
Justin Santa Barbara c4a2631593 Mount logic breaks if /var/lib/kubelet is a symlink 
Pass the correct kubelet root-dir on AWS
 2015-05-29 20:13:09 -04:00
Rohit Jnagal 6274d845b9 Merge pull request #8931 from dchen1107/clean 
Fix the race when monit is used to monitor services which are also
 2015-05-29 11:23:21 -07:00
Rohit Jnagal 50b9d6284a Merge pull request #8681 from vmarmol/system-container 
Create a system container
 2015-05-29 09:41:06 -07:00
Tim Hockin ac3cc3c518 Rename PORTAL_NET all over 2015-05-28 16:10:44 -07:00
Tim Hockin 3005471100 Add new apiserver flags for clusterIP (nee portal) 
Leave old flags but marked as deprecated
 2015-05-28 16:10:44 -07:00
Dawn Chen 79a2a4f18d Fix the race when monit is used to monitor services which are also 
started at boot time via init.d at boot time.
 2015-05-28 11:58:33 -07:00
Eric Tune b76dd0e585 Merge pull request #8722 from roberthbailey/dead-flags-salt 
Remove deprecated nodecontroller flags from the salt configuration.
 2015-05-27 08:22:26 -07:00
Victor Marmol c97dda068d Create a /system system container in Debian. 2015-05-26 20:52:09 -07:00
Saad Ali fae1636eda Merge pull request #8586 from vmarmol/use-root-containers 
Run Docker containers as top-level containers in Debian.
 2015-05-26 18:44:36 -07:00
Saad Ali bed9f826d3 Merge pull request #7435 from jsafrane/devel/volume-tests 
V2: Add e2e tests for NFS and Gluster
 2015-05-26 17:49:41 -07:00
Robert Bailey e46b5641ae Remove deprecated flags from the salt configuration. 2015-05-26 12:36:09 -07:00
Jan Safranek 20004e0c16 Add e2e tests for Gluster and NFS tests. 
- add appropriate server containers into contrib/for-tests/volumes-tester
- the tests are off by default (they need kubelet --allow_privileged=True)
  - enable by 'go run hack/e2e.go ... --ginkgo.focus=Volume'
- add glusterfs tools to list of installed packages on each node
 2015-05-25 12:02:41 +02:00
CJ Cullen d6e0aedec1 Re-add the cluster_name flag that the ServiceController and RouteController need. 2015-05-22 16:31:22 -07:00
Victor Marmol 656be14e84 Run Docker containers as top-level containers in Debian. 
Part of #8215
 2015-05-22 13:27:25 -07:00
Dawn Chen 677a4aa1a7 Merge pull request #8164 from cjcullen/cloudprovider 
Route creation reconciler loop.
 2015-05-22 12:27:50 -07:00
Dawn Chen 626d25d50e Initial cpu limit for fluentd pods (file based). 2015-05-21 23:44:26 -07:00
Dawn Chen 456262c43c Merge pull request #8475 from dchen1107/clean 
Create default LimitRange object for cluster: 100m cpu.share per container
 2015-05-21 09:14:20 -07:00
CJ Cullen bf646abf8d Revert "Revert "Modify nodes to register directly with the master."" 
This reverts commit c53786ab31.
 2015-05-20 14:21:03 -07:00
Quinton Hoole c53786ab31 Revert "Modify nodes to register directly with the master." 2015-05-20 13:47:51 -07:00
Satnam Singh 32842b4d22 Update Fluentd GCP image 2015-05-19 16:12:50 -07:00
Dawn Chen 061155c1eb Create LimitRange object for cluster before addons service 2015-05-19 10:23:41 -07:00
Dawn Chen 55501a6314 Have default LimitRange created for cluster: 100m cpu.share per container 2015-05-19 10:23:40 -07:00
Robert Bailey 8e356f8439 Configure the cloud provider for the kubelet. 2015-05-19 09:55:07 -07:00
Robert Bailey 01467e0bb8 Modify nodes to register directly with the master. 
- Delete nodes when they are no longer ready and don't exist in the
cloud provider.
 - Label each node with it's hostname.
 - Add flag to skip node registration.
 - Add a test for registering an existing node.
 2015-05-19 09:55:07 -07:00
Jeff Lowdermilk 553f9f822b Add ga-beacon analytics to gendocs scripts 
hack/run-gendocs.sh puts ga-beacon analytics link into all md files,
hack/verify-gendocs.sh verifies presence of link.
 2015-05-15 18:56:38 -07:00
Daniel Smith ce4b54ec70 Merge pull request #8209 from krousey/v1beta1_cluster 
Removing some v1beta1 uses in cluster/
 2015-05-15 14:56:41 -07:00
Kris Rousey 98c457c397 Updating /cluster to use v1beta 3 specs, and change a lot of polling to 
healthz instead of api endpoints.
 2015-05-15 14:17:55 -07:00
Victor Marmol d3c6fb0d6a Merge pull request #8275 from dchen1107/clean 
Install monit_watcher as a cron job (1m) to check if monit is running,
 2015-05-15 08:02:55 -07:00
Dawn Chen e6c159eda2 Install monit_watcher as a cron job (1m) to check if monit is running, 
and oom protect monit processes.
 2015-05-14 15:19:11 -07:00
Rohit Jnagal e4c98a05ea Merge pull request #8182 from GoogleCloudPlatform/revert-7863-etcd_maxprocs 
Revert "Set GOMAXPROCS for etcd"
 2015-05-14 11:19:56 -07:00
Victor Marmol 5e7a7b9206 Merge pull request #8241 from gmarek/remove_name 
Remove unused ClusterName from NodeController
 2015-05-14 09:49:34 -07:00
Victor Marmol 2b1b85468a Merge pull request #7330 from chrismoos/volume_format_script_updates 
Add safe_format_and_mount to hyperkube image.
 2015-05-14 09:17:20 -07:00
gmarek 8a4717d3f5 Remove unused ClusterName from NodeController 2015-05-14 18:01:23 +02:00
Dawn Chen 309a157665 Merge pull request #7984 from cjcullen/kubelet 
Kubelet configure cbr0 instead of configure-vm.sh
 2015-05-13 17:32:52 -07:00
Yu-Ju Hong ad83197c63 Merge pull request #8187 from satnam6502/nosecret 
Remove the use of secrets from node level logging agents
 2015-05-13 11:12:08 -07:00
Jeff Lowdermilk f8ebc4e21e Merge pull request #8104 from roberthbailey/kubelet-using-kubeconfig 
Replace the auth config file with a kubeconfig file when starting the kubelet on GCE.
 2015-05-13 11:09:27 -07:00
Dawn Chen e32daf4229 Monit watch kubelet and kube-proxy on master node too. 2015-05-13 10:36:00 -07:00
Satnam Singh 53ec5e38c7 Remove the use of secrets from node level logging agents 2015-05-13 10:11:07 -07:00
Filip Grzadkowski eecd7b8c0f Revert "Set GOMAXPROCS for etcd" 2015-05-13 17:53:04 +02:00
Robert Bailey c47b9178b4 Replace the auth config file with a kubeconfig file when 
starting the kubelet on GCE.
 2015-05-13 01:03:28 -07:00
CJ Cullen 31ea7d1295 Put cbr0-modifying stuff behind a flag. Address some other comments. 2015-05-12 23:00:28 -07:00
Alex Robinson a98d883134 Merge pull request #8121 from satnam6502/fluentd-gcp 
Convert node level logging to Cloud Logging to use token-system-logging
 2015-05-12 14:26:46 -07:00
Alex Robinson 96cb4360fa Merge pull request #8120 from satnam6502/fluentd-es 
Convert nodel level logging to ES to use token-system-logging
 2015-05-12 14:26:28 -07:00
Nikhil Jindal d75bd8bf2a Merge pull request #7101 from liggitt/service_account 
ServiceAccounts
 2015-05-12 10:23:41 -07:00
Satnam Singh ade18e4ff3 Conver node level logging to Cloud Logging to use token-system-logging 2015-05-12 10:17:45 -07:00
Satnam Singh f9af0658fa Convert nodel level logging to ES to use token-system-logging 2015-05-12 10:07:49 -07:00
Alex Robinson e2b7ef8d2c Merge pull request #8078 from satnam6502/fluentd-gcp 
Convert Fluentd to Cloud Logging pod specs to YAML
 2015-05-11 16:55:14 -07:00
Satnam Singh e246685468 Convert Fluentd to Cloud Logging pod specs to YAML 2015-05-11 14:55:46 -07:00
Satnam Singh 2f5ba28a17 Switch Fluentd to ES manifest back to YAML 2015-05-11 14:37:41 -07:00
Jordan Liggitt db1f0dc906 JWT token generation/verification 2015-05-11 17:18:06 -04:00
Robert Bailey 6c42cb154b Stop copying certs into /usr/share/nginx/ since nothing relies 
on that location any longer.
 2015-05-11 11:44:19 -07:00
Wojciech Tyczynski 600b00f2b1 Merge pull request #7863 from fgrzadkowski/etcd_maxprocs 
Set GOMAXPROCS for etcd
 2015-05-07 10:09:29 +02:00
Filip Grzadkowski 7e8bde53dd Set GOMAXPROCS for etcd 2015-05-07 08:11:39 +02:00
CJ Cullen f0c26052eb Fix controller-manager manifest for providers that don't specify CLUSTER_IP_RANGE 2015-05-06 17:50:04 -07:00
Maxwell Forbes 7426b533df Merge pull request #7862 from cjcullen/cbr0 
Remove restriction that cluster-cidr be a class-b
 2015-05-06 15:46:57 -07:00
saadali 4569de7a46 Enable Google Cloud Monitoring and Google Cloud Logging instead of 
Influxdb for Google Compute Engine deployments.
 2015-05-06 15:23:40 -07:00
CJ Cullen fbd125e4e2 Remove restriction that cluster-cidr be a class-b 2015-05-06 15:01:13 -07:00
Mark Lamourine b7c1b7eed2 Added port 6443 to kube-proxy default IP address for api-server 
added  api_servers_with_port variable and settings.

removed extra port string

split azure api_server host and port out
 2015-05-06 12:43:24 +00:00
Robert Bailey 357df12c02 Set allocate_node_cidrs to be blank by default. 2015-05-05 23:41:30 -07:00
Jerzy Szczepkowski e967ffd522 Added flag to set cluster class B network address for pods, add flag to disable allocation CIDRs for Pods. Fixed synchornization bug in NodeController registerNodes(). 2015-05-05 16:10:43 -07:00
Jeff Lowdermilk 87cdc3f032 Merge pull request #7672 from eparis/boilerplate-python 
Check license boilerplate for python files
 2015-05-05 09:23:36 -07:00
Chris Moos 1b814ffd2c Add safe_format_and_mount to hyperkube image. 
* Add volume format script into hyperkube image.
* Fix issue with format script on non-redhat platforms.
 2015-05-04 21:25:39 -07:00
Zach Loafman 2b3e52c230 Use built-in ContainerVM Docker version rather than installing lxc-docker 2015-05-04 16:33:00 -07:00
Eric Paris f416289a85 update all python with boilerplate 2015-05-04 18:37:47 -04:00
Robert Bailey e3ff2dc02d Merge pull request #7678 from justinsb/aws_api_on_443 
AWS: Change apiserver to listen on 443 directly, not through nginx
 2015-05-04 13:52:57 -07:00
Daniel Smith fbe3ec7513 Merge pull request #7316 from ArtfulCoder/master_components_logs 
logs for master components
 2015-05-04 11:20:03 -07:00
Justin Santa Barbara 57f7b658bb AWS: Change apiserver to listen on 443 directly, not through nginx 
Mirrors changes in GCE.  I think the same changes will be needed for vagrant.
 2015-05-02 16:59:04 -04:00
Eric Paris 6b3a6e6b98 Make copyright ownership statement generic 
Instead of saying "Google Inc." (which is not always correct) say "The
Kubernetes Authors", which is generic.
 2015-05-01 17:49:56 -04:00
Satnam Singh 941a457851 Trim Fleuntd tag for Cloud Logging 2015-04-30 13:39:35 -07:00
Satnam Singh e65f0aafd5 Trim cluster log tags to pod name and container name 2015-04-29 18:24:13 -07:00
Abhishek Shah 96439b5a52 logs for master components 2015-04-29 09:51:09 -07:00
Robert Bailey 8206aa9eac Salt configuration to add basic auth to GCE. 2015-04-28 14:07:54 -07:00
Satnam Singh 4a0cc7905d Convert node level logging agents to v1beta3 2015-04-27 17:21:42 -07:00
CJ Cullen 39c5bf363b Merge pull request #7303 from erictune/kube_env3 
kube-proxy uses token to access port 443 of apiserver
 2015-04-27 14:33:53 -07:00
derekwaynecarr 81e9395533 Fix vagrant regression due to #7326 2015-04-27 17:18:30 -04:00
Eric Tune 9044177bb6 Generate a token for kube-proxy. 
Tested on GCE.
Includes untested modifications for AWS and Vagrant.
No changes for any other distros.
Probably will work on other up-to-date providers
but beware.  Symptom would be that service proxying
stops working.

 1. Generates a token kube-proxy in AWS, GCE, and Vagrant setup scripts.
 1. Distributes the token via salt-overlay, and salt to /var/lib/kube-proxy/kubeconfig
 1. Changes kube-proxy args:
   - use the --kubeconfig argument
   - changes --master argument from http://MASTER:7080 to https://MASTER
     - http -> https
     - explicit port 7080 -> implied 443

Possible ways this might break other distros:

Mitigation: there is an default empty kubeconfig file.
If the distro does not populate the salt-overlay, then
it should get the empty, which parses to an empty
object, which, combined with the --master argument,
should still work.

Mitigation:
  - azure: Special case to use 7080 in
  - rackspace: way out of date, so don't care.
  - vsphere: way out of date, so don't care.
  - other distros: not using salt.
 2015-04-27 08:59:57 -07:00
Robert Bailey 846ffcff83 Pass the CA root cert into the apiserver so that the apiserver will 
perform client cert checks for authorization. Only enable on GCE where
the apiserver is terminating SSL connections from end users.
 2015-04-24 22:01:56 -07:00
Nikhil Jindal 84cb48be11 Merge pull request #7246 from satnam6502/es 
Convert Elasticsearch logging to v1beta3 and de-salt
 2015-04-24 09:21:41 -07:00
CJ Cullen 80af1c9e40 kube2sky using kubeconfig secret: take 2. Point system secrets at https://kubernetes. Override in clients that can't use DNS. 2015-04-23 18:13:16 -07:00
Satnam Singh c9b9e7651e Convert Elasticsearch logging to v1beta and de-salt 2015-04-23 13:06:15 -07:00
Dawn Chen f9156c281a Merge pull request #7123 from satnam6502/logging 
Propagate pod and container name for log files
 2015-04-23 10:13:08 -07:00
Satnam Singh 2444c1f943 Propagate pod and container name for log files 2015-04-22 15:08:51 -07:00
Dawn Chen 87e0d5da08 Merge pull request #7186 from ArtfulCoder/no_log_pod_on_master 
removed elasticsearch and fluentd-gcp pods from master
 2015-04-22 14:40:07 -07:00
Abhishek Shah 8cf11fd608 removed elasticsearch and fluentd-gcp pods from master 2015-04-22 12:54:57 -07:00
Robert Bailey dc45f7f9e6 Remove nginx and replace basic auth with bearer token auth for GCE. 
- Configure the apiserver to listen securely on 443 instead of 6443.
 - Configure the kubelet to connect to 443 instead of 6443.
 - Update documentation to refer to bearer tokens instead of basic auth.
 2015-04-22 11:11:20 -07:00
Eric Tune b944049fe9 Merge pull request #6916 from ArtfulCoder/controller-in-a-pod 
kube-controller-manager in a pod.
 2015-04-21 09:58:29 -07:00
Justin Santa Barbara 29592356f7 Rename salt minion_ip to hostname_override 
Prep work for making AWS node identifier -> instance id
 2015-04-21 06:24:49 -07:00
CJ Cullen 1be193098a Merge pull request #7072 from dchen1107/cleanup 
Remove fqdn dependency for node name
 2015-04-20 18:00:21 -07:00
Abhishek Shah f513aad380 kube-controller-manager in a pod. 2015-04-20 16:06:37 -07:00
Eric Tune c3203cba6d Create system secrets in kubeconfig format 
Was previously kubernetes_auth format.

Added defaults file which uses salt to fill in an env var
with the master's IP.

More thought needs to be given soon to how to make this
connection use a cert for the master, and how to support
multiple masters, and whether to use the DNS record
instead of an IP address.  But this PR unblocks some other
more urgent things, so doing it this way.
 2015-04-20 15:02:45 -07:00
Kenjiro Nakayama 2e702b0c61 Replace hostname -f with uname -n 2015-04-20 14:16:21 -07:00
Eric Tune 424ae1d67a Fix kube-addon retrying. 2015-04-20 07:38:39 -07:00
Robert Bailey 972cf60d2b Merge pull request #6957 from ArtfulCoder/no_monit 
remove unused file
 2015-04-17 11:40:36 -07:00
Abhishek Shah b08102d358 remove unused file 2015-04-16 21:51:51 -07:00
Abhishek Shah 647d5948a1 Docker Image install script - Sleep before each iteration of while loop 2015-04-16 21:46:35 -07:00
Derek Carr 6c394e4d81 Merge pull request #6911 from erictune/del_kub_user 
Remove unused kubelet user from salt.
 2015-04-16 11:45:26 -04:00
Eric Tune 955cc1e1df Remove unused kubelet user from salt. 
Fixes #1512.

Cluster creation and validation succeeds with this change.
 2015-04-16 07:46:56 -07:00
litian54 b5e4111234 Fix Jinja render bug if 'portal_net' or 'cluster_name' is undefined 2015-04-15 17:17:38 -07:00
Abhi Shah 72a694b6b3 Merge pull request #6823 from ArtfulCoder/scheduler-in-a-pod 
Scheduler in a pod
 2015-04-15 08:27:28 -07:00
Abhishek Shah a908bb78ea kube-scheduler in a pod 2015-04-14 17:53:12 -07:00
Quinton Hoole cd6daae014 Merge pull request #6271 from a-robinson/asynclb 
Manage load balancer creation and deletion asynchronously in a ServiceController
 2015-04-14 17:50:07 -07:00
Abhishek Shah 37eb4d85e1 Install kube-controller-manager image 2015-04-14 16:45:39 -07:00
Abhishek Shah 682e8cfa99 Load Kube Scheduler Image. 2015-04-14 15:20:24 -07:00
Alex Robinson ccc300289f Implement a ServiceController that watches services and handles keeping 
external load balancers up-to-date based on the service's specs, using
the new DeltaFIFO watch queue class. Remove the old registry REST
handler code for creating/updating/deleting load balancers.

Also clean up a bunch of the GCE cloudprovider code related to load balancers.
 2015-04-14 18:56:24 +00:00
Justin Santa Barbara 72687184b9 Don't always use aws.conf 
We don't usually need it, and it makes it harder to put apiserver into a container.
 2015-04-14 09:30:00 -07:00
Abhishek Shah c9d7010461 kube-apiserver in a pod. 2015-04-13 15:15:45 -07:00
Dawn Chen 67169ec87d Merge pull request #6719 from litian54/master 
Fix Rendering SLS 'base:nginx' failed if 'cloud' is not defined in minio...
 2015-04-13 10:42:48 -07:00
Dawn Chen 7f2f165797 Revert "kube-apiserver in a pod." 2015-04-10 22:54:52 -07:00
litian54 36ef786228 Fix Rendering SLS 'base:nginx' failed if 'cloud' is not defined in minion grains 2015-04-10 18:27:22 -07:00
Abhishek Shah 2f70395396 kube-apiserver in a pod. 2015-04-10 15:46:49 -07:00
Justin Santa Barbara a366f9ee88 Create the /usr/share/google dir in salt 2015-04-10 13:25:43 -07:00
Justin Santa Barbara b9fd56050a Add safe_format_and_mount script to aws 
Apache licensed :-)
 2015-04-10 13:25:42 -07:00
Derek Carr d2b6920a32 Merge pull request #6496 from ArtfulCoder/docker_image_install 
Load docker images of kubernetes components after docker is installed.
 2015-04-10 16:19:10 -04:00
Maxwell Forbes 7eb7387d97 Merge pull request #6403 from zmerlynn/salt_docs 
Update the existing salt.md, add the start of a Salt README tree
 2015-04-09 10:34:25 -07:00
Abhishek Shah 4996ccbf2d Load docker images of kubernetes components after docker is installed. 2015-04-09 09:06:42 -07:00
Dawn Chen 8ede54c511 Upgrade kubernetes to gcr.io/google_containers/etcd:2.0.9 2015-04-07 16:26:11 -07:00
Justin Santa Barbara 2426366ec8 Update cAdvisor with moved docker root on AWS 
We set up a symlink now, and we also pass docker_root into the kubelet.

The symlink is probably sufficient, but doing both feels safer.
 2015-04-07 10:40:22 -07:00
Abhishek Shah a0ea2c6718 Disable nginx service 2015-04-06 18:38:03 -07:00
Zach Loafman c292d2e8d6 Update the existing salt.md, add the start of a Salt README tree 
Starts to fix #6070
 2015-04-06 14:41:21 -07:00
Abhishek Shah 9231fae998 Prevent docker load. 2015-04-06 11:09:27 -07:00
Abhishek Shah fb665ede4c Run etcd on localhost for all providers. 2015-04-03 14:00:44 -07:00
Abhishek Shah 23c42cd8b2 Etcd listens on localhost for GCE 2015-04-03 11:40:09 -07:00
Yu-Ju Hong 76f1232a2e Merge pull request #6404 from dchen1107/master 
Using gcr.io/google_containers/etcd:2.0.8
 2015-04-03 08:51:41 -07:00
Zach Loafman de67b96ff7 Merge pull request #6396 from ArtfulCoder/master_logging 
Enable log collection from master.
 2015-04-02 18:50:30 -07:00
Abhishek Shah 39bb6d3777 Enable log collection from master. 2015-04-02 18:37:07 -07:00
Zach Loafman 12cf7681a9 Merge pull request #5470 from erictune/for-abhis 
Make secrets at cluster startup.
 2015-04-02 17:43:56 -07:00
Dawn Chen 814177c21b Using gcr.io/google_containers/etcd:2.0.8 2015-04-02 17:04:05 -07:00
Abhi Shah a918a719e0 Merge pull request #6334 from brendandburns/nginx 
Add an nginx docker image for use on the master.
 2015-04-02 16:58:02 -07:00
Brendan Burns 9f48a2f4e6 Add an nginx docker image for use on the master. 2015-04-02 16:42:17 -07:00
Eric Tune 59daeabaee Make secrets at cluster startup. 
These secrets will be used in subsequent PRs by:
scheduler, controller-manager, monitoring services,
logging services, and skydns.

Each of these services will then be able to stop using kubernetes-ro
or host networking.
 2015-04-02 15:58:45 -07:00
Zach Loafman c627a3598c Merge pull request #6384 from erictune/cleanup-addons.sh 
Use same addons script for init.d and systemd.
 2015-04-02 14:49:44 -07:00
Derek Carr 4ae016e022 Merge pull request #6382 from derekwaynecarr/fix_vagrant_kubelet 
Fix vagrant with etcd in pod
 2015-04-02 17:22:44 -04:00
Eric Tune b9570b3daa Use same addons script for init.d and systemd. 2015-04-02 14:07:23 -07:00
derekwaynecarr c6bf46acd7 Fix vagrant with etcd in pod 2015-04-02 16:55:29 -04:00
Dawn Chen 9b2f835a01 Merge pull request #6326 from ArtfulCoder/docker_image_creation 
Create Docker images for master components
 2015-04-02 13:12:17 -07:00
Abhishek Shah b1b779a8d5 docker image creation 2015-04-02 12:39:12 -07:00
Filip Grzadkowski 65c2942a82 Increase ulimit -n for apiserver. 2015-04-02 17:01:23 +02:00
CJ Cullen 5e6e67ba59 Add an alternative TokenSource to the GCE CloudProvider. 2015-04-01 17:52:30 -07:00
Dawn Chen ab1a8b1e7c Merge pull request #6298 from roberthbailey/kubelet-config 
Remove salt configuration that matches the default settings in the Kubelet.
 2015-04-01 14:31:31 -07:00
Derek Carr 2af9b54147 Merge pull request #6259 from zmerlynn/fix_cloud_provider 
Eliminate grains.cloud_provider (in preference to grains.cloud) from SaltStack
 2015-04-01 17:04:05 -04:00
Robert Bailey 40ebed845f Remove salt configuration that matches the default settings in 
the Kubelet.
 2015-04-01 12:13:40 -07:00
Abhishek Shah 8e3a41b52c Run etcd 2.0.5 in a pod 2015-04-01 11:38:21 -07:00
Zach Loafman b581320bf7 Eliminate grains.cloud_provider (in preference to grains.cloud) from SaltStack 
This variable can be entirely derived from grains.cloud, and it
simplifies the configuration somewhat. (Or someone convince me I'm
wrong. I'm happy to be wrong here.)
 2015-04-01 08:32:32 -07:00
Satnam Singh 73a9ab2928 Rename logging image to use google_containers 2015-03-31 16:00:52 -07:00
Victor Marmol f0b3493c26 Merge pull request #6240 from zmerlynn/fix_apiservers 
s/apiservers/api_servers/ in Salt
 2015-03-31 13:06:09 -07:00
Zach Loafman 6c219885e0 s/apiservers/api_servers/ in Salt 
It looks like api_servers finally won this battle. Kill off the
last remaining places passing it, but allow the kubelet Salt to
accept apiservers for a period of time.

(This was bothering my OCD.)
 2015-03-31 12:29:46 -07:00
Abhishek Shah 541219db77 Retry kube-addons creation if kube-addons creation fails. 2015-03-30 18:14:54 -07:00
Robert Bailey bc8af553a3 Configure the kubelet to bind a simple healthz server to a localhost 
port for monitoring by monit. This is in preparation for the standard
kubelet port to switch to SSL only (and eventually to only accepting
connections on the SSL port that present a proper client SSL cert).

Also standardize the formatting of the monit config files a bit.
 2015-03-30 14:20:30 -07:00
Victor Marmol b9214d4fcf Merge pull request #6122 from satnam6502/fluentd-gcp 
Update Salt config to use gcr.io Fluentd to Cloud Logging agent
 2015-03-30 08:02:12 -07:00
Zach Loafman 848134fdfa Revert "Running etcd 2.0.5 in a pod on master" 2015-03-28 07:36:39 -07:00
Dawn Chen 179fe870cc Merge pull request #4442 from ArtfulCoder/docker_etcd 
Running etcd 2.0.5 in a pod on master
 2015-03-27 18:33:25 -07:00
Daniel Smith 525bbfd175 Merge pull request #6103 from zmerlynn/remove_gce_node_names 
Remove the --machines SaltStack configuration on GCE
 2015-03-27 17:55:16 -07:00
Abhishek Shah 7750f35e7c Running etcd 2.0.5 in a pod on master 2015-03-27 17:50:10 -07:00
Satnam Singh 666e2b69ac Update Salt config to use gcr.io Fluentd to Cloud Logging agent 2015-03-27 16:39:12 -07:00
Satnam Singh ce1e73fccc Update Salt config use gcr.io Fluentd to ES image 2015-03-27 16:09:37 -07:00
Zach Loafman 68ccb97907 Remove the --machines SaltStack configuration on GCE 
Per https://github.com/GoogleCloudPlatform/kubernetes/issues/6072#issuecomment-87074456, this is no longer necessary.
We now no longer need a static node list. Woo!
 2015-03-27 14:44:19 -07:00
Abhishek Shah d84ff79cfd Added a catchall to set api server ip address in kubelet cmd params 2015-03-27 08:44:52 -07:00
Piotr Szczesniak c516aba8e8 Merge pull request #6043 from ArtfulCoder/kubelet_salt 
Modified salt config to get master ip address correctly
 2015-03-27 10:03:05 +01:00
Abhishek Shah f35a352782 Modified salt config to get master ip address correctly 2015-03-26 21:17:06 -07:00
Abhishek Shah d1015e785f Add salt configuration to remove cadvisor.manifest from master. cadvisor is now part of kubelet 2015-03-26 10:15:30 -07:00
Robert Bailey 268b617b84 Add local babysitting for the kube-proxy. 2015-03-25 23:24:24 -07:00
Brendan Burns 8bac135f85 Configure docker and kubelet on AWS master. 2015-03-24 17:19:06 -07:00
Satnam Singh 69f867dafb Upgrade etcd to v2.0.5 on the master 2015-03-23 13:52:04 -07:00
Justin Santa Barbara 60a07e972b Raise the file ulimit for kube-proxy for init.d systems 2015-03-19 19:36:13 -04:00
Satnam Singh 985d5c1f95 Add a prefix to the POS file for Fluentd to Elasticsearch 2015-03-17 13:13:03 -07:00
Satnam Singh 0212e68e91 Adjust POS file for Fluentd to Cloud Logging 2015-03-17 11:44:04 -07:00
Timothy St. Clair 140330fca8 Fix to address issue #5461 kube-proxy too many files open 
https://github.com/GoogleCloudPlatform/kubernetes/issues/5461
 2015-03-17 10:13:25 -05:00
Satnam Singh d395efa463 Make Fluentd to Cloud Logging collector read from head 2015-03-16 16:19:07 -07:00
Derek Carr 440c720805 Merge pull request #5330 from zmerlynn/try_addons_again 
Retry object creation with --validate in kube-addons
 2015-03-16 10:43:39 -04:00
Victor Marmol 1a7f7245e7 Remove cAdvisor manifest from cluster startup. 
cAdvisor is now integrated into the Kubelet and runs inside of it.
 2015-03-13 16:06:42 -07:00
Brendan Burns 966e8301a0 Add kubelet to the master. 2015-03-13 12:09:24 -07:00
Satnam Singh e16bcceceb Merge pull request #5390 from brendandburns/shell_sucks 
Configure docker on the master like we do on workers.
 2015-03-12 11:42:03 -07:00
Brendan Burns 17ff8fb421 Configure docker on the master like we do on workers. 2015-03-12 10:37:30 -07:00
Saad Ali 7629b616a7 Merge pull request #5348 from dchen1107/docker 
Check docker unixsocket periodically through monit. If failed, restart d...
 2015-03-11 17:42:58 -07:00
Dawn Chen fe9e0ca5ce Check docker unixsocket periodically through monit. If failed, restart docker. 2015-03-11 17:20:03 -07:00
Zach Loafman 018b11bc3a Retry object creation with --validate in kube-addons 
The better solution is some fence with Salt, but the actual logs
provided in the bug don't support any race condition here, plus the
ordering in the Salt configuration seems correct.

We haven't seen this again in a while, but given the results of the
situation (a borked cluster), I'm proposing a relatively simple
workaround.

Fixes #4357 (dubiously)
 2015-03-11 14:40:37 -07:00
derekwaynecarr f71a91a131 Fix regression introduced when moving to standalone salt 2015-03-11 11:06:00 -04:00
derekwaynecarr 468bf1da75 Enable common set of admission controllers across salt providers 2015-03-11 11:06:00 -04:00
Brian Grant b3d3db08ef Merge pull request #5064 from justinsb/ignore_stopped_aws_instances 
Ignore EC2 instances that are stopped
 2015-03-10 11:03:34 -07:00
Alex Robinson 6700a7f230 Merge pull request #5253 from satnam6502/fluentd 
Modify Fluentd to ES setup to pass flags
 2015-03-10 10:46:26 -07:00
Brendan Burns 00f3779c6f Merge pull request #5119 from zmerlynn/salt_drones 
Change GCE to use standalone Saltstack config
 2015-03-10 10:29:59 -07:00
Satnam Singh 3ce0ede78e Modify Fluentd to ES setup to pass flags 2015-03-10 10:11:26 -07:00
Alex Robinson 8fd33476f0 Merge pull request #5218 from satnam6502/fluentd-es 
Change Fluentd config to read from head of Docker container logs
 2015-03-10 09:57:02 -07:00
Zach Loafman 120dba474e Change GCE to use standalone Saltstack config: 
Change provisioning to pass all variables to both master and node. Run
Salt in a masterless setup on all nodes ala
http://docs.saltstack.com/en/latest/topics/tutorials/quickstart.html,
which involves ensuring Salt daemon is NOT running after install. Kill
Salt master install. And fix push to actually work in this new flow.

As part of this, the GCE Salt config no longer has access to the Salt
mine, which is primarily obnoxious for two reasons: - The minions
can't use Salt to see the master: this is easily fixed by static
config. - The master can't see the list of all the minions: this is
fixed temporarily by static config in util.sh, but later, by other
means (see
https://github.com/GoogleCloudPlatform/kubernetes/issues/156, which
should eventually remove this direction).

As part of it, flatten all of cluster/gce/templates/* into
configure-vm.sh, using a single, separate piece of YAML to drive the
environment variables, rather than constantly rewriting the startup
script.
 2015-03-10 09:04:29 -07:00
derekwaynecarr 22e559d769 Provision network before kubelet 2015-03-10 10:30:50 -04:00
Satnam Singh 2486de1287 Change Fluentd config to read from head of Docker container logs 2015-03-09 23:25:52 -07:00
Justin Santa Barbara a3ae480acc Ignore EC2 instances that are stopped 
Otherwise we pick up previous cluster instances (in EC2, stopped instances hang around for
a while - maybe 30 minutes?)
 2015-03-09 19:03:46 -07:00
Dawn Chen 7a47ea8913 Ask monit health check kubelet; if failed, restart. 2015-03-05 17:29:38 -08:00
derekwaynecarr 35ab4ccc6d Further vagrant reliability improvements. 2015-03-03 06:55:34 -08:00
Tim Hockin d46146da7d Stop using Docker's -r flag 
It has been deprecated since 1.3, replaced by restart policies.  We are
not supporting docker < 1.3 any more.
 2015-02-28 13:36:42 -08:00
Satnam Singh 16f49f6d09 Update Fluentd to Cloud Logging image 2015-02-26 15:18:22 -08:00
Eric Tune e77745672c Merge pull request #4577 from satnam6502/cluster-name 
Qualify the name of forwarding rules with INSTANCE_PREFIX
 2015-02-26 13:17:49 -08:00
Daniel Smith 3038a5624e Merge pull request #4859 from dchen1107/master 
Move salt based clusters to Docker 1.5.0
 2015-02-26 11:32:07 -08:00
Dawn Chen 59a9030413 Move salt based clusters to Docker 1.5.0 2015-02-26 11:06:53 -08:00
Satnam Singh 19b927ea57 Name a cluster and use it to make forwarding rules for GCE 2015-02-23 17:04:33 -08:00
derekwaynecarr 87a41b0934 Improve vagrant reliablility, fix race condition with openvswitch and docker 2015-02-21 13:31:50 -05:00
Victor Marmol ff7726d1ef Merge pull request #4655 from saad-ali/fix2884quick 
Add monit process monitoring for kubelet and docker
 2015-02-20 11:07:06 -08:00
David Oppenheimer ba749284d0 Have etcd listen on all interfaces so that monit probes succeed. Closes #3852. 2015-02-19 23:14:14 -08:00
saadali 4dea0488e6 Add monit process monitoring for kubelet and docker 2015-02-19 22:22:05 -08:00
Brendan Burns e4e68e2fa5 Merge pull request #4138 from justinsb/fix_duplicate_master_arg 
Remove duplicate --master arg from kube-scheduler
 2015-02-19 11:58:23 -08:00
David Oppenheimer c99a37d3a3 Merge pull request #4587 from gmarek/client2 
Fix monit config for etcd
 2015-02-19 11:37:47 -08:00
gmarek 19b7c01084 Fix monit config for etcd 2015-02-19 13:11:43 +01:00
Brendan Burns 664de16c81 Remove kubecfg from the client tools we attempt to distribute. 2015-02-18 22:26:26 -08:00
Dawn Chen 23b8b5ea6a Configure docker with larger open files. 2015-02-11 23:52:58 -08:00
Eric Tune 59a06b5073 Don't set --etcd_servers argument for kubelet. 
This is ignored by all recent kubelet versions when api_servers is
set.  And it is set in this salt file.
 2015-02-11 15:53:26 -08:00
Zach Loafman 6b3bb763f0 Merge pull request #4287 from derekwaynecarr/docker_group_gone 
Docker 1.4.1 on Fedora no longer provisions a docker group
 2015-02-10 14:28:02 -08:00
Zach Loafman 2fb4bd4724 Merge pull request #4286 from brendandburns/etcd 
Add restart to the kube apiserver.
 2015-02-10 12:09:34 -08:00
derekwaynecarr 46eb657722 docker does not provision a docker group anymore, so do not add vagrant user to it 2015-02-10 11:15:22 -05:00
Zach Loafman 2f231e0a2a Merge pull request #4268 from rajatchopra/master 
Fix vagrant networking round 2
 2015-02-10 07:39:48 -08:00
Brendan Burns c50e650d8b Add restart to the kube apiserver. 2015-02-10 07:35:00 -08:00
Rajat Chopra 1d12cad456 Fix vagrant networking. Include the master as part of overlay. And remove STP for efficiency. 
fix e2e services for vagrant
Kubelet should wait on SDN to finish
Do not require docker on master for default clouds
 2015-02-09 14:37:27 -08:00
Ravi Sankar Penta 3a5ef05f64 Make periodic sync nodes from cloud provider optional. 
--sync_nodes=false gives user flexibility to add/remove nodes in the
cluster using REST api/kubectl cli and at the same time can use
cloud provider for other resources like persistent disks, etc.
 2015-02-09 13:42:44 -08:00
Zach Loafman 6b8c332cd3 Revert "Fix vagrant networking" 2015-02-09 13:40:11 -08:00
Rajat Chopra 9f10108758 Kubelet should wait on SDN to finish 2015-02-09 11:22:14 -08:00
Rajat Chopra b61ea1bbf1 Fix vagrant networking. Include the master as part of overlay. And remove STP for efficiency. 2015-02-05 17:41:27 -08:00
Justin Santa Barbara e72d3392cc Remove duplicate --master arg from kube-scheduler 2015-02-04 19:05:05 -05:00
Justin Santa Barbara 7790435915 Fix for bool vs string comparisons in salt templates 
Fixes #4118
 2015-02-04 14:09:10 -05:00
Justin SB 9a29d9f92a Etcd data dir must be empty 
Otherwise we get the error "unknown wal version in data dir /var/etcd"

When this is the homedir for etcd, e.g. .bashrc confuses it
 2015-02-02 08:09:32 -08:00
derekwaynecarr 4dd50a18c3 Fix vagrant regression, add flag to easily enable v1beta3 2015-01-30 12:16:24 -05:00
Satnam Singh db416c4f39 Merge pull request #3970 from justinsb/fix_aws 
aws cluster/kube-up.sh fixes
 2015-01-30 07:56:01 -08:00
derekwaynecarr 3cb015c09c Missing boilerplate 2015-01-30 00:22:17 -05:00
derekwaynecarr aef084bf9d Various vagrant fixes, etcd 2.0 2015-01-30 00:15:34 -05:00
Brendan Burns a20ca78fb3 Update etcd to version 2.0.0 2015-01-29 17:53:53 -08:00
Justin Santa Barbara eb1d9d0a46 Add missing = to --machines arg 2015-01-29 20:35:36 -05:00
Justin Santa Barbara 5d864b006a Add aws to list of clouds that use tokens 2015-01-29 20:35:36 -05:00
Justin Santa Barbara 51225c76da Apply jbeda's conditional tar fix 2015-01-29 20:35:36 -05:00
Justin Santa Barbara 2f195578cc Create /srv/kuberentes/known_tokens.csv on AWS 2015-01-29 20:35:35 -05:00
Justin Santa Barbara bb0431dce8 Pass cloud & cloud_config to kube-apiserver for AWS 2015-01-29 20:35:35 -05:00
Brendan Burns c7c11d7731 Add monit support for restarting etcd to the master. 2015-01-27 20:15:24 -08:00
Victor Marmol 94988aece2 Update cAdvisor to 0.8.0. 2015-01-23 11:11:15 -08:00
Zach Loafman a305269e18 Deferred creation of SkyDNS, monitoring and logging objects 
This implements phase 1 of the proposal in #3579, moving the creation
of the pods, RCs, and services to the master after the apiserver is
available.

This is such a wide commit because our existing initial config story
is special:

* Add kube-addons service and associated salt configuration:
** We configure /etc/kubernetes/addons to be a directory of objects
that are appropriately configured for the current cluster.
** "/etc/init.d/kube-addons start" slurps up everything in that dir.
(Most of the difficult is the business logic in salt around getting
that directory built at all.)
** We cheat and overlay cluster/addons into saltbase/salt/kube-addons
as config files for the kube-addons meta-service.
* Change .yaml.in files to salt templates
* Rename {setup,teardown}-{monitoring,logging} to
{setup,teardown}-{monitoring,logging}-firewall to properly reflect
their real purpose now (the purpose of these functions is now ONLY to
bring up the firewall rules, and possibly to relay the IP to the user).
* Rework GCE {setup,teardown}-{monitoring,logging}-firewall: Both
functions were improperly configuring global rules, yet used
lifecycles tied to the cluster. Use $NODE_INSTANCE_PREFIX with the
rule. The logging rule needed a $NETWORK specifier. The monitoring
rule tried gcloud describe first, but given the instancing, this feels
like a waste of time now.
* Plumb ENABLE_CLUSTER_MONITORING, ENABLE_CLUSTER_LOGGING,
ELASTICSEARCH_LOGGING_REPLICAS and DNS_REPLICAS down to the master,
since these are needed there now.

(Desperately want just a yaml or json file we can share between
providers that has all this crap. Maybe #3525 is an answer?)

Huge caveats: I've gone pretty firm testing on GCE, including
twiddling the env variables and making sure the objects I expect to
come up, come up. I've tested that it doesn't break GKE bringup
somehow. But I haven't had a chance to test the other providers.
 2015-01-21 12:25:50 -08:00
Joe Beda 7ee327146a Fix typo in initd scripts 2015-01-15 09:12:20 -08:00
Joe Beda 228659a177 Fix up for posix flags 2015-01-15 09:12:19 -08:00
Satnam Singh 295bd3768d Launch Elasticsearch and Kibana automatically 2015-01-09 13:41:46 -08:00
Eric Tune c2b3d678c0 Merge pull request #3248 from derekwaynecarr/admission_control_hooks 
Implement basic admission control framework
 2015-01-07 16:52:49 -08:00
Dawn Chen 6cb7f0451c Using google/cadvisor:0.7.1 instead of :latest 2015-01-07 14:14:20 -08:00
derekwaynecarr 520ae3ef27 Implement basic admission control framework 2015-01-07 14:42:31 -05:00
Satnam Singh 8ab2e9c61a Use DNS for Elasticsearch 2015-01-06 07:33:48 -08:00
Dawn Chen 86aa38547c Remove duplicate volumes type. 2014-12-30 14:40:16 -08:00
Tim Hockin 59164ca844 Make DNS an option for cluster turnup 2014-12-29 09:18:12 -08:00
Vishnu Kannan 2d5432bc91 update cadvisor static manifest file completely. 2014-12-25 18:22:59 +00:00
Vishnu Kannan 9a4476a60c Update cadvisor manifest to bind mount "/sys" from global namespace. 2014-12-22 23:49:22 +00:00
saadali ec6d4c6078 Fix kube-controller-manager failing to start because "Jinja variable 'cloud_config' is undefined" 2014-12-19 17:20:22 -08:00
Brendan Burns 2369f0d50d Update aws. 2014-12-19 14:16:57 -08:00
Brendan Burns bd712b61e6 Merge branch 'master' of https://github.com/dutchcoders/kubernetes into dutchcoders-master 
Conflicts:
	cluster/aws/config-default.sh
	cluster/saltbase/salt/kube-controller-manager/default
 2014-12-19 11:45:46 -08:00
Jordan Liggitt 2de478f16c Make cert CN unique to avoid certificate serial number clashes 2014-12-18 15:13:32 -05:00
Jernej Kos 430a04af45 Added missing Salt requisites. 2014-12-18 10:29:39 +01:00
Dawn Chen c053ec63c5 Move salt based clusters to Docker 1.4.1 2014-12-16 15:54:26 -08:00
Joe Beda 911a437c1b Make the test for the 'docker_opts' grain more robust. 
Fixes #2982
 2014-12-16 14:26:36 -08:00
Dawn Chen 04b136a7db Move salt based clusters to Docker 1.4.0 2014-12-16 11:58:45 -08:00
Joe Beda 056fdbb9b8 Merge pull request #2943 from brendandburns/flags 
Add a log level flag.
 2014-12-15 12:59:08 -08:00
Brendan Burns 878d85df0c Add a log level flag. 2014-12-15 12:00:29 -08:00
Satnam Singh b4fcdd7e47 Collect Kubelet logs with Fluentd->Elasticsearch and Fluentd->Cloud Logging 2014-12-14 01:47:48 -08:00
Joe Beda c9acafc98a Move salt based clusters to Docker 1.3.3 
Part of #2896
 2014-12-12 12:18:07 -08:00
Brendan Burns 149881e13e Actually set repository QPS in the kubelet. 2014-12-11 12:06:40 -08:00
Joe Beda 89011f26f4 remove SSLv3 support in nginx config 2014-12-10 13:17:02 -08:00
Eric Tune edeb96dca9 Merge pull request #2772 from rlarocque2/apiserver_auth 
Implement apiserver communication for Vagrant
 2014-12-10 07:13:40 -08:00
Brendan Burns 3a7dd589d5 Merge pull request #2735 from erictune/proxy_no_etcd 
Use apiserver instead of etcd for kube-proxy.
 2014-12-09 15:27:20 -08:00
Jeff Lowdermlk 0007257bce Fix conditional on grains.docker_opts in docker-defaults 2014-12-08 16:42:41 -08:00
Richard Larocque 7d37b24cc7 Implement apiserver communication for Vagrant 
This implementation is based on the GCE impementation from
618a367dbb48dbaa9c3b50e877858d854322fd0b.
 2014-12-05 12:49:51 -08:00
Eric Tune 65b38e7af7 Use apiserver instead of etcd for kube-proxy. 
For cloud-providers that use salt, have kube-proxy use
apiserver instead of etcd as a config source.
 2014-12-02 22:21:01 -08:00
Brendan Burns b07515c5ea Add some backward compatability for GKE. 2014-12-02 11:32:45 -08:00
Alen Komljen d83d269c27 Remove empty lines when file is generated from jinja template 2014-12-02 15:00:58 +01:00
Joe Beda f8505cd286 For GCE, allow insecure registries anywhere in 10.0.0.0/8. 2014-12-01 11:36:08 -08:00
Alen Komljen 249bbd7d9a Remove unused nsinit salt state 2014-11-27 09:59:25 +01:00
Remco 83f3982eee fix(aws): cloud provider wasn't set, corrected cloud config syntax 2014-11-26 17:30:35 +01:00
Tim Hockin 47141f05c7 standard config opt names: logging 2014-11-25 17:41:53 -08:00
Tim Hockin 52ad94d766 standard config opt names: node monitoring 2014-11-25 17:39:28 -08:00
Joe Beda 4a5a69b520 Upgrade docker to 1.3.2. 2014-11-25 14:46:25 -08:00
Joe Beda 2b712a744a Fix up salt ordering dependencies in kube-apiserver/init.sls. 2014-11-25 14:46:25 -08:00
Joe Beda 802d214ed0 Merge pull request #2461 from MSOpenTech/deploy 
Bring Azure deploy scripts up to date
 2014-11-20 17:15:44 -08:00
Joe Beda 658c5cdaaa Check in default blank file for kubelet/kubernetes_auth 
This lets us eliminate some per-cloud stuff in our salt.  This is an alternate fix for #2475.
 2014-11-20 17:01:28 -08:00
derekwaynecarr 1a894741b2 Fix vagrant regression 2014-11-20 10:39:09 -05:00
Brendan Burns 927088de83 Fix etcd unpack by adding a explicit x flag. 2014-11-19 11:33:50 -08:00
Jeff Mendoza 9934a0ace4 Update cert generation for Azure. 2014-11-18 14:10:50 -08:00
Eric Tune 67adf6473e Don't reference non-existent token file if not GCE. 2014-11-18 00:23:41 -08:00
Eric Tune 46dcacfa93 Kubelet talks securely to apiserver. 
Configure apiserver to serve Securely on port 6443.
Generate token for kubelets during master VM startup.
Put token into file apiserver can get and another file the kubelets can get.
Added e2e test.
 2014-11-18 00:23:41 -08:00
Brendan Burns 5122e4cc3a Remove deprecated nsinit from the node definition. 2014-11-17 22:20:57 -08:00
Abhishek Gupta 65a4041d26 Fixing systemd service files to point to correct config/executors 2014-11-17 12:24:10 -08:00
Joe Beda 66d287f7e4 Merge branch 'rename-to-kube' of https://github.com/eparis/kubernetes into eparis-rename-to-kube 
* 'rename-to-kube' of https://github.com/eparis/kubernetes:
  rename kube server binaries to kube-

Conflicts:
	docs/salt.md
 2014-11-17 09:52:10 -08:00
Joe Beda 06d838ffb1 Merge pull request #2369 from jbeda/vagrant-fix 
Vagrant fixes
 2014-11-14 16:43:06 -08:00
Brendan Burns c2485a4056 Merge pull request #2147 from justinsb/ipv6 
Initial ipv6 / iptables work
 2014-11-14 10:34:37 -08:00
Joe Beda 5a0159ea00 Get Vagrant to start using TLS client certs. 
Also fix up cert generation. It was failing during the first salt highstate when trying to chown the certs as the apiserver user didn't exist yet.  Fix this by creating a 'kube-cert' group and chgrping the files to that.  Then make the apiserver a member of that group.

Fixes #2365
Fixes #2368
 2014-11-13 22:16:45 -08:00
Eric Paris a99c3c7963 rename kube server binaries to kube- 
apiserver becomes kube-apiserver
controller-manager -> kube-controller-manager
scheduler and proxy similarly.

Only thing I promise is that right now hack/build-go.sh and
build/release.sh exit with 0.  That's it.  Who knows if any of this
actually works....
 2014-11-13 20:08:26 -05:00
Richard Larocque 63d6e7cd9f make-cert.sh: Create directory before writing cert 
Adds a "mkdir -p" to the make-cert.sh.  This fixes an issue where the
script could fail if /srv/kubernetes did not exist previously.
 2014-11-13 16:39:09 -08:00
Eric Tune 19ec2234f9 Make server certs accessible to apiserver user. 2014-11-13 13:02:48 -08:00
Joe Beda ee2f030623 Give the API server access to TLS certs. 
Moved the cert generation to a separate salt state and put it in a more appropriate sharable location (`/srv/kubernetes/`).
 2014-11-12 18:14:24 -08:00
Brendan Burns 62bd634c07 Update the AWS config. 2014-11-10 13:33:43 -08:00
Brendan Burns bdc4ab62c7 Merge pull request #2245 from jbeda/gcs-deps 
Start pulling external dependencies from GCS where possible
 2014-11-10 11:00:41 -08:00
bgrant0607 6c70227a2e Merge pull request #2249 from satnam6502/salt 
Make Docker container dirs writable for pos file
 2014-11-08 11:56:48 -08:00
Satnam Singh 9fac9949b2 Make Docker container dirs writable for pos file 2014-11-07 17:42:48 -08:00
Joe Beda a839599883 Install client tools on master. 
We now include kubecfg and kubectl in "server binary" tar and place them into /usr/local/bin on the master.
 2014-11-07 17:13:18 -08:00
Joe Beda d033dfcc2b Load etcd tarball from GCS. 
Also fix up comment about how to cache docker debs.
 2014-11-07 13:43:07 -08:00
Justin SB ddaa716de1 Allow specification of docker daemon args 2014-11-04 14:12:59 -08:00
Justin SB 8e70a664e8 Extend container_bridge.py (salt lib) to support IPv6 2014-11-04 06:33:34 -08:00
Brendan Burns f44bb9d673 Merge pull request #2104 from jbeda/gcs-docker-install 
Get docker deb from GCS
 2014-10-31 21:15:06 -07:00
Joe Beda 29455398e6 Get docker deb from GCS. 
Also lock to 1.3.0 due to issues in 1.3.1 (docker/docker#8889)
 2014-10-31 16:27:00 -07:00
Deyuan Deng 019b7fc74c Separate minion controller from master. 2014-10-30 20:24:15 -04:00
Pieter Noordhuis 918c8e2b96 Limit GCE specific mods to GCE 2014-10-29 12:39:07 -07:00
Pieter Noordhuis 1152471db7 Split reactor files 
There now are reactor files to run highstate on:

* New nodes only
* Master nodes
* Minion nodes
 2014-10-29 11:30:09 -07:00
Pieter Noordhuis e90f98a52c Make vSphere scripts work for binary deploys 
Fixes #1732.
 2014-10-29 11:30:08 -07:00
Brendan Burns c4d6f3a1cd Fix the etcd md5 since they uploaded a new binary. 2014-10-29 11:18:37 -07:00
Joe Beda 8100c063b8 Turn on auto security updates for debian based systems. 
Fixes #2008
 2014-10-27 13:49:51 -07:00
Joe Beda 25b1eea116 Merge pull request #1832 from doublerr/rackspace_switch_to_coreos 
Rackspace - Switch to CoreOS for standard cluster
 2014-10-22 09:30:55 -07:00
derekwaynecarr b836e31cf4 Fixup docker configuration for sdn 2014-10-22 11:32:38 -04:00
Satnam Singh 39212f5e2e Flunentd to GCP logging node level configuration 2014-10-20 18:05:00 -07:00
Satnam Singh b7722a631d Adjust test in salt file to use boolean value 2014-10-20 15:52:28 -07:00
derekwaynecarr 0d86be680b Do not require a writeable /vagrant 2014-10-20 16:59:15 -04:00
Ryan Richard a8bae68865 Rackspace - Switch to CoreOS for standard cluster 
The Rackspace setup for Kubernetes now leverages CoreOS instead of Ubuntu. We've dropped Salt completely for our cluster.
 2014-10-20 12:14:20 -05:00
Dawn Chen baf3bea79a Set --ip-masq to false 2014-10-16 12:17:18 -07:00
Tim Hockin e907011111 Core support for ip-per-service 2014-10-16 08:36:47 -07:00
Joe Beda 9bb20b41ee Break out dynamic salt files to enable clean kube-push. 
This lets us blow away salt files and replace them with a new version while keeping a tree of "overlay" files that are cluster specific and generated at cluster up time.

Fixes #1783
 2014-10-14 15:01:43 -07:00
Joe Beda 96c1bc17ca Set shell options for reliability. 
Tweak a few other small things in our shell scripts.
 2014-10-10 12:33:36 -07:00
Joe Beda d43a6ec5a3 Standardize how we refer to the kubernetes root. 
Now use $KUBE_ROOT as the variable pretty much everywhere.
 2014-10-10 12:33:36 -07:00
Joe Beda 881cf80182 Vagrant now using pre-built binaries. 2014-10-10 12:30:12 -07:00
Joe Beda 15cd6f07d6 Use binary releases for cluster push scripts. 
This is for GCE right now.  Other clouds/clusters are probably broken.
 2014-10-10 12:30:11 -07:00
Joe Beda 9a8807b033 get.docker.com is now the preferred place to get Docker 
Update salt configs
 2014-10-09 12:31:53 -07:00
Joe Beda 0de2ef496e Merge pull request #1595 from thockin/flageq 
Use '=' for flags or else parsing gets confused sometimes
 2014-10-06 13:14:35 -07:00
Tim Hockin 440d14c239 Use '=' for flags or else parsing gets confused sometimes 2014-10-06 12:25:56 -07:00
Deyuan Deng 253907ab04 Add etcdctl in master 2014-10-04 11:26:01 -04:00
Brendan Burns 77d13db307 Fix the vagrant build. 2014-09-26 14:02:56 -07:00
Joe Beda b3e7fd4209 Be more robust in how we pass flags. 
Using the '=' form of a flag makes things better if we have an empty list of machines.
 2014-09-25 11:52:18 -07:00
Brendan Burns cafd20b233 Complete the mitm prevention on GCE. 2014-09-23 16:51:04 -07:00
Brendan Burns d38b498b0f Add a script for generating CA signed certs, and a client cert. 2014-09-23 11:38:29 -07:00
Brendan Burns bb829ce8f6 fix salt. 2014-09-17 17:13:41 -07:00
Brendan Burns ffa4500e09 Add a salt flag for enabling privileged containers. 2014-09-17 14:25:28 -07:00
brendandburns c74060d465 Merge pull request #1282 from derekwaynecarr/add_vagrant_to_docker_group 
Add vagrant user to docker group
 2014-09-15 17:07:02 -07:00
Brendan Burns 61e90fdaff Fix the logrotate cron job. 2014-09-11 20:56:31 -07:00
Brendan Burns 76c0ce374d Start log rolling for docker containers too. Fix logrotate for other daemons. 2014-09-11 15:47:06 -07:00
derekwaynecarr 63bd987561 Add vagrant user to docker group 2014-09-11 16:48:17 -04:00
Brendan Burns 686c3bec84 Add initial logrotate support. 2014-09-09 13:35:06 -07:00
derekwaynecarr f42fcef620 Add explicit flag to use openvswitch 2014-09-08 15:31:22 -04:00
Brendan Burns e8afebf1e1 Update nginx to support websockets and not buffer watches. 2014-09-05 21:46:01 -07:00
Brendan Burns 4a934427db Exclude vendored etcd. 2014-09-04 12:05:34 -07:00
Brendan Burns 3864cf453a Fix travis and cluster installs. 2014-09-04 10:31:09 -07:00
Joe Beda 847c87faa8 Fix docker start after minion reboot on GCE/debian 
Uncomment docker service state as it should no longer be affected by race bug in docker init.d start script.  Also make sure that the docker service state depends on the cbr state.

Fixes #802
 2014-09-03 11:11:28 -07:00
Ryan Richard 0b50244c72 workaround lack of jinja2.7 support 2014-09-02 19:05:56 -05:00
Joe Beda 2ac0770694 Make salt configs work with Jinja 2.6. 
Fixes #1089
 2014-08-28 16:54:14 -07:00
Ryan Richard 2565164a8d Add Rackspace support for dev-build-and-up 2014-08-28 10:36:58 -05:00
Clayton Coleman f5a835b067 Merge pull request #978 from rajatchopra/podnet 
Add IP per pod with Open vSwitch in Vagrant
 2014-08-26 19:29:42 -04:00
Joe Beda 5722eba780 Propagate version info into salt config and use to build. 
Fixes #1043.
 2014-08-26 16:17:06 -07:00
Rajat Chopra 2dd57898d4 add ip per pod across vagrant minions 2014-08-26 11:29:35 -07:00
Joe Beda 831a0d7e50 Merge pull request #1027 from MSOpenTech/azure-scripts 
Add salt configuration for openvpn for use on Azure.
 2014-08-26 11:04:03 -07:00