This change support running kubernetes master on Ubuntu Trusty.
It uses pure cloud-config and shell scripts, and completely gets
rid of saltstack or the release salt tarball.
To fix it, I just add openssl depedency on "generate-cert" state. It
should work on Debian-like and RedHat-Like systems. (and, Archlinux,
Opensuse, etc)
Fixed error :
$ sudo salt 'kubernetes-master' state.apply
----------
ID: kubernetes-cert
Function: cmd.script
Result: False
Comment: Command 'kubernetes-cert' run
Started: 06:57:06.634203
Duration: 208.719 ms
Changes:
----------
pid:
793
retcode:
1
stderr:
/tmpm24T3R.sh: line 22: openssl: command not found
chgrp: cannot access '/srv/kubernetes/server.key': No such file or directory
chgrp: cannot access '/srv/kubernetes/server.cert': No such file or directory
chmod: cannot access '/srv/kubernetes/server.key': No such file or directory
chmod: cannot access '/srv/kubernetes/server.cert': No such file or directory
stdout:
After applying my patch (success) :
----------
ID: kubernetes-cert
Function: cmd.script
Result: True
Comment: Command 'kubernetes-cert' run
Started: 07:17:04.172384
Duration: 1041.092 ms
Changes:
----------
pid:
1045
retcode:
0
stderr:
Generating a 4096 bit RSA private key
......................................................................++
...............................................................................++
writing new private key to '/srv/kubernetes/server.key'
-----
stdout:
----------
The version of Salt we're running doesn't do a good job of detecting
systemd. Inspired by https://github.com/saltstack/salt/issues/13926,
I added a provider-force to the services.
With this change, salt-call -l debug state.highstate succeeds, even for
repeated invocations.
The issue was (probably) benign, but definitely caused noised (e.g. #11297)
I got the package name wrong before, which meant that salt was failing
on invocations after the first (the name apparently doesn't matter on
the first invocation).
We've had a lot of salt problems with systemd on AWS; we have a
workaround in place that we use everywhere else, we should use that for
kube-node-unpacker too.
Fixes#19386
Issue #19388
To build the python image, BUILD_PYTHON_IMAGE should be set during make.
When the addon script is running, it will check if python is installed
on the machine, if not, it will use the python image that built previously.
When KUBE_E2E_STORAGE_TEST_ENVIRONMENT is set to 'true', kube-up.sh script
will:
- Install the right packages for all storage volumes.
- Use devicemapper as docker storage backend. 'aufs', the default one on
Debian, does not support extended attibutes required by Ceph RBD and Gluster
server containers.
Tested on GCE and Vagrant, e2e tests for storage volumes passes without any
additional configuration.
This ensures nfs-common is installed on GCE, and provides a more
functional explanation/example. I launched two replication controllers
so that there were busybox pods to poke around at the NFS volume, and
so that the later wget actually works (the original example would have
to work on the node, or need some other access to the container
network). After switching to two controllers, it actually makes more
sense to use PV claims, and it's probably a configuration that makes
more sense for indirection for NFS anyways.
We want to match the version of netcat that is installed on GCE. We
were having problems with netcat-openbsd having slightly different
timeout behaviour (on UDP packets; when there was no listener).
This scopes down the initially ambitious PR:
https://github.com/kubernetes/kubernetes/pull/14960 to replace just
`pause` and `fluentd-elasticsearch` to come through `beta.gcr.io`.
The v2 versions have been pushed under new tags, `pause:2.0` and
`fluentd-elastisearch:1.12`.
NOTE: `beta.gcr.io` will still serve images using v1 until they are repushed with v2. Pulls through `gcr.io` will still work after pushing through `beta.gcr.io`, but will be served over v1 (via compat logic).
OpenContrail is an open-source based networking software which provides virtualization support for the cloud.
This change-set adds ability to install and provision opencontrail software for networking in kubernetes based cloud environment.
There are basically 3 components
o kube-network-manager -- plugin between contrail components and kubernets components
o provision_master.sh -- OpenContrail software installer and provisioner in master node
o provision_minion.sh -- OpenContrail software installer and provisioner in minion node(s)
These are driven via salt configuration files
One can provision opencontrail by just setting "export NETWORK_PROVIDER=opencontrail"
Optionally, OPENCONTRAIL_TAG, and OPENCONTRAIL_KUBERNETES_TAG can be used to
specify opencontrail and contrail-kubernetes software versions to install and provision.
Public-IP Subnet provided by contrail can be configured via OPENCONTRAIL_PUBLIC_SUBNET
environment variable
At this moment, plan is to add support for aws, gce and vagrant based platforms
For more information on contrail-kubernetes, please visit https://github.com/juniper/contrail-kubernetes For more information on opencontrail, please visit http://www.opencontrail.org
changes to fluent-plugin-google-cloud to attach Kubernetes metadata to
logs.
Along with this, separate logs from containers in the cluster out from
logs from the daemons running on the node by instantiating two instances
of the output plugin, one which uses the new metadata (for containers)
and one which doesn't (for things like docker and the kubelet).
* Using Fedora 21 as the base box
* Discover the active network interfaces in the box to avoid hardcoding
them in configuration.
* Use the master IP for the certificate.
gunk when installing the google-fluentd agent.
Also let it log things by not redirecting to a file within the container
and only using -q (warning logs only) rather than -qq (error logs only).
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.
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.
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/
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.
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
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.
Justin Santa Barbara