github/kubernetes
4
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
15,882 Commits 1 Branch 31 Tags 1,019 MiB
72db123025
Commit Graph

27 Commits

Author SHA1 Message Date
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
gmarek
5bb7eec5ab Use salt pillars to pass test arguments during cluster creation. 2015-07-31 09:32:01 +02:00
Piotr Szczesniak
f48543aba5 Made enabling Kube UI configurable 2015-07-27 08:23:04 +02:00
Chris Hiestand
db1a73317c change shell var MINION_SCOPES to user setable 
Because bash arrays may not be environment variables
 (see: <https://stackoverflow.com/questions/5564418/exporting-an-array-in-bash-script>)
 we have to change the MINION_SCOPES array to a string
* prefers aliases instead of full URLs for scopes

tested under GKE, needs GCE testing

Close #10458
 2015-07-22 14:34:03 -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
Eric Paris
58df58f3d7 Remove unused enable_node_monitoring option 
Back in 1a7f7245e7 we dropped the one
place this was used, but left all of the variable and definitions and
garbage around cluster/
 2015-06-25 20:57:56 -04:00
Tim Hockin
280f99afd9 EOL our registry caching mirror 
Docker's v1 registry has gotten slower and slower, and they have no
interest in fixing it.  Using a mirror forces v1 mode.  Measurements
show that v1 with our mirror is slower than v2 with docker's registry in
just about all metrics.
 2015-06-24 09:56:59 -07:00
Zach Loafman
d8da39ecd0 Validate binaries downloaded from GCS: 
* Set SHA1 for Kubernetes server binary and Salt tar in kube-env.
* Check SHA1 in configure-vm.sh. If the env variable isn't available,
download the SHA1 from GCS and double check that.
* Fixes a bug in the devel path where we were actually uploading the
wrong sha1 to the bucket.

Fixes #10021
 2015-06-18 13:37:05 -07:00
Brendan Burns
2c59a3c4a4 Fix the scripts to default empty string if EXTRA_DOCKER_OPTS isn't set. 2015-06-12 21:47:30 -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
Prashanth Balasubramanian
f7c0f1c1e3 Set min-request-timeout in test clusters 2015-06-03 08:46:28 -07:00
Brian Grant
3da686fea5 Merge pull request #8894 from cjcullen/kmaster2 
Add an explicit variable to indicate whether an instance is master or not
 2015-06-02 20:37:43 -07:00
Brian Grant
21147d784a Merge pull request #9074 from roberthbailey/unbound-var 
Don't assume that CA_CERT_BASE64 will be set.
 2015-06-02 18:52:49 -07:00
CJ Cullen
dcf5b16cea Add an explicit KUBERNETES_MASTER variable to the kube-env. 2015-06-02 17:17:02 -07:00
Robert Bailey
0c62b71f8a Don't assume that CA_CERT_BASE64 will be set. 2015-06-01 11:35:26 -07:00
Tim Hockin
ac3cc3c518 Rename PORTAL_NET all over 2015-05-28 16:10:44 -07:00
Justin Santa Barbara
521cb0e85b Pass arg lists to gcloud as comma-separated single arg 
This should get rid of the scary warning messages on GCE up.

Fixes #8117
 2015-05-28 00:36:13 -04:00
Max Forbes
3437ac691a Rolling node upgrade 2015-05-19 11:42:34 -07:00
Robert Bailey
9ab41db7ea Static cert distribution for GCE. 
To make cert validation work, no longer use the
fqdn for the master name on the node VMs.
 2015-05-14 11:59:13 -07:00
Max Forbes
0acf8f9a00 Refactor GCE kube-env for both OSes. 2015-05-12 13:26:10 -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
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
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
Dawn Chen
13a0b033e2 Bring up a cluster using coreos image for worker nodes. 2015-04-28 16:13:26 -07:00
Dawn Chen
5fa11322f8 Factory out debian e.g. ContainerVM image specific support to its own 
helper utility library.
 2015-04-28 16:07:57 -07:00