github/kubernetes
1
0
Fork 0
mirror of https://github.com/k3s-io/kubernetes.git synced 2025-09-25 20:29:56 +00:00
Code Issues Projects Releases Wiki Activity

Examples/rethinkdb fixes - namespaces, api update, curl params

Browse Source
This commit is contained in:
Marcin Wielgus
2015-07-07 15:27:49 +02:00
parent bb6f2f7ad9
commit 0c25b1c272
7 changed files with 43 additions and 37 deletions
Download Patch File Download Diff File Expand all files Collapse all files

39
examples/rethinkdb/README.md
View File

@@ -10,16 +10,10 @@ Setting up a [rethinkdb](http://rethinkdb.com/) cluster on [kubernetes](http://k
Quick start
-----------
**Step 0**
change the namespace of the current context to "rethinkdb"
```
$kubectl config view -o template --template='{{index . "current-context"}}' | xargs -I {} kubectl config set-context {} --namespace=rethinkdb
```
**Step 1**
antmanler/rethinkdb will discover peer using endpoints provided by kubernetes_ro service,
Rethinkdb will discover peer using endpoints provided by kubernetes service,
so first create a service so the following pod can query its endpoint
```shell
@@ -29,8 +23,9 @@ $kubectl create -f driver-service.yaml
check out:
```shell
$kubectl get se
$kubectl get services
NAME LABELS SELECTOR IP(S) PORT(S)
[...]
rethinkdb-driver db=influxdb db=rethinkdb 10.0.27.114 28015/TCP
```
@@ -47,10 +42,10 @@ Actually, you can start servers as many as you want at one time, just modify the
check out again:
```shell
$kubectl get po
POD IP CONTAINER(S) IMAGE(S) HOST LABELS STATUS CREATED MESSAGE
rethinkdb-rc-1.16.0-6odi0 kubernetes-minion-s59e/ db=rethinkdb,role=replicas Pending 11 seconds
rethinkdb antmanler/rethinkdb:1.16.0
$kubectl get pods
NAME READY REASON RESTARTS AGE
[...]
rethinkdb-rc-r4tb0 1/1 Running 0 1m
```
**Done!**
@@ -65,16 +60,15 @@ You can scale up you cluster using `kubectl scale`, and new pod will join to exs
```shell
$kubectl scale rc rethinkdb-rc-1.16.0 --replicas=3
$kubectl scale rc rethinkdb-rc --replicas=3
scaled
$kubectl get po
POD IP CONTAINER(S) IMAGE(S) HOST LABELS STATUS CREATED MESSAGE
rethinkdb-rc-1.16.0-6odi0 10.244.3.3 kubernetes-minion-s59e/104.197.79.42 db=rethinkdb,role=replicas Running About a minute
rethinkdb antmanler/rethinkdb:1.16.0 Running About a minute
rethinkdb-rc-1.16.0-e3mxv kubernetes-minion-d7ub/ db=rethinkdb,role=replicas Pending 6 seconds
rethinkdb antmanler/rethinkdb:1.16.0
rethinkdb-rc-1.16.0-manu6 kubernetes-minion-cybz/ db=rethinkdb,role=replicas Pending 6 seconds
rethinkdb antmanler/rethinkdb:1.16.0
$kubectl get pods
NAME READY REASON RESTARTS AGE
[...]
rethinkdb-rc-f32c5 1/1 Running 0 1m
rethinkdb-rc-m4d50 1/1 Running 0 1m
rethinkdb-rc-r4tb0 1/1 Running 0 3m
```
Admin
@@ -92,6 +86,7 @@ find the service
```shell
$kubectl get se
NAME LABELS SELECTOR IP(S) PORT(S)
[...]
rethinkdb-admin db=influxdb db=rethinkdb,role=admin 10.0.131.19 8080/TCP
104.197.19.120
rethinkdb-driver db=influxdb db=rethinkdb 10.0.27.114 28015/TCP
@@ -124,8 +119,6 @@ since the ui is not stateless when playing with Web Admin UI will cause `Connect
**BTW**
* All services and pods are placed under namespace `rethinkdb`.
* `gen_pod.sh` is using to generate pod templates for my local cluster,
the generated pods which is using `nodeSelector` to force k8s to schedule containers to my designate nodes, for I need to access persistent data on my host dirs. Note that one needs to label the node before 'nodeSelector' can work, see this [tutorial](https://github.com/GoogleCloudPlatform/kubernetes/tree/master/examples/node-selection)