*[Accessing services running on the cluster](#otherservices)
*[Requesting redirects](#redirect)
*[So many proxies](#somanyproxies)
## Accessing the cluster API<a name="api"></a>
...
...
@@ -203,100 +202,7 @@ You may be able to put a apiserver proxy url into the address bar of a browser.
way that is unaware of the proxy path prefix.
## <a name="redirect"></a>Requesting redirects
Use a `redirect` request so that the server returns an HTTP redirect response and identifies the specific node and service that
can handle the request.
**Note**: Since the hostname or address that is returned is usually only accessible from inside the cluster,
sending `redirect` requests is useful only for code running inside the cluster. Also, keep in mind that any subsequent `redirect` requests to the same
server might return different results (because another node at that point in time can better serve the request).
**Tip**: Use a redirect request to reduce calls to the proxy server by first obtaining the address of a node on the
cluster and then using that returned address for all subsequent requests.
##### Example
To request a redirect and then verify the address that gets returned, let's run a query on `oban` (Google Compute Engine virtual machine). Note that `oban` is running in the same project and default network (Google Compute Engine) as the Kubernetes cluster.
To request a redirect for the Elasticsearch service, we can run the following `curl` command:
**Note**: We use the `-L` flag in the request so that `curl` follows the returned redirect address and retrieves the Elasticsearch service information.
If we examine the actual redirect header (instead run the same `curl` command with `-v`), we see that the request to `https://104.197.5.247/api/v1/redirect/namespaces/default/services/elasticsearch-logging/` is redirected to `http://10.244.2.7:9200`:
* HTTP 1.1 or later with persistent connection, pipelining supported
< HTTP/1.1 307 Temporary Redirect
< Server: nginx/1.2.1
< Date: Thu, 05 Mar 2015 00:14:45 GMT
< Content-Type: text/plain; charset=utf-8
< Content-Length: 0
< Connection: keep-alive
< Location: http://10.244.2.7:9200
<
* Connection #0 to host 104.197.5.247 left intact
* Closing connection #0
* SSLv3, TLS alert, Client hello (1):
```
We can also run the `kubectl get pods` command to view a list of the pods on the cluster and verify that `http://10.244.2.7` is where the Elasticsearch service is running:
```
$ kubectl get pods
POD IP CONTAINER(S) IMAGE(S) HOST LABELS STATUS CREATED
