added better openshift instructions

openshift/README.md

@@ -1,10 +1,20 @@
-# Deploy to OpenShift
+# OpenShift
 
-## Runtime Setup
-TBD
+## Build Setup
 
-## Deployment
-TBD
+Build using the stock NodeJS 4.x image on OpenShift.  Simply `Add to Project` and point it 
+to the latest release version number NOT the head of master.  For example, `Source ref: v4`.
+
+## Deployment Setup
+After some experimenting the optimal deployment profile for this service is:
+
+```
+2 pods or autoscaler 2-4 pods set to 70% CPU trigger
+CPU Request: 5 millicores
+CPU Limit: 1 core
+Memory Request: 100 MiB
+Memory Limit: 1 GiB (although we never hit this limit in our testing) 
+```
 
 ### Change Propagation
 To promote runtime image from one environment to another, for example from *dev* to *test*, run
@@ -14,7 +24,5 @@ oc tag <yourprojectname-tools>/mygovbc-captcha-service:latest <yourprojectname-t
 ```
 The above command will deploy the latest/dev runtime image to *test* env. The purpose of tagging runtime image of *test* env in both \<yourprojectname-test\>/mygovbc-captcha-service:latest and \<yourprojectname-tools\>/mygovbc-captcha-service:test is to use \<yourprojectname-tools\>/mygovbc-captcha-service:test as backup such that in case the image stream \<yourprojectname-test\>/mygovbc-captcha-service, which is used by *test* runtime pods, is deleted inadvertently, it can be recovered from \<yourprojectname-tools\>/mygovbc-captcha-service:test.
 
-The command can be setup as a Jenkins task to faciliate using Jenkins to orchestrate deployment of entire application.
+The command can be setup as a Jenkins task to facilitate using Jenkins to orchestrate deployment of entire application.
 
-## Tips
-To find source code commit point of a runtime instance on OpenShift, open a terminal on one of the running pods and run command `git rev-parse HEAD` in cwd.