Name	Name	Last commit message	Last commit date
parent directory ..
db-migration	db-migration
README.md	README.md
build.sh	build.sh
config.properties	config.properties
start.sh	start.sh

Deploying as a Cloud Foundry Application

Deployment Steps

The quickest way to install Stratos is to deploy it as a Cloud Foundry application.

You can do it in two ways:

Deploy Stratos from source
Deploy Stratos from docker image

You will then be able to open a web browser and navigate to the console URL:

https://console.<DOMAIN>

Where <DOMAIN> is the default domain configured for your Cloud Foundry cluster.

To login use the following credentials detailed here.

If you run into issues, please refer to the Troubleshooting Guide below.

The console will pre-configure the host Cloud Foundry endpoint. No other CF instance should be registered unless the instructions in the section Associate Cloud Foundry database service are followed. All other deployment methods (helm, docker-compose, docker all-in-one, etc) allow the registration of multiple CF instances by default.

Note:

You need the cf CLI command line tool installed and available on the path.
You need to have configured the cf cli to point to your Cloud Foundry cluster, to be authenticated with your credentials and to be targeted at the organization and space where you want the console application be created.
You may need to configure Application Security Groups on your Cloud Foundry Cluster in order that Stratos can communicate with the Cloud Foundry API. See below for more information.
The Stratos Console will automatically detect the API endpoint for your Cloud Foundry. To do so, it relies on the cf_api_url value inside the VCAP_APPLICATION environment variable. If this is not provided by your Cloud Foundry platform, then you must manually update the application manifest as described below.

Running Stratos in Production Environments

Please be aware of the following when running Stratos in a production environment:

Configure a Session Store Secret

Stratos uses a Session Store Secret to protect the user session cookie. We recommend that you set your own value for this secret - choosing an alphanumeric string of your choice.

You can configure this secret by editing the application manifest and adding to the env section, e.g.

applications:
- name: console
  ... memory, disk settings here
  env:
    SESSION_STORE_SECRET: <your session store secret here>

Pre-configure UAA client used for user invites

You can skip this step and configure any CFs invite clients via the Stratos UI.

To set the UAA client for user invites, supply the client id and client secret as environment variables as shown below:

INVITE_USER_CLIENT_ID=<UAA_CLIENT_ID>
INVITE_USER_CLIENT_SECRET=<UAA_CLIENT_SECRET>

This will set the the UAA client and UAA secret used to invite users for the default CF only.

See the invite users guide for more information about user invites in Stratos.

Use of the Default Embedded SQLite Database

We do not recommend deploying Stratos to a production environment using the default embedded SQLite Database. Instead we recommend creating and binding a database service instance to Stratos - for more information see here.

Deploy Stratos from source

To do so, clone the stratos repository, cd into the newly cloned repository and push to Cloud Foundry. This can be done with:

git clone https://github.com/cloudfoundry-incubator/stratos
cd stratos
git checkout tags/stable -b stable
./build/store-git-metadata.sh
cf push

If the cf push exceeds the time allowed see the instructions here

Pre-building the UI

Due to the memory usage of the Angular compiler (see below), when deployed to Cloud Foundry via cf push, Stratos does not use AOT (Ahead-of-Time) compilation.

If you wish to enable AOT or reduce the push time, you can pre-build the UI before pushing.

This can be done with:

git clone https://github.com/cloudfoundry-incubator/stratos
cd stratos
npm install
npm run prebuild-ui
cf push

You will need a recent version of Node installed locally to do this.

The prebuild-ui npm script performs a build of the front-end UI and then zips up the resulting folder into a package named stratos-frontend-prebuild.zip. The Stratos buildpack will unpack this zip file and use its contents instead of building the UI during staging, when this file is present.

Memory Usage

The Stratos Cloud Foundry manifest.yml states that the application requires 1512MB of memory. This is required during the build process of the application since building an angular2 app is a memory intensive process. The memory limit can be scaled down after the app has been pushed, using the cf CLI.

Deploy Stratos from docker image

Deploy Stratos using the splatform/stratos docker image

NOTE: Your Cloud Foundry must have docker support enabled.

cf push -o splatform/stratos:stable -m 128M -k 384M

Alternatively cf push using a manifest

download manifest-docker.yml or create your own manifest file:

applications:
- name: console
  docker:
    image: splatform/stratos:stable
  instances: 1
  memory: 128M
  disk_quota: 384M

now, you can simply push it to Cloud Foundry:
```
cf push -f manifest-docker.yml
```

Associate Cloud Foundry database service

Follow instructions here.

Use SSO Login

By default Stratos will present its own login UI and only supports username and password authentication with your UAA. You can configure Stratos to use UAA's login UI by specifying the the SSO_LOGIN environment variable in the manifest, for example:

applications:
- name: console
  ... memory, disk settings here
  env:
    SSO_LOGIN: true

When SSO Login is enabled, Stratos will also auto-connect to the Cloud Foundry it is deployed in using the token obtained during the SSO Login flow.

For more information - see Single-Sign On.

Troubleshooting

Creating logs for recent deployments

To create a log file of the push

cf push | tee cfpush.log

To create a log file of recent console output

cf logs console --recent | tee cfconsole.log

NOTE If the name of the application has been changed from console in the manifest file please also change the name in the logs statement

Turning on backend debugging logs

The LOG_LEVEL environment variable controls the backend logs

cf set-env console LOG_LEVEL debug 
cf restart console
cf logs console

would output more debugging output such as

  2018-10-24T14:47:36.91+0200 [RTR/1] OUT console.my.domain - [2018-10-24T12:47:36.850+0000] "GET /pp/v1/-o1F0L956QhAIK7R56Uc1lMh5L4/apps/3ddc0bc6-a645-4449-9d1b-6fe86146cf61/ssh/0 HTTP/1.1" 500 0 0 "-" "Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:60.0) Gecko/20100101 Firefox/60.0" "10.228.194.8:42182" "192.168.35.91:61044" x_forwarded_for:"10.228.194.8" x_forwarded_proto:"https" vcap_request_id:"182dddeb-d877-4d58-45f7-0bd886d1caf6" response_time:0.066925325 app_id:"0ba408ef-d0e6-4ab8-96bb-0bc078b8d8fb" app_index:"0" x_b3_traceid:"d166622a0d612fea" x_b3_spanid:"d166622a0d612fea" x_b3_parentspanid:"-"
   2018-10-24T14:47:36.91+0200 [RTR/1] OUT 
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] sessionCleanupMiddleware                     
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] errorLoggingMiddleware                       
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT INFO[Wed Oct 24 12:47:36 UTC 2018] Not redirecting this request                 
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] getSession                                   
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] getSessionValue                              
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] getSession                                   
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] setStaticContentHeadersMiddleware            
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] urlCheckMiddleware                           
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] sessionMiddleware                            
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] getSessionValue                              
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] getSession                                   
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] xsrfMiddleware                               
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] GetCNSIRecord                                
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] Find                                         
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] decryptToken                                 
   2018-10-24T14:47:36.85+0200 [APP/PROC/WEB/0] OUT DEBU[Wed Oct 24 12:47:36 UTC 2018] Decrypt                                      
   [...]

Application Security Groups

If you have problems when deploying Stratos UI as a Cloud Foundry application, check that the Application Security Group you have will allow Stratos to communicate with the Cloud Foundry API.

For information on the default ASGs, see here.

To configure a new ASG for the organization and space that are using Stratos, first create a new ASG definition, for example:

[
   {
      "destination":"0.0.0.0-255.255.255.255",
      "protocol":"all"
   }
]

Save this to a file, e.g. my-asg.json.

Note: this allows example all network traffic on all IP ranges - we don't recommend using this.

Unbind the existing ASG for you organization (ORG) and space (SPACE) with:

cf unbind-security-group public_networks ORG SPACE

Create a new ASG using the definition you saved to a file and give it a name, with:

cf create-security-group NAME my-asg.json

Bind this ASG to your ORG and SPACE with:

cf bind-security-group NAME ORG SPACE

Console fails to start

The Stratos Console will automatically detect the API endpoint for your Cloud Foundry. To do so, it relies on the cf_api value inside the VCAP_APPLICATION environment variable. To check if the variable is present, use the CF CLI to list environment variables, and inspect the VCAP_APPLICATION variable under System-Provided.

$ cf env console
Getting env variables for app console in org SUSE / space dev as admin...
OK

System-Provided:


 {
  "VCAP_APPLICATION": {
   "application_id": ...,
   "application_name": "console",
   "application_uris": ...
   "application_version": ...,
   "cf_api": "http://api.cf-dev.io",
   ...
 }

 No user-defined env variables have been set
 ...

If the cf_api environment variable is absent then set the CF_API_URL variable. See the following Setting the CF_API_URL env variable in the manifest section.

However, if the cf_api environment variable is present, and an HTTP address is specified, it is possible that insecure traffic may be blocked. See the following Setting the CF_API_FORCE_SECURE env variable in the manifest section.

Setting the `CF_API_URL` env variable in the manifest

To specify the Cloud Foundry API endpoint, add the CF_API_URL variable to the manifest, for example:

applications:
- name: console
  memory: 256M
  disk_quota: 256M
  host: console
  timeout: 180
  buildpack: https://github.com/cloudfoundry-incubator/stratos-buildpack
  health-check-type: port
  env:
    CF_API_URL: https://<<CLOUD FOUNDRY API ENDPOINT>>>

Setting the `CF_API_FORCE_SECURE` env variable in the manifest

To force the console to use secured communication with the Cloud Foundry API endpoint (HTTPS rather than HTTP), specify the CF_API_FORCE_SECURE environment in the manifest, for example:

applications:
- name: console
  memory: 256M
  disk_quota: 256M
  host: console
  timeout: 180
  buildpack: https://github.com/cloudfoundry-incubator/stratos-buildpack
  health-check-type: port
  env:
    CF_API_FORCE_SECURE: true

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

cloud-foundry

cloud-foundry

README.md

Deploying as a Cloud Foundry Application

Deployment Steps

Running Stratos in Production Environments

Configure a Session Store Secret

Pre-configure UAA client used for user invites

Use of the Default Embedded SQLite Database

Deploy Stratos from source

Pre-building the UI

Memory Usage

Deploy Stratos from docker image

Associate Cloud Foundry database service

Use SSO Login

Troubleshooting

Creating logs for recent deployments

Turning on backend debugging logs

Application Security Groups

Console fails to start

Setting the `CF_API_URL` env variable in the manifest

Setting the `CF_API_FORCE_SECURE` env variable in the manifest

Files

cloud-foundry

Directory actions

More options

Directory actions

More options

Latest commit

History

cloud-foundry

Folders and files

parent directory

README.md

Deploying as a Cloud Foundry Application

Deployment Steps

Running Stratos in Production Environments

Configure a Session Store Secret

Pre-configure UAA client used for user invites

Use of the Default Embedded SQLite Database

Deploy Stratos from source

Pre-building the UI

Memory Usage

Deploy Stratos from docker image

Associate Cloud Foundry database service

Use SSO Login

Troubleshooting

Creating logs for recent deployments

Turning on backend debugging logs

Application Security Groups

Console fails to start

Setting the CF_API_URL env variable in the manifest

Setting the CF_API_FORCE_SECURE env variable in the manifest

Setting the `CF_API_URL` env variable in the manifest

Setting the `CF_API_FORCE_SECURE` env variable in the manifest