In this lab you will deploy and test the IBM Stock Trader Lite V2.0 application in the IBM Cloud Kubernetes Service.
The IBM Stock Trader Lite V2.0 application is a polyglot stock trading sample, where you can create various stock portfolios and add shares of stock to each for a commission. It keeps track of each porfolio’s total value and its loyalty level, which affect the commission charged per transaction. It also lets you submit feedback on the application, which can result in earning free (zero commission) trades, based on the tone of the feedback. (Tone is determined by calling the Watson Tone Analyzer).
The architecture of the app is shown below:
-
Tradr is a Node.js Express UI for the portfolio service that implements the Backend for Frontend (BFF) pattern.
-
The stock-quote is a Open Liberty service that queries an external service to get real time stock quotes via an API Connect proxy.
-
The portfolio microservice is a Spring Boot application that sits at the center of the Stcok Trader V2.0 app. This microservice:
- persists trade data using JDBC to a MariaDB database
- invokes the stock-quote service to get stock quotes
- sends completed transactions to a local DNS proxy to IBM Event Streams running in the IBM Cloud
- calls the trade-history service to get aggregated historical trade data
-
The event-consumer microservice is a POJO application that receives the transaction data from Event Streams and calls the trade-history service to publish the data to the reporting database.
-
The trade-history microservice is a Python Flask application the manages historical trading data in a PostgreSQL database.
This lab is broken up into the following steps:
1.1 Verify command line access to your cluster by entering the following command from web based terminal you're using for the labs. Substitute you cluster name for CLUSTER_NAME
ibmcloud cs cluster get [CLUSTER_NAME]
You should get output that looks looks something like the following:
Retrieving cluster user001-cluster...
OK
Name: user001-cluster
ID: bpqalm9w0fg8b11rdi1g
State: normal
Created: 2020-03-20T11:36:57+0000
Location: wdc04
Master URL: https://c1.us-east.containers.cloud.ibm.com:29560
Public Service Endpoint URL: https://c1.us-east.containers.cloud.ibm.com:29560
Private Service Endpoint URL: -
Master Location: Washington D.C.
Master Status: Ready (2 days ago)
Master State: deployed
Master Health: normal
Ingress Subdomain: user001-cluster-5290c8c8e5797924dc1ad5d1b85b37c0-0000.us-east.containers.appdomain.cloud
Ingress Secret: user001-cluster-5290c8c8e5797924dc1ad5d1b85b37c0-0000
Workers: 2
Worker Zones: wdc04
Version: 1.16.8_1526
Owner: -
Monitoring Dashboard: -
Resource Group ID: 2a926a9173174d94a6eb13284e089f88
Resource Group Name: default
Point to your cluster by running
ibmcloud ks cluster config --cluster [cluster name]
1.2 Verify you have Helm V3.1 by entering the following command from web based terminal you're using for the labs
helm version
You should get output that looks looks something like the following:
version.BuildInfo{Version:"v3.1.1", GitCommit:"afe70585407b420d0097d07b21c47dc511525ac8", GitTreeState:"clean", GoVersion:"go1.13.8"}
If you don't have version 3, set an alias
alias helm="helm3"
1.3 USERNAME or STUDENTID env var set in your web based terminal. Ask your instructor for the correct setting for these variables if both of the following commands return empty values:
echo $USERNAME
echo $STUDENTID
1.4 SETUPURL env var set in your web based terminal. Ask your instructor for the correct setting for this variables if the following command returns an empty value:
echo $SETUPURL
Like a typical Kubernetes app, Stock Trader use secrets and ConfigMaps to store information needed by one or more microservices to access external services and other microservices. We've provided that info in a file hosted in Cloud Storage and there is a script that you'll use to retrieve it.
2.1 From your web terminal window add the public Helm repo that contains the Helm charts for MariaDB and PostgreSQL.
helm repo add bitnami https://charts.bitnami.com/bitnami
2.2 From your web terminal window clone the Github repo that has everything needed to deploy the aggregated Stock Trader app.
git clone https://github.com/IBMStockTraderLite/stocktrader-iks.git
2.3 Go to the directory required to run the setup scripts
cd stocktrader-iks/scripts
2.4 Retrieve credentials and other details needed to create secrets and/or ConfigMaps. Ask you instructor for the SETUPURL for the command below.
./setupLab.sh $SETUPURL
2.5 Verify that the output looks something like the following:
Script being run from correct folder
Using user001-cluster as IKS cluster name
Using stocktrader-user001 as Kakfa Topic name
Validating setup URL ...
Getting Ingress subdomain for cluster user001-cluster ...
Updating Stock Trader Helm chart with ingress subdomain: user001-cluster-5290c8c8e5797924dc1ad5d1b85b37c0-0000.us-east.containers.appdomain.cloud
Updating variables.sh with Kafka topic : stocktrader-user001
Setup completed successfully
2.6 Also verify that there is now a file called variables.sh in the current folder
In this part you'll install the prereqs step by step before installing the Stock Trader V2.0 application.
3.1 Install MariaDB by running the following command. Verify that no errors are displayed by the installation script.
./setupMariaDB.sh
3.2 Install PostgreSQL by running the following command. Verify that no errors are displayed by the installation script.
./setupPostgreSQL.sh
3.3 Create the DNS Proxy and store all the access information as secrets for the external Kafka installation. Verify that no errors are displayed by the script.
./setupKafka.sh
3.4 Store all the access information as secrets for the API Connect proxy to the external realtime stock quote . Verify that no errors are displayed by the script.
./setupAPIConnect.sh
3.5 Store all the access information as secrets for the external Watson Tone Analyzer service . Verify that no errors are displayed by the script.
./setupWatson.sh
3.6 Verify your progress so far. Run the following to see the running pods:
kubectl get pods
The output should show the 2 database pods which should be running and in the READY state.
NAME READY STATUS RESTARTS AGE
stocktrader-db-mariadb-0 1/1 Running 0 70s
stocktrader-hist-postgresql-0 1/1 Running 0 57s
3.7 Next look at your services
kubectl get svc
3.9 Verify that the output includes services for MariaDB, PostgreSQL and your DNS proxy to Kafka
NAME TYPE CLUSTER-IP EXTERNAL-IP
stocktrader-db-mariadb ClusterIP 172.21.216.8 <none>
stocktrader-hist-postgresql ClusterIP 172.21.235.35 <none>
stocktrader-pubsub-kafka-dns-proxy ExternalName <none> broker-0-0mqz41lc21pr467x.kafka.svc01.us-south.eventstreams.cloud.ibm.com
In this part you'll install all the Stock Trader Lite V2.0 microservices using a Helm chart that has templates for all the microservices. Note that the microservices require the information stored via secrets in the scripts you ran in the previous section.
4.1 Go back to the top level folder of the cloned repo
cd ..
4.2 Install the Helm chart. Verify that no errors are displayed
helm install stocktrader stocktrader
4.3 The output of the helm command should look something like the following:
NAME: stocktrader
LAST DEPLOYED: Sun Mar 22 16:01:54 2020
NAMESPACE: default
STATUS: deployed
REVISION: 1
TEST SUITE: None
NOTES:
Stock Trader Lite V2.0 is deployed.
The Tradr web application is available at:
http://user001-5290c8c8e5797924dc1ad5d1b85b37c0-0000.us-east.containers.appdomain.cloud/tradr
This is the base external URL for the Trade History Service:
http://user001-5290c8c8e5797924dc1ad5d1b85b37c0-0000.us-east.containers.appdomain.cloud/trade-history
4.4 Verify that all the pods are running by running the following command. Note you may have to run this command multiple times before all the pods become READY.
kubectl get pods
4.5 Keep running the command until the output looks something like the following. NOTE: It can take 30 seconds or more for all the pods to reach the READY state.
NAME READY STATUS RESTARTS AGE
event-streams-consumer-5dbfc5bb46-lvbhg 1/1 Running 0 89s
portfolio-6d4b99dc88-28b8v 1/1 Running 0 89s
stock-quote-695cc88b48-jzmwx 1/1 Running 0 89s
stocktrader-db-mariadb-0 1/1 Running 0 12m
stocktrader-hist-postgresql-0 1/1 Running 0 12m
trade-history-59cbc8fb59-8p26d 1/1 Running 0 89s
tradr-7fcd7c4bdf-5td8h 1/1 Running 0 89s
4.6 Look at the services available to be called outside of the cluster by running the following command:
kubectl describe ingress stocktrader-ingress
The output should look like the following indicating that only the tradr web app and the trade-history service are configured to be accessed from outside your cluster. All the other services are configured to be accessed from within the cluster only.
Name: stocktrader-ingress
Namespace: default
Address: 169.63.109.34
Default backend: default-http-backend:80 (<none>)
Rules:
Host Path Backends
---- ---- --------
user001-cluster...cloud
/tradr tradr-service:3000 (172.30.170.91:3000)
/trade-history trade-historyservice:5000 (172.30.170.92:5000)
Annotations:
Events: <none>
In this part you'll verify that the various microservices are working as designed.
5.1 Get the external URL of the tradr web app by running the following command:
helm get notes stocktrader
5.2 Open the displayed URL with your browser to bring up the login page:
5.3 Log in with the following credentials (note these are the only values that will work)
username: stock
password: trader
5.4 Click Add Client and fill in the requested values to create a new client. Click Save
5.5 Click on the link in the Portfolio ID column to see the details of the new client.
5.6 Do 3 or 4 "Buy" operations with different stocks.
5.7 Sell part of one of the holdings you just bought and verify that the Shares column in the table is updated appropriately.
5.8 Click on Feedback and submit some client feedback to be analyzed by Watson Tome Analyzer. Try to sound as angry as possible with the feedback. The app will award 3 free trades for very angry feedback and 1 free trade for feedback that is only slightly angry or not angry at all.
5.9 Verify that the following data flow:
Portfolio Svc--->Kafka--->Event-Consumer-Svc--->Trade-History-Svc--->PostgreSQL(Trade History DB)
works by querying the trade-history service for all trades in a portfolio.
Run the following command to get the base external URL of the Trade History Service:
helm get notes stocktrader
5.10 Copy the URL into a new browser tab and append the path /trades/1000 to it. For example if the URL is http://user049-cluster.us-east.containers.appdomain.cloud/trade-history/ then the full URL will be http://user049-cluster.us-east.containers.appdomain.cloud/trade-history/trades/1000
5.11 Verify that the returned output is in JSON format and that the history has captured all the trades you did while testing. A screenshot of the kind of the data that should be returned is shown below.
Free up resources for subsequent labs by deleting the Stock Trader app.
6.1 Run the following commands to cleanup (note: you can copy all the commands at once and post then into your command window)
cd scripts
helm uninstall stocktrader
./cleanupWatson.sh
./cleanupAPIConnect.sh
./cleanupKafka.sh
./cleanupPostgreSQL.sh
./cleanupMariaDB.sh
cd -
You installed and then tested the Stock Trader Lite V2.0 microservices sample application and got some insight into the challenges of deploying microservices apps in a Kubernetes cluster.