Gain practical experience in managing a cloud-based MySQL database with a focus on implementing connection pooling and performing database migrations. You will work with both Azure and Google Cloud Platform (GCP) for this assignment.
- Azure:
- Go to Azure Database for MySQL servers --> Click Create --> Choose Flexible Server --> Choose a Resource Group/Create new one --> Enter a server name and change compute + storage to Standard_B1s --> Enter server admin login name, password, confirm password --> Click Next: Networking --> Click + Add 0.0.0.0 - 255.255.255.255 and + Add current client IP address --> Hit create
- Once created, go to Server Parameters and change the max_connections to 20 and connect_timeout to 3, change require_secure_transport to OFF
- GCP:
- Go to SQL--> Create Instance --> Choose MySQL --> Enter an Instance ID and a Password --> Choose Enterprise and change the preset to Sandbox --> Click Show Configuration Options --> Go to Machine Configuration and choose Shared Core and 1vCPU, 0.614 GB --> Go to Connections and Add a Network --> Name it Allow-All and set it as 0.0.0.0/0
- Azure and GCP: I created two tables (Patients and Laboratory Test) and populated them.
- I have used these tables in a previous assignment.
- There is a one to many relationship between patients and laboratory tests
- You can check if your table was made by entering the terminal
mysql -u <user> -h <server name/IP address> -p - I then populated the tables with random generated data.
- Once I was done with the file, I went into the terminal and typed in
python populate_gcp.pyorpython populate_azure.py(whichever one you start with). - Then check in mysql terminal with
SELECT * FROM patients;andSELECT * FROM laboratory_test
- After connecting to MySQL workbench for both Azure and GCP, I used the reverse engineer function to generate the ERD for my database.
- I used the templates I have used in a previous assignment.
- I made a
.envfile and a.gitignorefile with my credentials in it. I also usedapp.pyfor Flask. - Please check out the Screenshots folder to see the Flask App.
- To set up Alembic, I first typed in the terminal
alembic init migrations - Then I went to
alembic.inito editsqlalchemy.url = mysql+mysqlconnector://username:password@host/database_namewith my own credentials. Make sure to addalembic.iniin your.gitignorefile. - I went to
env.pyto change models tofrom <your .py file that was used to generate the tables> import Baseandtarget_metadata = Base.metadata. Make sure to comment thetarget_metadata = Noneline. - Then go back to the terminal and create a migration with
alembic revision --autogenerate -m "create tables" - Then run the migration with
alembic upgrade head - Then save it as
alembic upgrade head --sql > migration.sqlwhich will create a filemigration.sql - Go to gcp.py and azure.py and add a new column in one of the tables.
- For example: In azure.py I added a column
tetris_score = Column (Integer). Though it does not relate to the database, I added it to note that this column was generated through Alembic. - In gcp.py I added a column
suika_score = Column(Integer)which also does not relate to the database but I added it since I was playing the game at the moment 😄 (also to note that this column was generated through Alembic).
- For example: In azure.py I added a column
I ran into many errors in this assignment, mainly in integrating with Flask and populating the data. The populating portion ran into errors because the range was small and I had to check out what the Faker package had. Flask integration was problematic because I was able to deploy it but I wasn't able to deploy it with the randomly generated data I populated. Migration with Alembic was tedious for me since I kept deleting and restarting the process for my data because I didn't realize I already had an Alembic_version column in my table. Once I dropped the column with DROP TABLE alembic_version, I was able to go through the steps for migration with Alembic.