Just published uc-local-apex-dev v25.1 π₯³. Run #orclAPEX 24.2 and 23ai 23.6 containerized locally on your machine. Automatic backup, migration, SSL and much more: https://github.com/United-Codes/uc-local-apex-dev
Comments
Log in with your Bluesky account to leave a comment
And Oracle please deliver faster container upgrades π€. New ORDS and container releases take week to land on your own container repo... Otherwise this would be 23ai 23.7 already....
Is there an upside of using two containers vs. just install ORDS in the database container? I find the later easier. It's roughly the same footprint on disk. This blog documents the steps https://pretius.com/blog/oracle-apex-docker-ords/
In my opinion yes:
- There are prebuilt ORDS images so I donβt actually need to install it. I just spin up a container and connect to the DB.
- I can easily upgrade both independently. Upgrading ORDS to a new version is as fast as changing the version number and running docker-compose up -d
ok, not really a big task in the db container. Just a dnf install command. Updating ORDS is then just a dnf update. Same as I have been doing in my Vbox VM's. I'm not using docker, rather use podman on my mac for multiple reasons.
But what if there is a new DB version? You need to upgrade to a newer image and thus throw away the current container. Then you have to install ORDS from scratch again (and other changes).
Or when ORDS requires a different Java version? You would have to mingle in the DB container to upgrade.
All scripted and you have yourself created scripts to backup the database content. Very nice and I will take a look at those. I don't view these containers as anything more than test/dev environments so if there is a new database version my scripts will rebuild the container in minutes.
I like to keep things the way Oracle provides them. That's why I mostly don't want to touch the containers. Maybe @oraclesean.com (author of Oracle on Docker) wants to share his view?
I also use podman, but still with the Docker CLI so I didn't need to change my worklfow :)
Comments
- There are prebuilt ORDS images so I donβt actually need to install it. I just spin up a container and connect to the DB.
- I can easily upgrade both independently. Upgrading ORDS to a new version is as fast as changing the version number and running docker-compose up -d
Or when ORDS requires a different Java version? You would have to mingle in the DB container to upgrade.
I also use podman, but still with the Docker CLI so I didn't need to change my worklfow :)