Skip to content

Blog

Using Umbrel

Installing UmbrelOS in VM

Using OracleVM, set up a new machine as usual. You will need to manually select Linux and Oracle Linux (64-bit) and also be sure to Enable EFI.

After installing the medium to the disk the machine will turn off. Change the boot order ot only include the hard disk. Go to storage and click on the installation disk. There is an option to eject the disk on the far right.

You can then navigate to umbrel.local in the browser.

Serverless Relational Databases

Here are some services I should look into that would likely end up costing a lot less than something like RDS.

Service Compute Pricing Storage Pricing Ideal For
Amazon Aurora Serverless Billed per ACU per second. $0.10 per GB-month. Variable workloads with MySQL/PostgreSQL compatibility.
Azure SQL Database Serverless Billed per vCore per second; e.g., ~$0.5218 per vCore-hour for Gen5. Billed separately per GB. Intermittent workloads with auto-scaling and pausing capabilities.
Google Cloud SQL On-demand and committed use pricing; e.g., ~$0.0825 per hour for 1 vCPU instance. $0.17 per GB per month for SSD. Managed relational databases with automated scaling.
PlanetScale Free tier available; paid plans based on resource consumption. Included in usage-based pricing. MySQL-compatible databases with serverless scaling and modern development features.
CockroachDB Serverless Free tier available; usage-based pricing beyond free tier. Included in usage-based pricing. PostgreSQL-compatible distributed SQL databases requiring high availability and global scaling.
  • in Flutter
  • 1 min read

Flutter Docs Android

Changes in build.gradle in the android directory

ext.kotlin_version = '1.8.10'

and 

classpath 'com.android.tools.build:gradle:8.1.0'

I also changed the gradle-wrapper.properties to version 8.1

I also manually set the JDK version to 17 (default 21) in gradle.properties

org.gradle.java.home=/Library/Java/JavaVirtualMachines/amazon-corretto-17.jdk/Contents/Home

2024 10 06 ESPN Fantasy Football SpringBoot

Quickly reviewing what I learned looking through the api.

In the in-match api return for an active week, i.e. scoringPeriodId is set to the current week, and our view contains mBoxScore, mScoreBoard, mLiveScoring, mMatchupScore, we'll get each teams active roster, currrent total projected points, and current live points. In each roster spot, we'll get two different stat objects, one representing the original projection and one the live stats. In the original projection, it's important to note that we have a appliedTotalCeiling value, we'll need that.