The datalake has been running for the past couple of months and there has been no issues as of yet. The AWS Batch jobs are running as expected and the data is being sent to S3 correctly. AWS Glue runs over the data in the bucket every two weeks. I could easily have changed it to once every month, but I prefer it to run it every two weeks because I want to ensure that the crawling works fine.

So what have I have been doing for the past couple of months? I have been working and learning lots of things regarding datalake technology on AWS. This datalake stuff is very interesting because I needed to convert data from a PostgreSQL database to Parquet-files. A container is started using AWS Batch that connects to the database and queries for specific data pertaining to a specific customer. The data is then convert in the format: CUSTOMER/YEAR/MONTH/file.

I went on a trip with my family to Florida in the United States for two weeks. It was a wonderful trip with ups and downs. The downs were related to our trip to the United States. Let us start from the beginning. We were supposed to wake up at 3AM and then head to Billund Airport by car. From there we would go on the first leg of the flight to Amsterdam.

For the past several months I have been reading the books in the “Wheel of Time” series. I wanted a series of books that were kind of like “The Lord of the Rings” with a rich history. And I found it in these books. I read the first one over several months because I was busy with other things, however I got really into them and I started to care about the characters, so I just had to read them.

So two days ago I ordered 64GB of memory for my laptop. It is two modules where each one is 32GB. I got a Dell XPS 15" 9510 that has two removable memory slots. Some laptops these days have soldered on memory slots that really sucks. Making them removeable is, from my perspective, very convenient and necessary. Manufacturers should do that - all of them. The next part came with opening up the bottom of my laptop.

This is a rather cumbersome guide because lots of moving parts must be aligned perfectly together to make this work. At the company I work at we are 12 people that uses Gerrit. I needed to have these 12 people, and some of them were from different departments, to sit in a virtual meeting at the same time. Thankfully this was done over the internet, but it did take some time to find 30 minutes in their busy schedules to run through a guide on migrating their users from LDAP to OIDC.

I have started watching Community again. It is one of those series that I can watch over and over again without ever getting tired of it, or skipping episodes. I can do the same thing with Friends, How I met your mother, Scrubs, Frasier, Spin City (First 3 seasons), New Girl and Brooklyn Nine-Nine. It is October and the energy crises is hitting us all. The cost of electricity has increased somewhat.

My vacation ends tomorrow. It is the last day and then we are starting up our routines again. I need to figure them out because I have forgotten them already. Is that what happens after two weeks of relaxing? I guess so.

Last week I was in Copenhagen Airport. I was walking around the concourses to pass the time, because there was two hours before my flight. So I thought I would do a stroll around and look at things. I first wandered to concourse F. It is one of the farthers concourses there is - the other one is concourse A. So I was walking to concourse F - there was nothing to see.

For the past several years I have worked with Terraform and only Terraform. CloudFormation is not something that I have worked that much with and I feel like I’m in quicksand. I did you use it a lot back when CloudFormation Stacksets got released. I like that I could just deploy a stackset and it then got pushed to all other accounts that I had designated. It worked really well. However tools such as Terraform has come a long way, so doing multiaccount deployments could not be done directly in Terraform.