Trying Docker for the First Time 📖 Many Learnings! 

Same here!!!

I second this! How does one do local development with Docker (and git) and then deploy to a (running) live server? How do you deal with local test database but then deploy to live server with production database? Etc.

I'm doing this what I think is a very local development environment with the help with docker, mostly node stuff with debugging and nodemon restart on local file changes

Checkout Kubernetes

@@andersryanc then its not realy the docker containers that is awsome. You still need to configure a bigger system to be able to scale an aplication properly an you are tied to yet an other system and you added one more layer of software to your aplication.

That would be awesome. AMP is pretty powerful nowadays. And also discussion on the drawbacks of giving away your content etc.

I didn't know about that. Will Google make it one of key (or mandatory) protocol for improving ranking?

It's not really relevant here. When developing, you would run the web environment normally and use docker for all the dependent services, such as databases, graphql servers etc.

What you see is Visual Studio Code and Kitematic mostly.

Docker is mostly used for (web) services. Not software applications, unless they use a web service. So it doesn't really give you a GUI as such, because it runs a service. The HTML output you see from a web application is viewable through a browser, but it doesn't run the browser in the container itself. So a similar comparison could be made with software applications. For that you would need a Virtual Machine.

1. I'm not quite sure what you mean, if a docker image us based on 16.04, it runs 16.04 in the virtual environment. Docker is an environment for virtual linux environments, it doesn't have a single version of linux that it runs. As for keeping updated, you update the Dockerfiles for your services and re-deploy. You don't keep it auto-updating or stuff like that because then if it updates and things break, you have no idea what happened or what state the instance is in, which is the point of docker. There are generally various handy tags provided by the vendors - you can see the tags for Ubuntu here: hub.docker.com/_/ubuntu 2. You can set up SSH so that you can jump in an inspect a Docker instance, which might be handy, but you don't make changes to the instance - if you want to alter the state of the instance, you change the dockerfile and redeploy so that everyone using the image can be sure that what the hell is going on. It's the same principle of how you don't jump jump into a production service and change the script files there, because nobody would know the discrepancies between what was on the server and what was in version control.

1. Due to similar to lxc concept. Docker inherits Linux kernel of host machine. So you may need to update host machine with docker. And docker is known to have sec issues, even without privileged mode, which can be fully mitigated. In prod u may have hundreds of virtual machines and thousands of containers. Deployment sw help to orchestrate the zoo with good design in zero down time. 2. You are free to update container in dev. There is important part - volumes. Which allow you to start/stop container keeping the data (internal volumes). And may allow you to destroy container and start new (explicit volumes). In dev, when I work with non-compile languages I mount sources directly into container. And changing source code have immediate feedback on GUI. Without need to setup dirty environment.

You are probably right. We use that as a way of thinking about it to explain the concept to a beginner. Though that might not be the best way of explaining to a beginner anyways...

there are many different types of virtualization and the literal definition is: "In computing, virtualization refers to the act of creating a virtual (rather than actual) version of something, including virtual computer hardware platforms, storage devices, and computer network resources." So yes Docker is a virtualized env.

This is the big difficulty that I have in understanding Docker. If you're running a Linux distribution on a Windows machine, how is it different form running a VM? It has its own file system, its own ports? Or is it more something like how Cygwin works? What is "process isolation"?

Thanks *so* much for the detailed reply! I don't know how I missed that SO question in my Googling, but that helps a lot! I'll definitely read further.

Good point. If you are interested here is a deeply technical explanation of what docker is and how are they different for Virtual Machines.. ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-0Xb421-9CTo.html

you all probably dont care but does someone know of a trick to get back into an instagram account..? I stupidly lost my password. I love any tips you can give me.

@Nicholas Arthur Instablaster :)

@Angelo Kaiden I really appreciate your reply. I got to the site through google and I'm in the hacking process now. Takes a while so I will get back to you later with my results.

@Angelo Kaiden it worked and I actually got access to my account again. I am so happy! Thank you so much you saved my account!

@Nicholas Arthur you are welcome =)

Cool! Check out this video on Fun Fun Function! ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-lAJWHHUz8_8.html

check out andrew mead's course

Correction: Only saved 50 minutes 42 seconds, that's 0.845 hour saved.

DevTips :p

For that, you need docker-compose. You will find good tutorials for this, i managed to get it working just fine. The only thing i got stuck on was: when you want to access container named mydatabase port 123 in compose from other container in the same compose file, you just write mydatabase:123 as url. Hopefully it will make sense after reading any guide.

The port mapping as shown in the video is also applicable to folders, effectively making the container run code straight from your local machine. These are called volumes in Dockerworld. As for restarting, you could use nodemon for example which will restart your app when it detects file changes. No image rebuilds or container restarts required.

Sounds awesome! Is this an actual thing or more of a "hacky" approach to solve this?

@@Fellc0de The possibilities are endless but I guess this is a typical development setup. Of course you would remove the volume and nodemon and package your files within the image when you're ready to deploy for production. Actually volumes are essential for persisting data as everything generated within the container will be lost when it is deleted.

You can do a single copy command, but then you lose some build caching features of docker. By splitting the lines, docker reuses the build cache of the package.json and npm install as long as the package.json file doesn't change. If you make it one copy command, any code change reruns the copy and npm install on every build command.

You rebuild it - think of it as the compiled artifact that comes from the Dockerfile.

I'd just like to clarify something.. You can't edit the image itself, but you can edit any volume-mounted files and see those changes reflected in a running container. An example of the dev workflow (with docker-compose) might be to build an image from a base image (such as FROM node:11), copy your dependency description (ie. package*.json) & install them, then volume-mount your app's /src files to the container. In this way, the image gets built without your src files. They, instead, get stuffed into the container once the container is spun up. If you edit your src files on the host, you'll see the changes reflected in the container without having to rebuild :)

I think you will need Windows 10 Professional to run it

@@paulrdrs Thant's the point! you need to run Docker toolbox where nothing works as expected!