GitHub Demo Days, 13 Sep 2021

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: From idea to AKS Deployment with GitHub! #DemoDays

Description

James Le Ha starts with an idea in an issue and takes it all the way to a Production deployment fully leveraging automations available in GitHub

0:00 - Start
1:13 - GitHub Automated Workflows
1:53 - Process overview and the new GitHub Issue forms
4:31 - GitHub Actions
7:30 - local build vs GitHub Codespace, the end of "Works on my machine!" problems
9:23 - Running the webapp
10:38 - Making changes to the application
13:31 - Digging into the Redis database
14:42 - Committing our changes to the repo
15:37 - GitHub CLI
18:19 - Action workflows triggered by pull requests
21:39 - Security vulnerability scanning with CodeQL
23:17 - Deploy flexibility with ChatOps in Microsoft Teams
28:10 - Verifying the deployment
29:30 - Merge Automation
30:32 - Available Code scanning tools
31:37 - AKS Cleanup PR, resource automation
33:18 - Workflow graph
36:03 - Environments and Secrets
38:44 - Release Automation
39:30 - Packages and Containers
42:24 - GitHub project tables
45:37 - All the Ops defined!

Demo repo: https://github.com/James-LeHa/DemoDay-AKS-to-Z

A

Hello, hello, hi friends good morning, good afternoon, good evening, good night, wherever you are nice to see everyone, hello, my friends from brazil happy friday, happy saturday for anyone who's staying up late to watch us welcome to demo days today. We're going to take a look, uh take a look at something very practical, workflow automation, specifically how you can speed up deployments with chat, ops and github automation.

A

Today, I've got james laha field solutions, engineer we'll run today's demo, and we've also got daniel selman enterprise solutions, engineer and he'll be in the chat up with any questions that you might have. As always, if you find today's program useful and want to talk about how you can speed up deployments in your organization feel free to use the qr code that I'm popping up on the screen right now in the corner to book a meeting and talk with us directly so before I turn it over to james. I've got one question for chat.

A

What tool do you use the most when automating your workday with that james? Take it away.

B

Thanks so much jj everyone thanks for joining us on this friday, github demo day. We do this every week and it's always about a different topic with github today's going to be about github automation, workflows, as aj mentioned at the top. We're going to be doing this on my public github repository you can find here at this link. We'll also drop that in chat here in a second. So, with this, let's go and get started.

B

We're gonna go through a couple: different scenarios of the sclc, hopefully some you find applicable and maybe some different ways of thinking of how you can automate to just improve the quality of life of your developers and your dev processes. So right at the back. When we're talking about development, it starts with an idea an inception of what are we going to do? What do we need to change so with that? There's probably a lot of different project board solutions out there you may use inside getup.

C

B

A native project board functionality, a lightweight common style board that you can set up and use to track different issues or work items or make just new items and notes.

B

So normally I could go ahead and just make an issue and start from there, but I want to sort of define a process of how we go through this flow. It's really going to start us off on this journey of modifying and creating our application.

B

So to do that, I'm actually go ahead to my issues: tab and we're going to use a concept called github issues templates, something you may actually be familiar with already. There's something new that we're going to look at today, which is actually called github issue forms. So it's a currently new beta functionality. You can try on public repositories, and with this you can see, I've actually defined that as a template, but included selectors and input fields as well.

B

So with this, I'm actually making a new feature request, because for today's demo we're actually updating our web app service. It's a python application with redis backend database, it's going to be deployed to kubernetes cluster and we're actually going to update that for a bake off service. We have coming so with this. Let's go ahead and update our issue.

B

I want to get a priority level. This values that I went ahead and selected in this issue, template that I'll show you here in a second. So this is a high priority for us category. We actually want to change the data.

B

We're going to submit this issue and we see a few things happening right off the bat right using the issue template functionality. We can go ahead and assign users to it, go and assign labels as well. The way the issue template forms look like if we take a look here used to find this in the root of your password in the dot github folder, you create a folder called issue templates and then within using yaml, you can actually define these issue. Template forms.

A

Sorry to interrupt james.

B

Oh zoom in uh great point: y'all: let's go ahead and do that and you let me know if we need a little bit more guys. I will.

C

Try to also be zooming.

B

In as we'll get additional content, so here this issue template form, let's look at actual raw yaml itself, it's pretty simple!

B

You define what you like name tile description labels. If you use issue templates before pretty similar here, though, when you find that body, that's where you can use things like drop downs, what attributes it has a label and different options or input fields and make that required or optional as well.

B

So now that we use this issue form normally, I have that project where we saw earlier and I'll go ahead and add my issue to that project board. But why not? Let automation do it for us, so we go to my github actions tab here we can see again action. Workflow has just ran and give actions at high level. It's automation with github the engine behind this is supporting both cicd for really any given activity.

B

You want to automate because it's simply just running code on different runners or machines that are up there in our cloud or you can also self-host yourself. We can see it's actually ran one minute ago, eventually to look at this. What triggered it?

B

It was actually from an issue being opened and let's take a look at this workflow file, it's a really simple example and we'll get to some more complex scenarios here in a little bit here you can see, namely it's once again still yaml, I'm giving it a name here at the top I'm defining this on environment. When do I want this workflow to actually be triggered here?

B

We're actually not even doing this as a pull request or a push or a merge, we're doing this whenever an issue is opened or closed or late, we'll go ahead and setting this environment token value for our github secrets using the github token. This is a token you can actually have inside your workflow files. You don't have to set it as an actual secret value.

B

It's within the contents of your rebuild so especially when trying to automate some things around your repository, maybe until it makes an issue or a label or change you can use this get a token to make some of those modifications here. We.

A

B

A job that's running on an ubuntu cloud runner, so it's our get a poster runner that exists up there in the cloud, and you can point to these for different jobs that you need to run and under each job. You're gonna have multiple steps and here's really the power of github actions in the git community right. This uses keyword right here.

B

Whenever we see this, we're actually gonna be referencing a github action, aka, a code source artifact that someone's created for us and thanking this user out here for creating an action for doing just that assigning to a project board reasons. A couple other conditional factors here saying: if the issue itself had a label that contained the value enhancement, go ahead and run here, I'm defining what project I wanted to point to project board id number one and what column I wanted to point to the backlog column.

B

So, let's look and see what actually has just occurred, though we see here, our feature: request board was updated three minutes ago. Timing seems pretty close to me and we see here. The issue that opened for that feature. Request was automatically added to our backlog board for us. If you want to build some sort of pattern around this right, whenever someone closes an issue, maybe send out an alert or tax.

B

Whatever someone makes an issue go and add to the board automatically based on certain parameters, because then you have multiple different boards for bugs for triage for future requests. Think of those scenarios that you want to use with github and get actions all right. So let's go ahead and go back to our code tab. So we have an idea now something we want to put in a future request. Now, let's actually start coding and modifying our application to accomplish this so once again, really simple web applications. Actually an example out there from microsoft.

B

You can find in the microsoft documentation it's actually down here in the readme file and what we're using here is python redis, we're actually employees in aks cluster. Now I could go ahead and pull my code down locally and add this in my vs code instance, but many people have probably seen before get code spaces which actually went ga not too long ago and getting code spaces. Let's take that development environment and put that up in the cloud. So I'm waiting for this new code spaces button.

B

You can see my browser's just changed. Let me know it's going to set up a code space for me, so it's actually building a container to host my vs code environment. It's actually have interactability with github, and it's actually going to give me those extensions and everything I'm used to working on locally. So from right. Here we can see it's launching, for us, we can see. Is that the sql experience many people are probably used to.

B

And now we can also see on the left-hand side. We see extensions as well that are being populated installed for us, including docker, some different python frameworks and some additional recommended actions. I'm sure everyone's gone through that classic use case or scenario where hey it works on my machine. But not yours and nothing is more of a pain than trying to figure out what framework will just be using what tools or packages do. I need to actually run the solution. We'll get a code space.

B

I guess, as you can see in a matter of seconds, go and spin up the development environment that you need with your specifications. You can actually make this even more customizable using this dot, dev container folder, so this dot dev container folder. You can find what sort of docker image you want to use, setup and user parameters additional settings as part of vs code itself, and here are additional ports that you want to reserve for your development process. Here you can see we're going to be using port, 8080 and 6379.

B

We have some additional scripts and a docker file, and really that's the great part about this. Every project may be unique but define that and have a reusable framework that everyone on your team can have and pull from all right now, we're in our codespaces environment.

A

Let's actually show you what this.

B

Application is doing so, I'm actually going to run some docker commands here. Let's just do a docker compose up and see what's going on.

B

This is something I would normally do locally right in my environment, I probably would have docker running eating a bunch of ram and memory, which is never fun, but it's what you need in order to develop and test out containerization on your machine right using my docker post file. Here, though, we're still in the context of my browser, we haven't even actually opened anything locally on my machine at all, and we actually see that looks like the state is now running. We take a look here at our ports that were reserved in our definitions.

B

We can actually see that processes are on there, one here for our database 6379 and one here for our front 8080..

B

So let's go and take a look and navigate to that url, and we can see here our codespace environment, that'll, split up for us on port 8080 is actually showing us the web application. You have it's really simple. It's a simple voting application! If you will right two database tables here for red and blue, as you select a value, it will just enumerate for us along this reset filter. You can use as well.

B

So that's great. We see how the application works. We see that it's actually up and running locally, let's go and make those changes, see them reflected once in our code space and actually deploy them and use github to help us get this up into our azure portal.

B

All right, navigating back here to our code space environment, I'm actually going to go ahead and shut down these containers. I can do that here from the terminal, control, z or control c, but here I'm actually going to go ahead and just stop them using the docker plug-in that's available for vs code and, of course, available for your github code space.

B

I've seen some questions here in chat. Some from alan amazing people that are joining.

C

B

Don't worry about it, this is actually document on the review file of this application, but at high level, what we're going to do is we're actually going to make changes to the web app based on a feature request from one of our users and we're going to see how give.

A

B

To help you deploy that up there and run some automation for containers, creating releases and actually working with name spaces in your aks cluster. So now that our container has stopped, let's go and make changes to this application.

B

First, I'm actually going to go and check out a development branch.

B

All right, so our new branch here in my code space and the actual main code for this exists in this azure vote. Folder one level deeper, there's. Actually this config file.cfg so inside this. This is actually where the application is taking those values and instantiating the database from so. Let's go ahead and update this.

B

Let's go ahead and put in two values here, so I would love to get some recommendations from chat here. Your favorite baked goods probably going to do cake empire my go-to, but if there's any further recommendations happy to see and accommodate those as well so give everyone a second there in chat to see if we have any recommendations for some values, if you guys have any issues with quality and such let us know trying to zoom in as much as possible, this will also be available back on demand.

B

If you need to re-watch it and maybe it'll be better on an upload site such as youtube all right. So let's go ahead and go for our default values. Here.

B

You're going to save that, let's go ahead and do our docker compose, but this time let's do build.

B

All right, let's see, doctor compose up, see this looks like and actually I did see some recommendations there, chad. So I appreciate that you know. So, let's see we had one that's pizza and one that is brownie from both allen and jason. Appreciate that y'all, let's go ahead and try this again. Let's do our documents build all right doctor composed up.

B

All right looks like our application is running, let's go ahead and take a look on port 8080 once again, and we can see that.

A

B

Our title change those two values for our database to vote on courtesy of chat, one additional thing that you may use quite often when you're developing and working on locally. Perhaps you want to manipulate and see what the database looks like with that. We have this nosql extension.

B

Let's actually add a database connection here, it's actually like redis, it's going to go and point us to that port. 6379, let's go to connect and let's actually see the tables we have available.

B

Everybody should go back here to your application. Let's go around. Some updates, you know seems like a pizza day. It is a friday. So if everyone can enjoy that we're actually going back here, you know- I don't think that's really therapy says such a lead. So let's go ahead and bump up brownies just a little bit all right. We did save those going back here to our code space.

B

Let me refresh this page and just like that we can see those values also reflected inside of our local dev environment, especially in terms of when you're working with data trying to see what changes occur. Have that on this localized instance, in your browser, it removes all those barriers of what kind of machine do you have they have enough ram or memory.

B

They have the right things installed on all those pins kind of go away when using a code space, that's a repeatable environment, all right, so it looks like that is up and running for us. Let's go ahead and commit our changes, go and stop our containers. First.

B

Let's do our git commit.

B

All right now, in this case I'll, do a good push, but, as you can see here, it's actually not connected up that origin. Yet, while this is on game.com, what is using our game code space, we still need to push and select that origin, just as when we're developing locally on our machine and make a new branch.

B

So let me go ahead and select and paste our did push command.

B

Let's start getting the stats, it's confirmed. It looks like everything's up to date with our origin on give.com from here. I would actually probably go back to my repository, make a pr and merge this in and start that kind of process. But if you didn't know, github also has a github cli, that's just using the gh command. It's actually something about installed in my code space as well to have available to me so that.

A

As you can see, there's plenty of examples of additional.

B

Commands and controls you can use in order to operate and work with github right from the context of your terminal.

B

So I'm actually going to go ahead and work with your actually going to go ahead and create one. So here.

A

B

Can see it's giving me some prompts whatever the power to be I'll. Have my defaults I'll need a body, and here it actually submitted right from my browser, or I can actually continue in browser and see it open up. Let's go ahead and select that middle option.

B

You can see here seconds back to our repository, it's actually on and creating us a pull request from our demo day, dev bridge to our main bridge everything we did in our code space from creating this branch pushing up there. Making modifications to now merging has brought us all the way back here into this pull request, I'm actually using a cord request, template which you can also find inside your github repository and the reason behind that is. I actually want to kind of standardize when you make a pull request.

B

What information can you put in there so for this first one I'm actually going to go ahead and press the pound sign and actually just go and select that issue number seven from earlier that we created.

B

If there's any other, related issues I want to link to. I can as well now, which will make additional comments or changes.

B

Add them here all right, so let's go ahead and create this for request.

B

If we scroll down russia and see some additional pieces of automation and running here, we'll give actions as part of our cic checks.

B

And while these run.

C

I'm actually going to go ahead.

B

And assign a reviewer to this, let's meet my colleague daniel who's in chat and helping answer. Questions really appreciate it and let's take some time here, let's look at chat and see if there's any questions or updates from y'all.

B

All right so seeing a couple things about give co-pilot that just came out. That's people are working and playing with that's great, seeing here's someone that just joined no problem so harsh for that we're actually doing a demo of deployment in aks cluster from github, so we're gonna, give actions, we're gonna, get some security scanning and some approval mechanisms you have as well. So don't worry about. This will also be available on demand and this repo we're working with.

C

Is also public.

B

And open source, if you want to catch.

C

B

Kind of look at the source code all right, so our pull request is running. We have a couple status checks that are running as well. I went ahead and add daniel as a reviewer. So once these finish up, you can take a look there for us. Let's actually just start in. Let's take a look at some of these because there's a looks like a lots going on, but it's pretty simple and consolidated as part of github action, workflow files.

B

C

B

Over here to my actions, tab, we can see here. Two actions have been triggered and ran about a minute ago. Right well for my pull request. The first one here is for my build test and scan action. So let's take a look at this, so this summary page that I showed you this sort of ui visual of what's going on in your workflow file. So each of these are different jobs and jobs can be related to each other or they can be standalone.

B

As you can see here, built-in tests they actually ran in parallel, while the build and container scan actually was relying on both of these to pass first before it run, that's actually not taking that workflow file for our build test and scan the first.

A

Thing you're going to.

B

Notice here is that on section is actually our trick. Previously, we triggered this for a issue being opened right. This is part of that automation, movement. Our project board right here, though, we're actually triggering this as part of pull request to our main branch under get. Actually you have one or more jobs, each of those jobs.

B

You're gonna, give it a name and you're gonna determine where do you want that job to run and what type of machine either in our cloud they get a post or on a self-hosted machine that you can have on premises locally, doesn't matter to us, you just connect that up and that can be used for workflow run references of give actions, but here I'm just using an ubuntu latest runner, I'm actually using a matrix strategy, so something you can define inside of action.

B

It's really great for you to test multiple programming languages or multiple different tests, or use cases to find this array here of 3 7 and 3 8. That's actually what's making our job run.

A

Two separate versions.

B

As you can see here for build and test three seven and three eight, that being.

A

Back here, let's look at our step section.

B

We have one more steps in each of these steps. When you see this uses, that's actually referencing a github action out there in our git of action marketplace nearly 10 000 actions from our community that give develops, as well as our ecosystem partners such as azure that we're going to take a look at in a little bit.

B

Here, though, we're checking our source code, we're actually sending python and we're actually saying which versions of python we want to. From that matrix definition. Up top I'm actually going to go ahead and install any dependencies. We have and actually go ahead and run flake eight go ahead and run some of my test files with pi test after that job. We also have this lint job and that's we're actually using some other open source actions from our community to actually run our flight 8 linger.

B

So this job is running in parallel with my build and tester scrolling down here, we then have our build skin image shop. As you can see, this is also running on an ubuntu cloud, mr runner, but this one actually has an additional parameter called needs where the lint job and the build test job have to pass before. This will actually run makes sense, because I don't want to waste compute resources. If something fails.

B

As part of my testing right, this building scan container, it's actually scanning our container image, using an azure action for container scan, we're going to look for any known abilities in our container image itself, all right, so that was one workflow file, run that we had here and going back.

B

We also had second one, and this one is actually for code ql, so code ql is github's proprietary code scanning technology that she scans your code as data, and it does this at a high level by making a database representation of your code, seeing how your code is built and then performing. We call cochlear analysis that ql is actually referring to queries that we have for the programming language. We support as.

A

Those run they're.

B

Going to find any vulnerabilities inside of your database based on definitions that we're adding to all the time the reason.

C

B

Using queries to do this is because, as a sql query right, you get a result set back from your database, the same thing here when you're looking for vulnerabilities, so it helps you find variations of vulnerabilities, and this is just code. It's actually code that is open source. So, if you want to write your own queries, you can, if you want to contribute back to our community, you can as well that's actually something you can look at and see on. Github.Com slash, github code.

B

That's our cocktail database of different queries that we're adding to every day, as you can see, we're actually getting a bunch of collaboration for our community with over 5000 plus pull requests closed and a few hundred open right now, all right so just looking through there, we had our pull request, building and testing validation, scanning container image and also just scanner application right until this python map. We want to make sure it's secured and now looking at progress again, we can actually see all of our sash checks and pass that's great.

B

We actually define some of these to be required status checks as part of our branch protections inside of github. So what that looks like those have been checked up in there. One thing, you're probably wondering is: why did we go ahead and deploy somewhere? Excuse me that could have been definitely part of our building test and scan could just add, deploy job into it, but I actually want to give some flexibility to when we decide new deployments and that's what this section here talking about review lab deployments is all about.

B

So whenever we want to make the deployments into our environment, we can actually use this keyword as part of an issue comment here in our pr now. I can do this right from the github ui, but why not take this chance to show you the additional extensible option with github, and really this will chat up flavor as part of our title.

B

Now, if I go over here, I actually have the microsoft teams extension for github installed, so what it's doing is actually feeding in content and subscribing to a feed of information come from get up, whether it be whenever I create a new issue, as I mentioned earlier at the top of the call.

B

My report quest is open, for instance, from here. I can also comment in.

B

And make changes that will be reflected inside of our quest idea.com. So I'm actually going to go ahead and use this keyword that we've defined here for making a review environment for aks.

B

We're going to submit this and now go back to github.com, let's refresh scroll down our pull request, you actually see hey. I made a comment here for microsoft teams because I get it with microsoft teams, integration once again, part of the readme file here, you'll find a reference to that. If you want to take a look and try out for yourself, let's scroll it up, what does it actually.

C

B

When we say you're making a review environment how's it going to benefit us, let's go take a look at our github actions. We actually see you get of action ran about 25 seconds ago from an issue common event.

B

Let's take a look at this. We can see. We have two jobs here and let's see how those are actually defined in our workflow file.

B

So as before, we see the on section this time. We're triggering this on an issue comment be made. You can also define certain environment file variables inside of your workflow file and here we're defining our cluster name and resource group name. These are actually coming from github secrets that you define your repository down.

A

B

Multiple jobs, the first one, is the chat ops. This is where I'm actually validating and verifying my trigger message. So here I'm saying if this keyword is found, go to continue and then go ahead and extract that string afterwards, which is the specific name space. I want to call it scroll down a little more we're getting the request.

B

Reference just make sure we point at the right place, put the right code and then our build deploy job he's actually going to go ahead and take that namespace value from our previous step, using some github action, syntax right, but just.

C

B

Since this is all in the context of workflow file, you can actually pass values through jobs, especially if they're dependent upon each other.

B

Go down here, some more pieces of automation with github actions, we're checking our code, we're actually going to log into the give container registry to store these images and we're doing that using my user account. So my actor and using the give secret for our get a workflow run scroll down a little bit more. We're gonna use a couple of other actions from docker and azure, and here we're actually deploying to our aks cluster.

B

That's where I'm providing actually my azure service principle, that's also a github secret, the resource group name and cluster name. Let's go.

A

Down here it's.

B

Actually creates the name space, create employment for us and go ahead and do that deployment itself.

B

And there's a lot of obviously some scripting writing in here and writing things with coop cuddle, but just.

C

Understand things.

B

You want to do with your ci cd or which is part of your software development. You can more or less put that inside negative action, pretty simply a repeatable pattern that you, the people that your company or team can use, there's two things.

C

I want to point out, though,.

B

I'm actually going to make a comment inside of my pull request, because I just want to update things that are in a readable way for everyone to know hey did we use them successfully or did something fail so here an additional condition we have for our step is if everything was successful, previously go ahead and run this so you're, actually using a github script to actually go and post and update a comment in cycle pr.

B

Conversely, if it fails same thing, I want to know why I failed. I want to give you that information that we can use to go trace through and debunk that all right. It looks, though, we're lucky and nothing has failed, seeing green checks here, let's go to pull request scroll down just a little bit.

B

We can see first daniel approved these changes for us earlier appreciate that we made our team comments earlier here that went ahead and triggered, and that comment we're just talking about. Let us know that the aks employment is successful, the namespace that was created so dead, review1 pr8 and the image being used as well.

B

If you scroll down here.

C

We can see this new, updated area saying.

B

That a branch was successfully deployed, I'm going to click in here. We actually click view deployment. We can actually see hey. Those are the changes we made from our data code, space right that was in the browser, but still local to our development environment in the browser itself up there in a public accessible ip address, just to verify this works looks good and just to verify. What's going on in the azure side of things, because I know a lot of smoker mirrors can occur during the demos, but let's.

A

Let's look at our azure portal.

B

Just real quick to understand here my azure portal. Look at my resource groups. I have one here for demo days for my demo test environment, cluster, here's two under name spaces, and we should actually see that name space for dev review, one pr8 that was created about three minutes ago. This seems to line up to me all right.

B

We need a lot there in the context of the pr. So let's go ahead and actually merge in this pull request and see the rest of the automation when it comes to flowing this through.

B

So we're going to merge this things are approved from data and our status checks, passing let's scroll back up and let's look at github actions tab. Once again, let's see what's available there.

B

So the first thing we can note is actually three different actions are kicking off and running all around the same time right some from a commit some from report bus itself, the first one.

A

I want to note is.

B

We see codekill.

A

B

Actually, writing again, that's.

A

Actually, by design, if you.

B

Want to be done that way, but it's not a workflow file for code ql. We actually are saying: hey, run this on any pushes to the main branch, as well as the pull request against the main branch and some logic here for the codeql workflow file. It's quite simple. It's just a couple extra lines you actually put in about. When checking out your code, you then initialize the code. Kill technology choose what languages you want to scan for, in our case, it's just python or so using that matrix strategy.

B

In case we have maybe a large project of multiple languages, we're going to.

C

Build that code.

B

And then we're going to perform our code to analysis. The cool thing about this is: if you have given security or if you're, using a public repository, you can actually set these up right from your security tab. So we have a code scan alerts. We actually want to add more scanning tools.

B

We can actually take a look and see the available options code. Develop is our technology that github develops. We add to all time for our security team, but there are additional tools for the marketplace from free and open source products, maybe container scanners, api scanners or additional infrastructures code or other tools out there maybe have use case for, and we actually recently added statcomp, which is a das solution for dynamic, app security testing.

B

If you actually just want to set up a codekill workflow, for example, you just click this button and right there. It fills it up with really the information you need to get started from the triggers to the job itself to different steps inside that workflow file.

B

All right going back here. Let's look at our other workflow files that also ran the first one we see is actually our cleanup pr. So.

C

This is something.

A

Because inside our pork.

B

Bus as you saw, we made multiple comments in there, as we did for microsoft teams earlier. They actually triggered this off a successful deployment for us, but we might do that multiple times. I don't want to eat up all those resources in my aks cluster and possibly not have room for other people to do deployments.

B

So, after a pull request is closed, I'm actually going to run this action.

B

What this action is doing is I'm running on pull request close against the main bridge, I'm grabbing that cluster name, that resource group name inside here it's actually going to extract the value from our environment.

B

It's going to go ahead and run our cube. Cuddle delete command to actually delete that name space for us and once it's done it's actually, unless nothing's been cleaned up or let us know, some failure happened actually as part of complete, because I should go back to the report. But let's just take a look and validate.

B

Support requests we open today we work on earlier. It looks like hey our pr cleanup was successful.

B

And we actually look at our azure resource group. Let's actually do a refresh.

B

I know that was real quick, but hopefully you saw that dev review environment for pr8. It's no longer there. It was gone ahead and we delete that user again of action all right. So look at workflow runs here. We have one here for actually our staging and production deployment. We.

A

Did everything inside.

B

Of code spaces from that initial issue that we opened up, we put into a pr, we made a review environment. We ran our different security scans against it, build a test against it. So now it's actually deployed up to our primary staging and product environments.

B

So opening this action right here, let's.

A

C

Talk about this get a workflow run.

B

Real quick, so what's going on here, so firstly, we're actually going to be building our image for our github container registry ghcr we're going to deploy to staging environment. We have a job to validate the staging environment.

B

Next, we're actually going to create a pre-release inside of github, so releases are something just package of the code, your project based on the sha or commit or branch we're going to deploy this to production and validate production.

B

So the first things that happened here actually look at the workflow file itself once again that on section where we have multiple different triggers in here or even using pads more because there may be some cases where, if someone modifies the workflow file, I don't want to go through that entire cic process. Again, it's just a workflow file. It's an actual code change! That's what.

A

I'm specified here making changes.

B

In my github folder scroll down once again use those environment variables for different environments inside of github.

B

We have a job here for that build image gcr here we're deploying to our staging one key thing, though, we're actually defining the environment of staging and it's url inside of our workflow here github.

A

B

Which is all we're taking a look at they allow you to break apart, specify, put rules around what jobs can occur as part of an action and what can they access and perhaps you can improve them. It's really important. If perhaps you want to date and protect your pre-pro product environments right, let's go down here, there's a lot more logic, but really similar.

B

What we did earlier when we're deployed into aks right, create a namespace or validate namespace, exist point to a secret and container registry ids we're going to deploy to aks, and here we have a job for validating staging a lot of these jobs. You could probably truncate have this as one, but here I just want to show you when the visualizations over that process flow the ghoster.

B

So this is another action for community where I'm just going to check the health.url here, I'm saying how max attempt's going to have, and what's the retry time just to make sure that that aks.

C

B

Is up and running and it's live all right so going back here to our profile, taking a look staging was deployed to let's actually click on this url all right. This seems to check out since we consist of what we did earlier. Our bake off against brownie and pizza.

B

All right and what does that actually look like across github, let's just scroll down here, we can see inside of our environments. We actually have one for staging as well, where it's actually denoted and it's tracked right throughout the process as we're deploying or making changes here throughout the day and you're going to see the laser flow was made five minutes ago. Once again, we give you the appointment from here and see our environment.

B

Next, let's go to settings and actually see what you can customize and configure when it comes to environments for here under settings we can see. First, you have our bit of secrets, which you can find.

B

We actually defined four of them earlier for aks cluster name, a resource group name our service principle, and this is actually from amazing modification to our environment as one of our workflows of clean it up and under the environments itself, you can see we have two environments, one's for staging, as you can see, there's no protection rules and one here is for production, and you can see I've actually specified a few different things.

B

Firstly, I've said who do I want to be a required reviewer before that workflow can actually run, and how many long should the weight tire be before the workflow is self deploy in case you want to add, maybe 30 minutes an hour or several hours you're going to find that right here next down here we're actually our protected branches, and can this be deployed from anywhere or should be a protected branch here we're seeing a protective branch which will apply to our main branch inside of our repository there's.

A

B

Section for environment secrets- and this is really useful- perhaps if you're trying to protect, maybe a database key or a production key, that's part of your deployment mechanism right that can only be accessed or ran if, for instance, here data goes ahead and approves this workflow for us all right. So now, we've looked at a couple things about our environments: let's go back here and actually see hey if daniel would be able to go and make an approval for us to get into our production environment.

B

Hopefully he's a time to check out and see the website, and maybe some of you have as well from the available url. That's here all right, great time in data it looks like here we can see in our events that has just been approved for our production environment and just as quickly we can see hey this workflow job is now running and went ahead and deployed up to our production instance, and now it's going to validate production just to make sure everything is looking good to access this production euro.

B

All right, let's see it looks like the same thing. The entire web app is up.

B

Rs database is also connected, go back and take a look here and see production staging we're also up and running and active in our aks.

B

Cluster navigating back up here, let's actually take a look at our different environments here on the right hand side, so here we see we have staging and production they're, both active. Once again, we can navigate here to see the history of different changes made and we.

A

Can navigate and see the most.

B

Recent deployment that's been done as well, once you get to our bake off service. That means he's actually persisting that database information from earlier.

B

Going back here to our code tab, we also create a release as part of that workflow run and, as I said about seven minutes ago, from github actions. The great thing when it comes to automating your workflows inside of github is the git of action, can actually be the context or the actor that does those sort of things from moving a project card, as we saw earlier or in this case creating a release from us.

B

I wouldn't create a pre-release environment, because once it's up and running, I actually want to go in there make the manual change and point out that hey. This is actually an official live.

A

B

Now, there's laser release of this code, the assets that were published earlier- and this is the commit that's referring back to from the changes that we made the additional thing to look at something that we were actually modified earlier.

B

Is our packages in containers right, of course, those are such an important part of deployment and cloud native technologies nowadays, so get up, gives you the option to host those locally, and when I was publishing things to the ghcr, the github container registry. It was all being pointed right here.

B

You can see your history of changes, I've made and that last published was about nine minutes ago, which I believe should line up about that seven. Eight minutes it's been since our deployments began.

B

We actually see this container image. We can pull this down.

B

You actually see the history of how many downloads it's had as well.

B

One really useful thing with git container registry and packages is the fact that we can actually define different settings and access to this package right from our github ui.

B

So here from our package settings we can see we can actually manage what repositories can access this package using github actions. Give actions are so important for pulling down images. Potion images right, so why not give other repos that you have context to inside your organization or just inside of your namespace to use? So here we see. Obviously the refill itself has permissions to write, upload and download the package, or we can change that to read for certain repositories. I want to add one I'm actually going to look at a previous repo that I have.

B

Perhaps I want this repository to have.

C

A

B

This because maybe they have some related cisd or maybe we're publishing certain content for microservices, even have all that permissioning local and part of github itself, and we.

C

Can see here, we also can see who's inheriting.

B

Permissions to this- and these are my two members of my repos daniel and phenomenal- an extra user account we have- if you want to you- also change the visibility of this. Maybe a public or private package.

B

All right, let's go back up here. Let's take a look at github, so we've.

A

Done a few things right, we've went ahead and.

B

Created a code space environment when it added make changes that we actually turn into a pull request, use issue in chat, ops using the team's integration point to see information and record and subscribe to different events occurring from deployments. Action workflow runs commits and changes made, and we hadn't accepted that and used our environment's approval mechanisms for my colleague daniel to go ahead and approve that up and be sent up into our aks cluster.

B

Now, let's go ahead and go back to our project tab and let's actually take a look. This bag look item so we actually move this from in progress, so I actually think we're moving this over to done.

B

So, let's go ahead and click in this and right here from our ui. We can actually go ahead and just close. This issue.

B

Awesome now one additional thing: that's part of github and something that we're actually currently a beta for you may have seen. Is this concept of github issues? Next, if you will so given, issues and projects have been super invaluable over the years for many companies and organizations and its users, we want to provide an option for people to take this for the next level and that's.

A

B

What our issues next initiative, but this will allow you to do actually look at issues and project boards, actually create those cross, read them and actually using different, bespoke.

B

Probably cut some fields and idles especially take a look here at a demo environment for how that would look like so pulling this up. This is our new iteration of data projects. Currently in beta, please go ahead and sign up for it. It's definitely available. If you guys want to take a look in use case and.

C

B

Of that, we could actually see over here that we can actually create several types of views of this information board views as well, based on different parameters and values, as well as creating custom content files, labels and fields from a variety of different types, selectives numbers values which gives you the flexibility and the power to work.

B

The way you need to writing for your teams and for your processes, curling beta, but check out our get a blog, because we're making some updates all the time about the product right here and we'd love to get you guys in the beta, your reactions and some of the next steps about it all right and with that sort of wrapping up our demo day for this friday september 10th. I'm james, hoggins solutions. Engineer really appreciate your time. So far. We're going to stay here on chat for a few.

A

More minutes to.

B

Answer any questions that people have had on the demo date, but hopefully be able to learn some and see some content and daniel there in chat, has shared out the demo repository where you can actually subscribe and fork this or use it for yourself as well.

B

You can see some comments in here from bart and alan allen. I think this is for any user organization that possibly wants to move different workflow processes open to get of actions right. So everyone nowadays needs to be using containers for kubernetes for something or another. So just point out, you have available marketplace actions you can leverage from azure or other ecosystem providers.

B

You can store and use your container images inside github and when it comes to just automating sort of that process of tracking items, issues and projects also that we get as well using that automation issue forms we saw earlier at the top of the call.

B

Let's hear your comment here from bart yeah. I appreciate that. Hopefully, that is more of a positive thing than anything else, but uh hopefully this and other demo days sessions that we do.

A

Weekly will also pique your.

B

Interest so make sure you all subscribe and see some of those as well.

B

Excellent all right, everyone appreciate y'all's time this week and we will see you guys next time do.

C

B

Ahead and click that qr code, if you're interested in github and want to see some more interesting features, activities or possibly talk to some. You know representatives about what that can mean for your enterprise or your organization.

C

Hey james, I have one more thing to ask: kind of. uh There was an earlier question in regards to uh chat ops, and I know that we're hearing terms like get ops chat, ops. uh There are a lot of ops that are being introduced today. You wanna uh kind of close on, uh maybe a note, just kind of giving a brief explainer when it comes to those concepts. I know we talked about using teams and github comments in order to do things on our pipeline.

B

Yeah great point data: you know that's exactly right, right issue, ops, chat, ops, infrastructure is code, so many things are there and available it really comes down to. Where is your source control? Where is your code living and.

A

B

Do you want to actually automate your different environments right so.

A

Chat, apps doesn't have to be just about.

B

Github they can be about a lot of things like azure or different environments. I think the big key in it is give bots or some automation as part of maybe your different services, whether it be slack or teams to maybe generate a new dev environment or create a new test review, rebound template, or in our case here, it's actually generating a damn environment from kind of itself based on that subscription event.

B

So here we're actually using github to kind of cut through that middleman of logic, because, instead of having a bot in their specifics, listening to keywords and doing some back-end api activity, we're saying give actions to that force as well. There's plenty of different examples of this being done with github or just other tools you want to integrate with.

B

I think that's really going to be a next evolution, just making people more productive, because when you cut the middleman out of requesting for an environment, it's just a simple chat command like how we do here at get of ourselves. We see.

C

B

Greatly increases productivity of teams and users.

C

Yeah and alan, I definitely echo your sentiment that there are so many ops and the emphasis that I think james was getting at. Is you know uh the ops? Are there or the chat bots? um Are there just to have another avenue in order to be productive? His ultimate goal.

B

All right: well, we appreciate everyone. This will be available uh on youtube as recorded later on today or tomorrow. So we'll look forward to any questions, comments or further activity. There make sure you subscribe and follow us on linkedin, twitch, twitter and other sources to get updates on future demo day sessions thanks. Everyone have a good day and good weekend.

B