Version control (Git) & Deployment Pipelines in Microsoft Fabric | DP-600 EXAM PREP (4 of 12)

[@LearnMicrosoftFabric]

Did you receive free DP-600 exam voucher today? To celebrate, let's go through version control and deployment in Fabric 😊 I'd be GRATEFUL if you dropped a LIKE, or a COMMENT - let me know what you think of the series so far! Have a great weekend everyone. Hopefully two more DP-600 videos next week 🙌

[@chealey1877]

Arrived in my inbox today so looking to get it scheduled asap. Liking the series and the timing is perfect!

[@Mohit-zj8ln]

Microsoft fabric going to be a game changing tech and this series helping me alot

[@LearnMicrosoftFabric]

@@chealey1877 Awesome - great work!! 8 more videos in this series, and I plan on releasing two a week, just FYI. Should give the voucher people 3/4 weeks after the series to book their tests!

[@LearnMicrosoftFabric]

@@Mohit-zj8ln Yeh Fabric is pretty cool. I'm glad the videos are helping 👍

[@chealey1877]

@@LearnMicrosoftFabric I've done plenty of exams in the past so hoping my Power BI certification will help at least.

[@amira_369]

At first I thought you’re a Microsoft technical trainer and you’re helping people to use Fabric but I was really surprised that you’re this good and you’re not working with Microsoft! AMAZING CONTENT

[@LearnMicrosoftFabric]

😂 thanks 🙏 hahaha no, I’m just a normal guy, but if anyone in Microsoft is watching and wants to sponsor the channel, reach out 🤣

[@kevindoherty8788]

This content is so good, I feel guilty watching it for free. Thank you so much.

[@LearnMicrosoftFabric]

Haha thanks for watching! I’m glad you are finding it useful ☺️

[@Jujubeterraba]

I was thinking the same...

[@diegozanettibraga6324]

You explained the git branching process in a very clear way! Never found somebody that could do that so well as you did! Thanks a lot!!

[@TheAncientEchoTales]

Bro you're too good. I received my free voucher recently and i scheduled to write the exam on 27th of this month.....your videos are well structured and very helpful.. keep it up 🙏

[@LearnMicrosoftFabric]

Best of luck for the exam 🙌😀

[@TheAncientEchoTales]

@LearnMicrosoftFabric and thank you so much for this . Kinda feel guilty learning this for free 🖤

[@kejmil1992]

I am mostly Power BI Developer and Admin Tenant and I am feeling overwhelmed because of the material to learn to pass DP-600. Thanks to Will I have hope I will pass it, thanks a lot with this course and a lot of work you dedicated to prepare it!

[@LearnMicrosoftFabric]

Ah sorry to the hear that :/ yeh DP-600 is a tough exam which covers a lot of different areas. I think it's normal to feel a little overwhelmed - I don't think anyone is an expert in all the areas covered. It's more of a generalist exam, which requires awareness about lots of topics, but I don't think it requires deep knowledge really about any of the topics. Feel free to join our community where there is lots of people in a similar boat going through the process: skool.com/microsoft-fabric

[@cassandraedmond8442]

I must say that this series of training is so much better to understand.

[@happyheart9431]

Million Thanks. God Bless you Will. You are sharing such a valuable content free. All your videos shows how much hard work you put .

[@LearnMicrosoftFabric]

no problem, thanks for watching! ☺️

[@HimanshuRana-fv7om]

The subscribers of the channel are not justifying the content... Great work Will👍

[@LearnMicrosoftFabric]

Haha in a good way or a bad way? 🤔😂 thanks for watching! 🙌

[@SteffiRs]

I absolutely love your videos! Your explanations go beyond just the technical concepts and you make them very easy to understand. I even prefer your content to the Microsoft provided content. I can't wait for your upcoming videos to learn more so that I can use that exam voucher with confidence 💪

[@LearnMicrosoftFabric]

You got the voucher? That's awesome. I'm glad you're finding the videos useful!

[@Nalaka-Wanniarachchi]

Nice stuff.What is very cool is showing how it is connecting the pieces.

[@LearnMicrosoftFabric]

Yes, all these things are closely connected. Version control is the glue that binds everything together. Thanks for watching!

[@roxanac9383]

This is golden, thanks! Can't way to finally implement version control on my Power BI projects. I also got the free DP-600 exam voucher and I'm currently preparing so I can have it scheduled in time and hopefully pass 🤞🏻

[@LearnMicrosoftFabric]

Nice work, good luck for the exam!

[@mattroberts9665]

Another great video Will. Thank you I know exactly why you chose Power BI Reports in your Deployment Pipelines 😉. They're probably the most mature. I know we've discussed this issue on SKOOL, and still feel the same at the moment. Deployment through environments (whether using Deployment Pipelines or other methods) seems to have been poorly covered by Microsoft in their delivery of Fabric. Deployment Pipelines in particular has a long way to go before we can say it's Production ready, especially for Warehouses and parts of the data pipelines. Fabric is such an amazing platform, and i'm sure it will only improve over the coming months, but the deployment of Fabric Items is a big miss by the product team.

[@LearnMicrosoftFabric]

Thanks Matt! Yes I think it's early days in the Git/Deployment Pipelines journey for Fabric - hopefully the functionality gets improved quickly!

[@mainajnabee]

Great videos! Subscribed! Can you please explain following? So let's say we have ADO repo with Dev, Test and main branches. And we've corresponding fabric workspaces (Dev, Test and Prod for main). Now when we promote Dev fabric WS to Test WS via Deployment pipeline, are we not promoting the Dev ADO branch's code to Test WS instead of Test ADO branch's code? And by that logic, eventually the Prod fabric WS will have the Dev branch code instead of main? What am I missing? Appreciate your work!

[@georgekarlatiras9576]

I have the exams tomorrow morning and this is a last moment treasure for me!

[@LearnMicrosoftFabric]

Good luck 🙌

[@atlevr]

Regarding the Switching of branches in the workspace. We have set up Dev-stage-prod environment locked to the main branch. For each developer we have a personal Dev workspace where we create new branches based on Dev-Main. Then we merge into Dev main with pull requests.

[@LearnMicrosoftFabric]

Nice setup - I think that is what Microsoft suggest now. A few new features introduced since this video was recoded which make it possible 👍 I hope to do more content soon on different approaches pros/cons 🙌

[@abhayverma2585]

Hi @atlevr @LearnMicrosoftFabric Can one of you please help me understand this approach in detail or share where it’s published/shared on the web?

[@atlevr]

@@abhayverma2585 sorry, I do not have a reference. It’s basically only workspaces. Ws (dev), ws(stage), ws (prod). Dev is connected to the main branch of the repo, then we have ws (dev)- [developer initials] for each developer. We create a new branch based on main to attach to the developer dev workspace. One thing to consider is to split code and lakehouses into different workspaces, to separate code and data. It’s a mess when you dont separate them.

[@rajbabs9846]

15:55: Looks like you didn't address why we don't want to delete the branch "in Fabric". Can you tell why?

[@LearnMicrosoftFabric]

Ah yes, it's because if you delete the branch you can't switch back to the main branch, and eveything breaks.

[@HarisAqeel-vf2hv]

I have a copy data activity in the development workspace. That copy data activity pulls the data from my Azure Blob and then copies the exact file in my Lakehouse. In Azure Blob, I have 3 containers something like (wf-dev, wf-stage, wf-prod). In three workspaces I have assigned different containers in the source of copy data activity. Now, as soon as I create a deployment pipeline and try to deploy dev to stage to prod. All of my source of copy data activities are overwritten by dev container. I know there is a selective deployment option, but is not there a way to dynamically provide parameter. I tried doing the same copy files from azure blob to lakehouse using just notebook but the notebooks also get overwritten. I tried making it dynamic by passing parameters from the pipeline parameters but these parameters are also overwritten.

[@saniashehzad9521]

Hi Will! Just wanted to ask that at 20:28 you mentioned we have to create a new branch whenever we make changes to a PowerBI file. I dont get why that is. Cant we just do it in 1 branch like the dev-branch and then push to main?

[@LearnMicrosoftFabric]

I was talking about the scenario when we have a Branch Policy on the main branch (to protect unwanted/untested changes). Another option is to NOT have any branch policies, and just push changes directly to your branch. This is obviously easier to set up, but doesn't give the control that many companies look for

[@saniashehzad9521]

@@LearnMicrosoftFabric I see that makes sense. Thank you for clarifying!

[@massimotin]

very good explenations .. you are the one

[@owennowakowski7558]

Super helpful videos! Keep up the great work!

[@LearnMicrosoftFabric]

Glad you're finding them helpful! Thanks for watching 🙌

[@hugodasilvadasilva]

Hi Will, First I reallly would like to thank you for enabling this videos for free. At around 25min you start to show how Fabric works with Azure devops. It's a little bit confusing because the way you configured your envirement isn't the best way. What happens is that when you have a main branche protection enabled, you are suppose to have one workspace for the main branch and each developer is suppose to have it's own workspace linked to it's specific feature branch. Once the developer finished his implementation, he goes to the Azure devops and create a new Pull Request. If you don't have a main branch protection enabled, you can keep just one workspace where every developer works on.

[@LearnMicrosoftFabric]

Hi Hugo, thanks for the comment - indeed there are many ways to configure Azure DevOps and version control within Fabric. The purpose of the video (in the context of studying for the DP-600) was just to walk through the concepts of Git, version control, branching etc. In the future, I will definitely go into more detail about different architectures and approaches and the pros/ cons of each 👍

[@Karenshow]

I absolutely love this, thanks for sharing.

[@LearnMicrosoftFabric]

Really glad you liked it, thanks for watching!!

[@divyanshdhingra5893]

Hi, if we implement the incremental refresh in production workspace and do the development work in dev workspace (without incremental setup) and everytime push the changes from dev to prod, will it impact the incremental setup or we'll have to apply incremental setup again?

[@sanishthomas2858]

Nice. quick question, but for pratice question 3 , order 2 could in the 1st place as well first that that is checkout fature branch as 1st step and then clone the local repo as 2nd? is it possibe?

[@pallaviroutaray5754]

Awesome content. very very informative Will

[@LearnMicrosoftFabric]

Thank you! Glad you enjoyed 😊

[@马勒-n3y]

Another helpful video @LearnMicrosoftFabric, one question here. How can we set up the workspace? based on projects or departments, orginzations? As in Fabric we have more item than privous PowerBI

[@jafarhussain4665]

Hi, Thank you for the wonderful video and Topic. A small doubt. Rather than using .pbit File, why can't we use same .pbix file as a template to reuse our report, like creating multiple copies of template in .pibx format and reuse it ? Thank you

[@skalsolutions]

Great video Will! I have a question regarding the integration of Git and Deployment Pipelines. How would we effectively combine both features when managing three workspaces: dev, qa, and prod? Would it be appropriate to use one branch for each workspace in Git? Additionally, after deploying the report from dev to qa and prod using deployment pipelines, how would we utilize Git in QA and PROD environments? I understand that GuyInACube has already created a video demonstrating the use of Git for two workspaces, but I am uncertain about its integration with deployment pipelines.

[@LearnMicrosoftFabric]

Thanks for watching. Yes, I put a slight caveat in the description, that this just covers the basic implementation of Git and Deployment Pipelines (enough for the DP-600). In reality there are lots of different strategies for version control and deployment, some of which I'll explore in more detail on the channel in the future definitely. Be careful if you combine both approaches. In general, I would say you only want ONE method of moving an item from Dev to Test. This can be achieved either through git branching or through the manual deployment pipeline in Fabric. If you do both at the same time, you run the risk of things getting out of sync. I personally prefer the first option, because it can be somewhat automated. Hope that makes sense?

[@skalsolutions]

@@LearnMicrosoftFabric thanks a lot for your answer! Yes, make sense!

[@abhayverma-xm7kg]

amazing content! Love it!

[@tecnocriollo]

With the workspace explanation I understood why the branch is blocked at workspace level. Because you have to set a different branch for every workspace.

[@TheAncientEchoTales]

Sorry .. please quick question, did you open your read .md file in notepad and still edit it over there ?? I'm kinda lost

[@NiharikaAggarwal-b9u]

Hi since we add a reviewer earlier when making changes in policy why we do need to again it after creating pull request

[@LearnMicrosoftFabric]

When setting up the policy, I was specifying who has the ability to perform a review (this might be several people in reality). When creating the pull request, I specify a particular person who I want to review it (from the list of approved reviewers. Hope that makes sense 👍

[@georgekarlatiras9576]

Awesome !!!

[@chat-jpt]

When you are dealing with semantic model pipelines, how would you explain the difference between this and the medallion architecture? I get the latter is due to data quality/transformation. Folks tend to have a panic reaction about overhead and technical debt when hearing they need 3 layers x 3 deployment stages

[@LearnMicrosoftFabric]

Thanks for the great question! In my view, deployment stages and medallion architecture have different core goals. In medallion, we are looking to progressively improve the quality and usefulness of our datasets. Deployment pipelines are mainly used to ensure any new things you develop don't break any of the existing things you have in production. Which combination you need depends of your requirements in your org. Some will see more value in one over the other. Some will need both. Hope that helps?

[@PARAMASIVAMJ-j9t]

i have trouble in skool community after login, i got a window like update profile, after filling it there is no update or cancel button, i cant even go back to home page

[@LearnMicrosoftFabric]

Hi, I think I might've heard of something similar before. Be careful the file type of the image you are using in your profile. Retry and make sure you are using a PNG or JPG (and if one of those doesn't work, try the other one)

[@terryliu3635]

Great video! Thank you!

[@miftahindustri]

How About Data Source in git?

[@azwarmzafar]

Thank you great content as usual, a question how to get the free voucher for the exam as it’s being discussed in the chat please?

[@LearnMicrosoftFabric]

I think the free voucher deal has expired, but I think you can still get 50% off in some places - take a look at the Learn Together series of videos Microsoft produced - you should find a link in those videos 👍

[@bijalpatel5164]

Amazing Video!! I have one question: What is the correct order to work on Fabric?

[@LearnMicrosoftFabric]

Thanks! I'm not sure I understand your question... can you explain a bit more? 😊

[@ArnoWakfer]

Love your videos Will - just a suggestion. In some slides we can't see all the text as your webcam image appears over it.

[@LearnMicrosoftFabric]

Thanks Arno, I will bear that in mind going forward 🙌

[@souvanikchaudhury8890]

HI Will, I didn't able to access the note from the link. it is blocked.

[@LearnMicrosoftFabric]

I see you have now found the entrance, welcome!

[@souvanikchaudhury8890]

@@LearnMicrosoftFabric Thanks! Will how you create a fabric workspace(Synapse Data Engineering) and link it with Azuredevops? That part is not clear.

[@LearnMicrosoftFabric]

@@souvanikchaudhury8890 You create the workspace, then go to Workspace Settings > Git integration

[@souvanikchaudhury8890]

@@LearnMicrosoftFabric Got it! Thanks

[@SimonTidd]

How do I sponsor you for your content Will? It's the best I've come across in this subject area. If you are not a Microsoft MVP already, I think you will be soon. I predict you have a very bright future ahead of you.

[@LearnMicrosoftFabric]

thanks Simon, very kind of you. If you head over to the Skool community: Skool.com/Microsoft-fabric I’ll be releasing something new over there in the next few weeks if you want to support, that would be awesome

[@Joe-tk8cx]

Could you explain whats meant by semantic models, other than that great video

[@LearnMicrosoftFabric]

learn.microsoft.com/en-us/power-bi/connect-data/service-datasets-rename

[@TheRamiroB]

god job!! congrats

[@LearnMicrosoftFabric]

Thanks for watching!

[@amrosaad9730]

Thank you.

[@LearnMicrosoftFabric]

No problem, hope you found it useful 👍

[@osmanbaba1485]

Will, you smart

[@LearnMicrosoftFabric]

I'm not sure about that, but thanks 😂

[@ShresthShukla-h9n]

how old are you will?

[@carlosnavia1361]

✅

[@LearnMicrosoftFabric]

Thanks for watching Carlos!

[@HarisAqeel-vf2hv]

I have a copy data activity in the development workspace. That copy data activity pulls the data from my Azure Blob and then copies the exact file in my Lakehouse. In Azure Blob, I have 3 containers something like (wf-dev, wf-stage, wf-prod). In three workspaces I have assigned different containers in the source of copy data activity. Now, as soon as I create a deployment pipeline and try to deploy dev to stage to prod. All of my source of copy data activities are overwritten by dev container. I know there is a selective deployment option, but is not there a way to dynamically provide parameter. I tried doing the same copy files from azure blob to lakehouse using just notebook but the notebooks also get overwritten. I tried making it dynamic by passing parameters from the pipeline parameters but these parameters are also overwritten.