How to register an azure pipeline from a Github Repository using Azure Devops API
Hello! I'm working on automating Azure Pipeline registrations. (to work similarly to GitHub actions) Our scenario is: Our repositories are located on Github Yes, the service connection is there. I can use it normally from Azure DevOps Web Our pipeline definition is located in a file in the path automation/pipeline.yaml for each repository. The question is, how I can use the Azure Devops API to do it? I'm checking the documentation here: https://learn.microsoft.com/en-us/rest/api/azure/devops/pipelines/pipelines/create?view=azure-devops-rest-7.0 But I didn't find anything relevant. The descriptions of the parameters don't say much about how to configure the repository etc. Am I missing any necessary documentation? Any directions on how to do it?
Require latest target branch to be merged as PR check
Hi, we are using feature branches actively. Each branch is merged via a PR. For a to PR to be merged we run extensive build validations (e.g. build the solution, deploy it, execute it, destroy it). However, some branches take a bit of time to complete. In this case the have branched of the master several days/weeks ago. This means that in the meanwhile lots of other PRs have overtaken and the master has progressed. I want every branch to be updated with its target branch and otherwise a build validation to fail. It seems this has been requested at: - https://github.com/MicrosoftDocs/azure-devops-docs/issues/8083 - https://stackoverflow.com/questions/64029333/vsts-how-to-require-a-branch-to-be-up-to-date-before-merging-doing-pull-reques However, I am wondering why there is no documented solution today? Do I miss something? Is it overrated because in Azure DevOps PR triggered runs are actually not validating my branch but the result of a merge of my branch with the target branch? Thanks for your support Felix
Freeze column headers in Azure DevOps Sprint Taskboard
Hello, Is there any way to keep the column headers in view while scrolling down the Azure DevOps Sprint Taskboard ? 1. Open a project in Azure DevOps and select Sprints in the left menu bar 2. Go to a Sprint's taskboard and scroll down 3. Notice that the Lane headers (New, In Progress, Resolved, Closed) are no longer visible; the user needs to scroll up in order to see in which column (or lane) a task resides.Migrating Builds from TFS 2017 to DevOps Server 2022, a few questions...
Hi all, We are going to be moving to DevOps Server 2022 and our on-prem build definitions will have to be converted to the infrastructure as code, YAML format. The question(s) I have relate to getting started. Currently, with TFS, I just choose New Definition, add/configure the steps or tasks and away the build goes. With the new format, it is my understanding that the build definition is now a YAML script file that will be kept in source control. How do I get started creating this file and where do I store it in source control, or is that all automatically done with a New Definition option? Once I figure that out, I'll be recreating our Definitions with the Classic option for task configuration until I'm up to speed with YAML to script on the fly. Any information or help is appreciated! Thanks!!
Non-SaaS Product GIT Branching Strategy
Dear Team, What’s your recommended approach? A non-SaaS product Two repos - Backend and Frontend Current Approach - Dev, QA and Prod Branches Sprint branch (can’t go with feature branch as multiple unlimited APIs and multiple user stories will impact the same set of APIs) created out of Dev and merged into Dev at end of sprint Post each Sprint Dev branch tagged and PR into QA branch Customers are given Docker images generated out of specific tags from QA branch Now comes the fun part- Say customer 1 on Tag v4.3.0, customer 2 on Tag v4.4.0 and product last release is Tag v4.5.0. Current active sprint once complete would be v4.6.0 Developers currently working on active sprint branch for v4.6.0 Bug 1 reported by customer 1 in v4.3.0 Bug 2 reported by customer 2 in v4.4.0 I can extract the specific tag code, make the changes, then manually make the changes in other tags and release to those customers if common bug, or else manually make the changes in active branch also so that next release it’s not missed What if Bug 1 is a Feature for customer 2 who doesn’t need it? So where will I store these changes? Which branch? I want to avoid having customer specific branches as it becomes a big overhead. Suggestions welcome!
Modifying Release Retention Policy via the API in Azure DevOps Server
Hi, Is there a way to adjust the Release Retention in the Project Settings, specifically the Maximum retention policy and Default retention policy via the API in Azure DevOps Server? I'm looking to automate this process but am struggling to find clear documentation or references on the appropriate API endpoints or methods.
Query DevOps for Feature duration
We're using the Feature Timeline to create a high-level planning of Features across Iterations. Some are single-iteration features, but some span multiple Iterations. I want to get the Feature information into Excel with the Team addin, including the beginning and ending iteration. However, when I query, I get a single Iteration for each Feature. Which Iteration is returned seems to depend on the first Iteration where the Feature was planned. So I cannot see from the query data whether a Feature spans multiple iterations, and not even if the Iteration returned with the Feature is the first or last Iteration the Feature is planned in. I've included all available Date/Duration/remotely relevant columns to see if I can find out if the information I need is available somewhere. Apparently not: all Date columns except Created/Changed are empty, and there's nothing Duration-like. Since the Feature Timeline 'knows' that some features span multiple Iterations, the information must be stored somewhere. But WHERE? Screenshots included: Feature 26417 in the Feature Timeline, spanning sprints 6 and 7; and the same Feature in a Team query in Excel, showing no date information and only the Iteration Path to sprint 6.
Function App access restrictions preventing ADO cloud Pipeline to Government tenant from succeeding
Hi there, I'm running into an access restrictions issue with Function App on our Gov tenant thats preventing the deployment from succeeding from our commercial cloud ADO. It seems that ADO cloud and / or the agents IP's need to be defined in the function app access restrictions, but unclear where to gather that IP info, or add an additional task on the release side. I read some other forums that reflected that noted the AgentCloud service tag should suffice, however that doesn't work either. Our service connection deploys app service code to the same RG the function app resides in just fine. Any feedback how solve this issue is greatly appreciated.Azure Devops Newbie Question
Hi all, I am setting up a test project with Agile process type and I am looking to view Epics, Features, User Stories and Tasks all within the same board however I am struggling with this one. If I set everything up on the backlog view I can expand them out and see everything fine. However when I select "View as Board" I dont get the view I am looking for (or maybe it is just not possible?). Within the boards views, on the right side I have the option to view Epics, Features and User stories but I can only select one at a time. If I select Epics I see Epics and Features. If I select Features I see Features and User Stories and if I select User Stories I see User Stories and Tasks. Is there a way to see all of them in one single view within the board? Any ideas are most welcome! Thanks
How should I publish my distributable release files in a release pipeline?
Hello community, I have been looking up quite some time on the subject of pipelines, but strangely, I cannot get to find an answer for this very simple scenario I have. Thing is, the releases to our software need to be published as a easily downloadable and versioned file/s or installer. Why? Because due to the nature of the bussiness, target environments do not have internet connection, so we need to offer manually downloadable distributable files/installer to manually install them. I have tried different things but these don't seem to work for me: - Publish as a "Pipeline Artifact": This option doesn't work in classic Release Pipelines (it throws an error "Cannot upload to a pipeline artifact from release environment."). - Publish as a "Build artifact": If I do it in a build pipeline, it will defeat the purpose of separation between "build" and "release" pipelines. Even if I did it anyway, it will be published as a "build" artifact, which will be very hard to reach in the Azure Devops web app: Pipelines>(Pipeline name)>(browse run name)>Artifacts>Artifact name, which also gets deleted like 30 days after. - Publish with "Universal Publish": This option seems to be very apt, since releases are added into the Azure Artifacts feed with a corresponding version number in a very "official" manner (i yet have to see how to assign this version number dynamically from build stage). But the problem is that apparently it's not possible to download universal packages from Azures Web App UI for technical reasons. Instead, I have to use Azure CLI. But i need to be able for me or any team member to download this artifact easily and without installing additional software or technical knowledge. Am I missing something here? Some considerations: we only have an Azure Devops plan, no other Azure services (storage, etc).
