Posts

The Problem I have an Azure DevOps YAML template that does some deployment actions using PowerShell. This template is used multiple locations. The problem is that the PowerShell step within the template needs a variable number of environment variables, set from values stored in an Azure DevOps variable Group. The number of variables is not fixed because they depend on the underlying commands the PowerShell script is triggering to do the deployment. ...

I have blogged in the past about Caching NVD Vulnerability Dependency data on hosted Azure DevOps Pipeline agents. Using the cache is a great way to speed up slow builds. However, today I was surprised to find I was getting cache misses on my pipeline, even though I was sure the cache should have been hit. There are rules over how the cache is used: The cache is specific to a pipeline definition, so there is no sharing of the cache between pipeline definitions The cache is only created if the pipeline is successful (running the post run tasks) The cache only lasts 7 days but what I had not realised was the cache is also specific to the branch in a not so obvious way. My pipeline was triggered off a PR, so the cache was being created on the ‘branch’ PR #123. This was working as expected, all runs of the PR triggered build used the cache after the initial run. However, if I manually triggered pipeline run of the same branch as the PR was using, there was a cache miss. ...

There is still time to register for the free Global DevOps Experience which is being run on the 15th of June at many venues around the world, including Black Marble’s offices. The Global DevOps Experience is a day full of learning and fun, as we immerse you in the world of DevOps and AI. You will learn about the latest trends and technologies, and work in a team to solve challenging exercises based around a realistic business scenario. ...

The Issue Whilst migrating some Azure DevOps classic pipelines to templated multi-stage YAML, I hit a problem that a job running on a self-hosted agent would not start. The YAML stage, which required approval, would be queued and approved, but the agent would just sit there with the message Starting job but would never start. The strange thing was even though the job was not started, the pipeline instantly showed as failed with no error message. ...

The Issue Whilst updating a client’s Duende Identity Server from versions 6 to 7, we experienced a problem. We followed the upgrade steps and all was working fine against our development instance i.e. the Identity Server Db was upgraded to the current schema and we could login from our test MVC web client without issues. The problem occured when we started to test using our UAT (a production replica) DB. On loading the Identity Server, it did the expected EF migrations and appeared to start, but when a client tried to connect we got an exception in the Identity Server logs in the form ...

The Issue Within one of our Azure DevOps builds we today started to see the following error when running the SonarQube analysis step: ##[error][SQ] Task failed with status FAILED, Error message: Unrecoverable indexing failures: 1 errors among 1 requests. Check Elasticsearch logs for further details. Nothing obvious pointed to why this should have started to occur, the SonarQube logs showed nothing more than a longer version of the same message. ...

Background In past years, pre Covid, Black Marble hosted a venue for the Global DevOps Bootcamp, a community run in person hackathon event. After a short hiatus, I am pleased to be able to say this event has a successor, the Global DevOps Experience which is being run on the 15th of June at many venues around the world, including Black Marble’s offices. What will be involved? The Global DevOps Experience is a day full of learning and fun, as we immerse you in the world of DevOps and AI. You will learn about the latest trends and technologies, and work in a team to solve challenging exercises based around a realistic business scenario. ...

The Issue We have a number of projects that use the old style SQL Server Reporting Services (SSRS) .RPTProj project format. These projects are not supported in Visual Studio 2022 out of the box, but there is an extension in the Marketplace that adds the functionality back so you can build them in the IDE. However, we want to build our RDL files as part of our CI process, and this is where we hit a problem. When we attempt a build with MSBuild it fails with an error about missing .NET 4.0 SDK/Targeting Pack. ...

Problem One of my client’s recently had a problem that a large number of Azure DevOps work items had been updated via Excel in error. They asked if there was a means to undo these edits. Unfortunately, a feature Azure DevOps does not provide. Solution So, I wrote a PowerShell script to do it. The script… Gets a list of work items edited today by a specific user, the one who did the problematic bulk update, using a Work Item Query Get the last update of each work item and check it was made by the user who did the bulk edit, incase someone manually fixed the problematic update already. For a limited list of fields, revert the change to the value prior to the last update Save the updated work item, or if the -whatif flag is set just validate the update against the Azure DevOps instance For my client, the script worked well enough, reverting over 1000 work items in about 5 minutes. The few work items it could not revert were fixed manually. ...

Background Programmatic connection to Azure DevOps cannot be done with your Active Directory credentials. This is because this involves a dialog being shown, and these days usually an MFA check too. Historically, the solution to this problem was to enable Alternate Credentials, which could be passed as username and password, without the dialog being shown. However, the use of these has been deprecated since 2020, and they have been completely removed since Jan 2024. ...