In my current organization we are using Azure DevOps Server (on-prem) to release our products. The current setup is that we have a bunch of build and release agents running on a set of VMs.
The servers that we actually release to are different machines than the ones running the release and build agents, therfor we end up using alot of "PowerShell on target Machine" tasks during the release to configure and setup dependencies for our products.(asp.net websites in this case)..
However.. what I find strange is that due to this setup we cant really utilize other tasks to setup our environment/target-machines. For instance, lets say that we would like to use the "extract files"-task, then this would not be possible since the extraction would happen in our agent and not in our environment/target Machine.
Are we missing something or are you actually supposed to basically only use the "PowerShell on target-machine"-task for such a scenario?
You are right, most of the tasks are running on the machine of where the agent is installed. if you want to run the agent in one VM and deploy with this agent to another VM you can use out-of-the-box with the PowerShell on target machine or Copy files to remote machine, or SSH tasks.
But, in the marketplace you can find many tasks that can do things to remote machines, include zip or unzip files.
If you can it's really recommended to install the agents on your target machines.
Related
I have a requirement to integrate the JMeter scripts, checked-in a Git repository, with a DevOps pipeline so that I can run the JMeter scripts using a specific VM in Azure.
Basically, I should have all my jmxs and csvs in a git repository and when I run the pipeline, having a parameter of the script name, it should run the script on a specific VM (not with a static IP) and copy the jtl in some storage.
What is the best way to achieve this?
With a DevOps pipeline so that I can run the JMeter scripts using a
specific VM in Azure. What is the best way to achieve this?
If the specific VM exists before the current pipeline, you can consider installing self-hosted agent there.
To do CI/CD using Azure pipelines, we need at least one agent. If we use microsoft-hosted agent, it will provide one fresh VM for us to run jobs. Since you need to run the script in your own specific VM, I suggest using self-hosted agent. You can follow the steps here to install one agent into your own VM. (The steps are quite easy and only cost several minutes)
After making your VM a self-hosted agent, the pipeline will call your VM to run the jobs. Now your original issue turns into how to run JMeter locally with command-line. See similar issues here: Five Ways To Launch a JMeter Test without Using the JMeter GUI and Run .jmx file through command line ....
1.So now we can use a command-line task in pipeline to run the JMeter related commands shared in the similar topics above. And these jobs are done in your specific VM.
2.I'm not sure which location you want to copy the jtl to, but you can use Azure File Copy task to copy files to Microsoft Azure storage blobs or virtual machines (VMs). Or a simple copy/xcopy command in your command line task to copy files to another location in same machine. (Specific VM)
Hope all above helps :)
I have Use following Task in Azure CD pipeline.
"Run Taurus" Task is as following.
Where "_WM WebClient TestArtifacts" is git/Azure Repo directory where .jmx file kept(in Code).
I have multiple ASP.NET web applications which I want to release into different VMs. Manually installing a DevOps agent in every single VM seems really inefficient. Is there a way to make this process faster? Is it possible to create release pipeline that could directly push the code to the public ip of the VM?
As workaround, you can prepare scripts to register each new agent. Here you can find parameters: Self-hosted Windows agents - Unattended config.
If you deploy your web application with IIS deployment task,
You can have a try using task Manage IIS, which can create website in a remote machine.
And then you can add a Windows machine file copy task to copy the build artifacts to the website Physical path in the remote machine.
Another workaround is that you can manage IIS with powershell script. So that you can add task PowerShell on target machines to run powershell script to manage IIS website. You can refer the example scripts at this page, and this page. For more information about IIS powrshell command you can refer here
I am getting my first VSTS build and release definitions setup.
Test server destination, on prem
project is using VS 2017 professional
Earlier I setup a private agent for doing the build, following this.
https://learn.microsoft.com/en-us/vsts/build-release/actions/agents/v2-windows?view=vsts
The agent installed fine and I can see it from build & release / agent pools
However, have now switched to using a hosted agent for that so I can remove the need to install visual studio on a test server.
Now I want to deploy the build to my test server, which is on premises. Do I need yet another agent or can I use the one I setup already. In other words are build agents separate from release agents? The agent I setup earlier is running as a service.
The build agent can be used in VSTS build and release, also the deployment group agent is configuring with build agent too.
You don’t need to setup another build/release agent, there are many tasks to deploy app to another server, such as IIS Web App Deployment Using WinRM, PowerShell on Target machine, Windows Machine File Copy.
The build and release agents are the same, but you have to install an on-premise agent to deploy your projects in your network. Is not that you can not use the hosted agent, but you would have to use WinRM and open you network, which may not be a good idea.
After you installed your on-premise agent, you don't have to install VS on the machine to run tests. You can add the following task to your release.
It doesn't look like this is possible but is there a way to use VSTS Continuous Delivery and deploy to a server outside Azure?
Sure. VSTS Continuous Delivery can deploy to anything that can run the VSTS agent or anything that can be targetted from a VSTS agent. Basically you have two options:
Let the VSTS (hosted) agent communicate directly to your deployment target. Out of the box tasks can use MsDeploy or WinRM. There are also tasks for Octopus Deploy, AWS and other systems available through the Marketplace. Powershell has options to connect directly to other systems and you can use many different protocols from a powershell script.
Install the VSTS agent on your target machine (or network). The VSTS agent can be downloaded and installed onto the target server. That way you have full local access to the target machine and can simply copy the files to the target folder straight from the Release Management artefacts. By installing a VSTS agent in you local network, you can use the options listed under 1 to bypass firewalls that might be blocking the hosted agent.
I have a Windows VM that hosts a VSTS build agent. Due to the number and length of builds that are running I would like to know whether multiple build agents can be hosted on one computer? That would allow a dedicated agent for slow builds, and a dedicated agent for quick builds.
https://www.visualstudio.com/en-us/docs/build/admin/agents/v2-windows
Yes you can run multiple agents in a single VM.
Make two directories say Agent1 and Agent2, extract the agent in each one of them and configure them with different names against your VSTS/TFS account.
It should work out of the box.
We run 4 agent jobs per machine concurrently with no issues. As mentioned above, should work out of the box. Just make sure you clean up directories. We have a script to do it every night
Yes, this works, I did the following:
Created a PAT for agent installation needs
Downloaded agent binaries from the agent creation page
Unpacked the archive contents into 2 different directories ("c:\ado-build-agents\agent1" and "c:\ado-build-agents\agent2")
Ran "config.cmd" and followed configuration instructions, provided by it.
Updated pipelines to build the agent pool, which those agents reside in ("Default" in my case)
To test the setup - triggered all 15 pipelines, that I had. As the result I was able to see two pipelines running at the same time, while others were in the "Queued" state (according to my expectations).
I will be also testing out how resources are consumed by the agents to try to understand if I should deploy more agents on the build machine.