I'm currently editing the Jenkins' job description using a function in my jenkinsfile:
currentBuild.description = "${get_user_name()} bla bla"
I want, after the build is done, to read a pytest html file (which is generated during the job) and extract number of passed / failed tests, and add that data to the description.
The extraction itself is already done in a python script I have, however, I'm not sure how can I add that data to the description.
One way I thought of was an http request to the jenkins web API, but that seems a bit convoluted, and specifically, I had some problems doing that in the past.
Is there a way to make it simple using http, or using another (built-in) way?
Related
I have a Jenkins pipeline defined that includes an input step. A human can provide the input by clicking in the Jenkins UI and there is an HTTP endpoint to provide input as well.
Is it possible to provide the input via Groovy API calls? For instance, could a parallel step in the same pipeline provide the input values? Or, could a completely different build provide input values via Groovy code?
The reason I'd like to use Groovy is to keep the input providing entirely in the Jenkins system and avoid having to provide authentication credentials for the HTTP endpoint.
We had a similar problem (one pipeline should be able to trigger input steps in other pipelines).
This worked in the Jenkins script console and should work in a pipeline:
import org.jenkinsci.plugins.workflow.support.steps.input.*
def build = Jenkins.instance.getItemByFullName("TestInputPipeline").getLastBuild()
def action = build.getAction(InputAction.class)
action.getExecutions().get(0).proceed([])
TestInputPipeline is the name of a test pipeline with a single input.
If your input has parameters, you will probably be able to provide them with the map in the proceed call.
This Input Step Plugin test code helped us: https://github.com/jenkinsci/pipeline-input-step-plugin/blob/master/src/test/java/org/jenkinsci/plugins/workflow/support/steps/input/InputStepRestartTest.java
JavaDoc can be found here: https://javadoc.jenkins.io/plugin/pipeline-input-step/index.html
When I create a jenkins job via the API, I use a previous job's config.xml, make my modifications and then make the POST call to create the job.
My questions is, is there a way to generate this programmatically? I.e. is there a structure of a config.xml, what XML entities it should have, what values, etc so I can write a small module to generate one and send it to the jenkins API call?
I don't think there's any mandatory XML entities. Submitting an empty structure should result in a job that has default values for all settings.
What you want to do is exactly what's done by the Jenkins Job Builder. It provides a YAML-based framework for creating Job configuration XML files and submitting them to Jenkins. It's a common alternative to the Job DSL plugin. I wouldn't recommend to re-implement such a solution yourself -- handling all the plugin-specific XML configuration parts will be a nightmare.
We create our jobs using Job DSL plugin. You can try the playground http://job-dsl.herokuapp.com/.
At first, it seems that it is hard to learn, but after the first seed job, it is much better.
When we started writing our scripts we were afraid that there will not be suitable API methods for our needs. It turned out that we had one such case, which was solved using the configure block.
Get started guide here.
I have JMeter setup on Jenkins. I am running some performance test with it. On Jenkins, I have setup the TestLink plugin.
On JMeter, I have 'View Results Tree' tree setup, and export the results to a xml file. With the results with the following format:
<testResults version="1.2">
<httpSample t="81" lt="81" ts="1463588977147" s="true" lb="Unit_EdgeSpanTest_transmission-submit_Case1" rc="200" rm="OK" tn="CinemaCMSAPITest 1-1" dt="text" by="382">
<assertionResult>
<name>Response Assertion-ResponseCode</name>
<failure>false</failure>
<error>false</error>
</assertionResult>
<assertionResult>
<name>Response Assertion-TextResponse</name>
<failure>false</failure>
<error>false</error>
</assertionResult>
</httpSample>
Now I have problem get JMeter results report back to testLink. Googled, not much resource out there tell me what to do. I could only find one resource from Jenkins:https://wiki.jenkins-ci.org/display/JENKINS/Integrating+TestLink++Jenkins++JMeter
But it did not help much. I am not using Ant...
Does anyone has successfully made this work? And could give me some hint of what to do?
Thank you very much in advance!
I recall I had to submit JMeter results to TestLink some time ago. In my case each JMeter sampler represented an individual test case so I used the Beanshell Assertion and TestLink Java API Client combination. If you're not familiar with Beanshell - check out How to Use BeanShell: JMeter's Favorite Built-in Component
If you're not comfortable with coding I guess you can access the API endpoints using HTTP Request or SOAP/XML-RPC Request samplers.
There is also a number of discussions regarding the API on TestLink forum
Hope this helps.
I followed the instruction described in the question to configure the testLink, JMeter, and Jenkins. Just with one problem: I first tried to use the script from the link above, but it seems always returned the wrong result. I wrote a simple java script to transform JMeter XML test result file to TAP format. Except that the link from the question is a good instruction to follow to setup the integration between JMeter/Jenkins/TestLink:
To summarize:
From Jenkins:
install TestLink plugin
Configure TestLink from Jenkins > Configuration
Create a project > Configure:
Invoke TestLink: Put the TestLink Version/Test Project Name/Test Plan Name/Build Name/Custom Fields/Test Plan Custom Fields according to the settings from the TestLink.
Add build action to run the java function/shell script
Result Seeking Strategy: Add the tap directory in 'Include Pattern' Add JMeterTestCase to Key Custom Field.
From TestLink:
Create custom Fields,
Add the JMeterTestCase custom field to test specs. And set the test specs to 'Automated'
Assign the test specs/cases to the according test plan.
From JMeter:
Make sure the test result listener generates xml format result.
Have a script transform JMeter Test result from xml format to Tap format.
In the link from the question, it has a script for doing that, and I also attached a simple script wrote in Java for transforming JMeter Result xml file to TAP file in java:
https://github.com/yueran/jmeterResult_xmlToTap
I have three Jenkins projects (Maven jobs, to be exact) that have many available parameters. I've set up a way to automate running these weekly with a set of parameters using the Build Flow plugin so that I end up running a total of 12 builds. I also have a follow-up project that exists only to send an email containing some information about the builds that ran, including a link to all 12 of the artifacts.
My problem is figuring out a way to get the artifact names from all 12 jobs. I think the best way to do that is using the REST API, but I could use some help with that.
I am using the Editable Email Notification plugin, and I have access to the job name and build number. I probably just need a script to grab the artifact name from each job.
Does anyone have experience with this?
I haven't tried the exact case you want, but you should be able to use the depth and tree parameters of the REST API to narrow down the data you need. Start with the API URL for one of the builds, which will be of the form
http://jenkins/job/jobname/buildnumber/api/json
My question may be silly but I've been trying several ways and I still can't do what I want, i.e.:
use the scp target of Ant to target a remote machine and execute
a script there
this script creates a dynamic list of files
get this list of files (only their names) back in Hudson to use it in the next build step (another scp from Ant)
I tried to use environment variables but they are interpreted by Hudson so I'm stuck here...
Globally my question would be: how to get a result from an Ant build step ?
Thanks for your ideas,
Emmanuel
You may find File parameter useful. This allows you to create an input file, pass it to build. You may need to write script/ant script to process the file though.
In the long term you may evaluate a Hudson farm. This will allow to create tasks that span multiple machines , pass results around. (https://wiki.jenkins-ci.org/display/JENKINS/Plugins)
You can get the ID(s) of the job that triggered your job via the API and fetch their status.