Project path and agent name information #26
Conversation
| @@ -64,7 +67,8 @@ public boolean hasReport() { | |||
| .addField(BUILD_STATUS_MESSAGE, build.getBuildStatusSummary().message) | |||
| .addField(BUILD_RESULT, result) | |||
| .addField(BUILD_RESULT_ORDINAL, ordinal) | |||
| .addField(BUILD_IS_SUCCESSFUL, ordinal < 2 ? true : false) | |||
| .addField(BUILD_IS_SUCCESSFUL, ordinal < 2) | |||
There was a problem hiding this comment.
Just a simplification of the code
|
Hi, |
|
Done :) |
|
Hi, While testing in production and exploiting those metrics in Grafana, I realised that the agent name should actually be a tag, and not a field. |
|
I'll soon append a correction. |
| @@ -24,6 +26,7 @@ public AbstractPointGenerator(MeasurementRenderer projectNameRenderer) { | |||
| Point.Builder builder = Point | |||
| .measurement(name) | |||
| .addField(PROJECT_NAME, renderedProjectName) | |||
| .addField(PROJECT_PATH, build.getParent().getRelativeNameFrom(Jenkins.getInstance())) | |||
There was a problem hiding this comment.
An even better alternative could be to remove PROJECT_PATH (tag and field) completely and add a tag like PARENT which contains the parent folder like folder1/subfolder5 of a job with path folder1/subfolder5/job2.
With this more normalized schema it would be possible to filter and group by the parent (which could be for example folders for different areas in Jenkins or the main folder/job of a multi-branch job where each branch / pull request / tag is a separate job).
What's your opinion on this @asimell and @dcoraboeuf?
This PR adds:
Best regards,
Damien