| Table of Contents |
|---|
How to report a bug
- Figure out the steps to reproduce the bug
- If you have precise steps to reproduce, the team will be able to solve the issue much quicker!
- If you can only occasionally reproduce the bug, you should provide additional information such as logs.
- If you cannot reproduce a bug, it is unlikely it can be fixed.
- Please test with the latest released version of CPS/NCMP.
- Create a new bug ticket in the CPS team backlog.
- Be aware that CPS is an Open Source project. If you uncertain about publicly posting information such as logs, please provide Jira issue link where they may be accessed, etc.
Where to report bugs
Issues may be reported here on the open source CPS Team Jira board:
Writing steps to reproduce
| What to include in the report? | Good example | Bad example |
|---|---|---|
| Indicate whether you can reproduce the bug reliably or not. | I can reproduce using the following steps:
| Our test fails, here are the logs... |
| Precisely describe which URLs with example payloads to reproduce the issue. | When I send a POST request to http://localhost:8080/ncmp/v1/ch/id-searches with the following JSON: {"cmHandleQueryParameters":[{"conditionName":"hasAllModules","conditionParameters":[{"moduleName":"ietf-yang-types-1"}]}]} | I perform a search for a module |
| Accurately describe the expected and actual results. | Expected Results: The search takes 2 seconds to complete. | It takes a long time |
Bug Reporting Template
A bug report should contain the following information, where applicable:
- Clear description of the issue
- Clear Steps to Reproduce
- Affected version(s)
- Expected behavior
- Actual behavior
- Impact - this is important for us to prioritize tickets
- Possible work-arounds
- (Optional) Attached artifacts: Screenshots, Logs, Test data, etc.
- Alternatively, attach the corresponding client-specific Jira or other place where artifacts can be shared.
Additional information for Performance-related bugs
- Environment info should include available resources in the deployment (memory and CPU cores, number of application instances).
- What is the load on the system at the time? (how many concurrent operations)
- What is the size of data set being used? How many CM-handles?
- Measured CPU and memory consumption, if known.
- For Out Of Memory Errors (OOMEs): a heap dump from during the OOME would be useful.
- Generally, a Database SQL dump that can reproduce the issue would be ideal.
...