| Summary: |
Tests need a better way of determining if a package it's installed |
| Product: |
[QA/Testing] Runtime Testing
|
Reporter: |
Stefan Stanacar <sstncr> |
| Component: |
general | Assignee: |
Stefan Stanacar <sstncr> |
| Status: |
RESOLVED
FIXED
|
QA Contact: |
|
| Severity: |
enhancement
|
|
|
| Priority: |
Medium
|
CC: |
jessica.zhang
|
| Version: |
1.6 | |
|
| Target Milestone: |
1.6 | |
|
| Hardware: |
x86 | |
|
| OS: |
Multiple | |
|
| Whiteboard: |
|
|
OS type for building Yocto:
|
---
|
Type of Regression:
|
---
|
|
Verified:
|
|
Documentation change:
|
No (bug/feature does not impact docs)
|
| Bug Depends on: |
5410
|
|
|
| Bug Blocks: |
|
|
|
The current oeRuntimeTest.hasPackage method in meta/lib/oeqa/oetest.py relies on ${WORKDIR}/installed_pkgs.txt file (which is generated at do_rootfs with data from the package manager) for determining if a image has a certain package. We should find a better/alternative solution, since current solution presents some problems: - builds with rm_work enabled delete that file hence tests will start to fail - that file's name/location/purpose could change at any time - that file might not contain all packages (there's and extra file too) These aren't visible problems right now because we control how we run the tests, on which images and so on, but to make this task truly a test task it should be as independent on enviroment and previous outputs/operations (do_rootfs creating that file) at possible. I don't have any alternative now, but this should get fixed in 1.6 or so.