| Summary: | Multilib test is sub-optimal | ||
|---|---|---|---|
| Product: | [QA/Testing] Runtime Testing | Reporter: | Ross Burton <ross.burton> |
| Component: | general | Assignee: | Alexandru Georgescu <alexandru.c.georgescu> |
| Status: | RESOLVED FIXED | QA Contact: | Libertad <libertad.gonzalez.de.la.cruz> |
| Severity: | normal | ||
| Priority: | Medium+ | CC: | bogdanx.a.voiculescu, jose.perez.carranza |
| Version: | 1.9 | ||
| Target Milestone: | 2.0 | ||
| 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: | |||
| Bug Blocks: | 8249 | ||
|
Description
Ross Burton
2015-08-27 14:46:59 UTC
Right I now understand the point of the test - to verify that building and installing something non-trivial works. Of course connman-gnome is still a bad choice due to the long list of dependencies, in local testing it failed because lib32-bluez5 and bluez5 conflict. I've a patch that adds the generic multilib sanity test alongside fixing the connman-gnome piece now. Changing the test to use libglib2.0-utils would test the same interactions but with a lower bar to entry. Patch sent that hopefully fixes the failure. Also sent a patch to change the connman-gnome reference to libglib-2.0-utils. Thoughts? We'd need to update the autobuilder too though. Fixed in oe-core c51d24710b48eb3930edb8c661100705c6203e78. |