| Summary: | sysvinit reproducibility issue (shutdown user leakage) | ||
|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Richard Purdie <richard.purdie> |
| Component: | core | Assignee: | Richard Purdie <richard.purdie> |
| Status: | RESOLVED FIXED | QA Contact: | |
| Severity: | normal | ||
| Priority: | High | CC: | meta.mr.watcher, meta.watcher, randy.macleod |
| Version: | 3.0 | ||
| Target Milestone: | 3.1 M3 | ||
| Hardware: | x86 | ||
| OS: | Multiple | ||
| Whiteboard: | |||
| OS type for building Yocto: | --- | Type of Regression: | --- |
| Verified: | Documentation change: | No (bug/feature does not impact docs) | |
|
Description
Richard Purdie
2020-02-03 12:27:03 UTC
Extra data so far: * This group ownership is performed in the recipe's do_install. * The shutdown group is defined in etc/group in base-passwd's group file. * There is a DEPENDS on base-passwd. If we could create a QA test to make the build fail then we can run that in master-next. I found how to reproduce: bitbake sysvinit bitbake sysvinit -c clean bitbake sysvinit -c package_write_ipk -f which changes the sysvinit package from having two files with owner/group "root/shutdown" to "root/70". The reason is the base-passwd files are not restored to the sysroot when do_package is setscened so there is no information to resolve "70" to a name whe building the ipk. I think adding do_package_setscene[depends] = "base-passwd:do_populate_sysroot" to the recipe fixes this. This does have wider implications unfortunately. |