| Summary: | clutter-gst qemu gtkdoc failure | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | [Build System, Metadata & Runtime] OE-Core | Reporter: | Richard Purdie <richard.purdie> | ||||||
| Component: | multimedia | Assignee: | Unassigned <unassigned> | ||||||
| Status: | RESOLVED FIXED | QA Contact: | |||||||
| Severity: | normal | ||||||||
| Priority: | Medium+ | CC: | alex.kanavin, meta.mr.watcher, meta.watcher, randy.macleod | ||||||
| Version: | 3.2 | ||||||||
| Target Milestone: | 3.3 M1 | ||||||||
| Hardware: | x86 | ||||||||
| OS: | Multiple | ||||||||
| Whiteboard: | AB-INT | ||||||||
| OS type for building Yocto: | --- | Type of Regression: | --- | ||||||
| Verified: | Documentation change: | No (bug/feature does not impact docs) | |||||||
| Attachments: |
|
||||||||
|
Description
Richard Purdie
2020-09-29 02:43:29 UTC
I did log into the fedora32-ty-1 worker pulled the build directory back and ran "bitbake clutter-gst-3.0" on this build. It worked. It suggests there was some kind of transient race issue. I'll attach the failure and success logs Created attachment 4731 [details]
Original failure log
Created attachment 4732 [details]
'success' log
Note that whilst the build succeeded, the same error is present in the second log. Bad error handling? Also worth noting this is a build with api-documentation enabled in DISTRO_FEATURES in case that isn't clear. /usr/libexec/gstreamer-1.0/gst-plugin-scanner: symbol lookup error: /home/pokybuild/yocto-worker/qemux86-world-alt/build/build/tmp/work/core2-64-poky-linux/clutter-gst-3.0/3.0.27-r0/recipe-sysroot//lib64/libc.so.6: undefined symbol: _dl_fatal_printf, version GLIBC_PRIVATE ^^^ looks like something in the build is trying to run the gst plugin scanner from the host against target libc. I am looking into it. On second thought, that might not be the actual error, as it's present in a successful run too, and the transient gtkdoc binary may be failing without diagnostics. I ran this a few times, and out of ideas at this point. If the failure happens again, save the build directory but do not touch or rebuild anything inside, as something is wrong in the transient gtkdoc introspection binary. Thanks Alex, if we see it again I'll try and preserve the build dir. I should know better than trying to rerun it! |