[yocto] How to deal with patch failures due to new source?

Burton, Ross ross.burton at intel.com
Tue Feb 23 15:31:59 PST 2016


On 23 February 2016 at 20:16, Paul D. DeRocco <pderocco at ix.netcom.com>
wrote:

> All of a sudden, I'm getting massive patch failures in OE's Samba 4
> package. (Last night it was fine.) Looking at the main log, the
> samba-4.1.12-r0 recipe did a do_fetch, do_unpack and do_patch. This is a
> huge patch, a megabyte in size, applying to 525 files, and it now has 31
> failures, because the patch is from last September and wasn't updated.
>

If OE didn't change the recipe, and none of your other layers changed, then
what *did* change?  The tarball upstream can't have changed as the
checksums wouldn't have matched.

Try doing bitbake samba -cclean to wipe away the local build state and
rebuilding it, this will verify that it's not some local state causing
problems.  What can happen is that bitbake wants to do a rebuild so re-runs
steps and ends up attempting to patch generated files and the patch fails
to apply (in this case the patch is broken).

Probably worth posting to openembedded-devel@ too as samba is part of
meta-oe.

Ross
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.yoctoproject.org/pipermail/yocto/attachments/20160223/e9218b17/attachment.html>


More information about the yocto mailing list