[oe] concurrency issue at hand?

[Leon Woestenberg]

Hello,

during the last few weeks I have run into erroneous builds against
.dev and bitbake head. (80% of all my builds fail).

MACHINE=at32stk1002, DISTRO=angstrom-2008.1, uclibc
MACHINE=mpc8313e-rdb, DISTRO=angstrom-2008.1, both glibc or eglibc

which reproduced, so I have been trying to understand the build errors
at the package level, going back and forth between known-good and
known-bad revisions, etc.

However, building this on different hosts, seeing different behaviour,
 and now with concurrency disabled and enabled, I think we have a
concurrency issue at hand.

Sometimes restarting the build helps, sometimes not.

I strongly suspect threaded bitbakes in combination with packaged-staging.

I just wondered if anyone else is seeing the same?

Regards,
-- 
Leon