On Tue, 9 Aug 2022 at 15:49, Alban Bedel via lists.openembedded.org
wrote:
> > How are you corrupting sstate files? They're moved into place
> > atomically at the end of processing so this should be really hard to
> > do.
>
> We are still hunting for this problem, but as it happen on a shared buil
ë]úçݺ `
Q ëN÷ëß±vm:Nâr«v«v'¥{±~§u«b¢z+¸
ªÝ¦êÝébì_¢éÝjبࢧzfÞu×rÞ+-²^éy×^vࢧzfÞu×rÞ+-²^éy×^vàJæãyËQxá ¢·0²ËZµàèÖ¢»b~)쵩ex²ËZµêZrF
yö¢ÀÚµäî{Ð.m6Û]øó~~ÓM4
íz{Kjx.jD¢{^Ôò¥ë^Æß騧r«±ë
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to th
Hi Alban,
Alban Bedel via lists.openembedded.org escreveu no dia terça, 9/08/2022 à(s)
09:36:
> Currently if the installation of a sstate package fails the whole
> build is aborted although bitbake could just fallback on running the
> real task.
>
> To fix this issue we replace the call to bb.fa
On Tue, 2022-08-09 at 10:35 +0200, Alban Bedel via
lists.openembedded.org wrote:
> Currently if the installation of a sstate package fails the whole
> build is aborted although bitbake could just fallback on running the
> real task.
>
> To fix this issue we replace the call to bb.fatal() in
> ssta
Apologies, but no. This fixes the symptoms, not the problem, which is
that your infrastructure is malfunctioning. Why does it supply broken
artifacts?
Alex
On Tue, 9 Aug 2022 at 10:36, Alban Bedel via lists.openembedded.org
wrote:
>
> Currently if the installation of a sstate package fails the w
Currently if the installation of a sstate package fails the whole
build is aborted although bitbake could just fallback on running the
real task.
To fix this issue we replace the call to bb.fatal() in
sstate_setscene() with a simple sys.exit() to pass the error to bitbake
without aborting the buil
