On 1/31/21 7:36 PM, Richard Purdie wrote:
[Please note: This e-mail is from an EXTERNAL e-mail address]

On Wed, 2021-01-20 at 15:05 +0800, Yu, Mingli wrote:
From: Mingli Yu <mingli...@windriver.com>

When multilib enabled, there comes below error for packages which
inherit useradd bbclass randomly when build eSDK.
  $ bitbake image-name -c populate_sdk_ext
  Exception: subprocess.CalledProcessError: Command 
'/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/lib32-recipe-sysroot/usr/bin/postinst-useradd-lib32-avahi'
 returned non-zero exit status 1.
  Subprocess output:
  
/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/recipe-sysroot-native/usr/sbin/useradd
  Running useradd commands...
  NOTE: lib32-avahi: Performing useradd with [--root 
/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/recipe-sysroot
 --system --home /run/avahi-daemon                               
--no-create-home --shell /bin/false                               --user-group 
avahi]
  useradd: Warning: missing or non-executable shell '/bin/false'
  abort()ing pseudo client by server request. See 
https://wiki.yoctoproject.org/wiki/Pseudo_Abort for more details on this.
  Check logfile: 
/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/pseudo//pseudo.log
  ERROR: lib32-avahi: useradd command did not succeed.
  ERROR: Logfile of failure stored in: 
/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/temp/log.do_sdk_depends.20567

  $ cat  
/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/pseudo//pseudo.log
  [snip]
  path mismatch [1 link]: ino 26553244 db 
'/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/recipe-sysroot/etc/group'
req 
'/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/recipe-sysroot/etc/group.33605'.
  [snip]

  Check the PSEUDO_IGNORE_PATHS in [1], the recipe-sysroot part is
  
/prj/tmp-glibc/work/corei7-32-wrsmllib32-linux/lib32-avahi/0.8-r0/recipe-sysroot.

  Use ${RECIPE_SYSROOT} for PSEUDO_IGNORE_PATHS to guarantee the
  recipe-syroot part to be 
/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/recipe-sysroot
  to fix the above issue.

  [1] 
/prj/tmp-glibc/work/intel_x86_64-wrs-linux/wrlinux-image-std/1.0-r5/lib32-recipe-sysroot/usr/bin/postinst-useradd-lib32-avahi

Signed-off-by: Mingli Yu <mingli...@windriver.com>
---
  meta/conf/bitbake.conf | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

Could you retest this with the latest pseudo upgrade in master please?

Hi Richard,

I indeed test against with the latest pseudo and the issue still exists and can 100% be reproducible in our env and this patch can resolve our issue.


I'm not convinced this is the correct fix, I suspect there is something
else going on here. We don't see this on the autobuilder and it would
be good to understand why.

It's strange that the issue doesn't occur in autobuilder and I'm still debugging the reason.

Thanks,


Cheers,

Richard

-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#149090): 
https://lists.openembedded.org/g/openembedded-core/message/149090
Mute This Topic: https://lists.openembedded.org/mt/79973790/21656
Group Owner: openembedded-core+ow...@lists.openembedded.org
Unsubscribe: https://lists.openembedded.org/g/openembedded-core/unsub 
[arch...@mail-archive.com]
-=-=-=-=-=-=-=-=-=-=-=-

Reply via email to