If you look at the log.do_rootfs it will give you more or less the exact failure(using the message below as a guide), the error message below is basically the final catchall that says "at least one of them failed, please go fix it" kind of message.

On 2/14/2020 10:30 AM, Andrew Boos wrote:
Thanks Jeremy for the hint about update-alternatives. It turns out that lmbench was installing a binary called "stream" that conflicts with ImageMagick's own stream. All we had to do was remove lmbench from our IMAGE_INSTALL and the problem went away. I'm posting the original error for anyone that is having the same issue.

ERROR: imx8-image-3.0b3-r0 do_rootfs: Postinstall scriptlets of ['imagemagick'] have failed. If the intention is to defer them to first boot,
then please place them into pkg_postinst_ontarget_${PN} ().
Deferring to first boot via 'exit 1' is no longer supported.


On Thu, Feb 13, 2020 at 8:02 PM Jeremy A. Puhlman <jpuhl...@mvista.com <mailto:jpuhl...@mvista.com>> wrote:

    In these kind of cases, by providing the actual error message,
    people can provide more help.

    The recipe inherits update-alternatives and sets a large number of
    alternatives. The alternatives are done in the automatically added
    post install scripts. In all likelihood one of the listed links is
    wrong, or already a symlink or something similar.
    On 2/13/2020 4:49 PM, Andrew Boos wrote:

    I am working on a bitbake build and I am running into an issue
    where a postinst scriptlet is failing via "exit 1" to defer to
    first boot and I cannot figure out which recipe is causing the
    issue. Does anybody have any tips on identifying which recipe is
    actually exiting with a failure? The error log says that it is in
    imagemagick but the recipe does not have any postinst scripts
    defined. Also, there are no imagemagick .bbappend files in my


        Virus-free. www.avg.com


-- Jeremy A. Puhlman
    jpuhl...@mvista.com  <mailto:jpuhl...@mvista.com>

*Andrew Boos* | Avilution, LLC
(c) 251-895-5509

Jeremy A. Puhlman

Openembedded-core mailing list

Reply via email to