[Openstack-operators] diskimage-builder error

Ignazio Cassano ignaziocassano at gmail.com
Mon Jul 2 05:55:13 UTC 2018


Hi Tony,
applying the patch reported here (https://review.openstack.org/#/c/561740/)
the issue is solved..
The above path was related to another issue (distutils) but is solves also
the cleanup error.
Anycase I could unpatch the diskimage code and send you the log.
What do you think ?
Regards
Ignazio


2018-07-02 3:34 GMT+02:00 Tony Breeds <tony at bakeyournoodle.com>:

> On Sun, Jul 01, 2018 at 12:25:24PM +0200, Ignazio Cassano wrote:
> > Hi All,
> > I just installed disk-image builder on my centos 7.
> > For creating centos7 image I am using the same command used 3 o 4 months
> > ago, but wiith the last diskimage-builder installed with pip I got the
> > following error:
> >
> > +
> > /usr/share/diskimage-builder/lib/common-functions:tmpfs_check:20
> > :   '[' -r /proc/meminfo ']'
> > ++
> > /usr/share/diskimage-builder/lib/common-functions:tmpfs_check:21
> > :   awk '/^MemTotal/ { print $2 }' /proc/meminfo
> > +
> > /usr/share/diskimage-builder/lib/common-functions:tmpfs_check:21
> > :   total_kB=8157200
> > +
> > /usr/share/diskimage-builder/lib/common-functions:tmpfs_check:24
> > :   RAM_NEEDED=4
> > +
> > /usr/share/diskimage-builder/lib/common-functions:tmpfs_check:25
> > :   '[' 8157200 -lt 4194304 ']'
> > +
> > /usr/share/diskimage-builder/lib/common-functions:tmpfs_check:25
> > :   return 0
> > +
> > /usr/share/diskimage-builder/lib/common-functions:cleanup_image_dir:211
> > :   timeout 120 sh -c 'while ! sudo umount -f /tmp/dib_image.azW5Wi4F; do
> > sleep 1; done'
> > +
> > /usr/share/diskimage-builder/lib/common-functions:cleanup_image_dir:216
> > :   rm -rf --one-file-system /tmp/dib_image.azW5Wi4F
> > +
> > /usr/share/diskimage-builder/lib/img-functions:trap_cleanup:46
> > :   exit 1
> >
> >
> >
> > Anyone knows if is it a bug ?
>
> Not one I know of.  It looks like the image based install was close to
> completion, the tmpfs was unmounted but for some reason the removal of
> the (empty) directory failed.  The only reason I can think of is if a
> FS was still mounted.
>
> Are you able to send the full logs?  It's possible something failed
> earlier and we're just seeing a secondary failure.
>
> If you can supply the logs, can you supply your command line and link to
> the base image you're using?
>
> Yours Tony.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20180702/1e0635ad/attachment.html>


More information about the OpenStack-operators mailing list