Package: fai-client Version: 4.2.3 Severity: grave
The following commit from fai.git which was released as of FAI 4.2.3 seems to cause a serious regression: ,---- [ git log ] | commit 22ce409188cee3f9abb6f7396da25496474841f7 | Author: Thomas Lange <la...@informatik.uni-koeln.de> | Date: Mon Aug 25 23:35:26 2014 +0200 | | make mount --move work in jessie, Closes: #758613 `---- It executes: | # get the partition for making mount --move work | local part=$(stat -c %m -- $disk/..) | mount --make-runbindable $part which modifies the underlying device in a way that it breaks unrelated operations. For example while fai dirinstall was executed the situation was as follows: | # cat /proc/mounts | [...] | /dev/vdb /srv ext4 rw,relatime,user_xattr,acl,barrier=1,data=ordered 0 0 | /dev/vdb /srv/jenkins/jobs/grml64-full_testing/workspace/grml_chroot/grml-live/sources ext4 rw,relatime,user_xattr,acl,barrier=1,data=ordered 0 0 | tmpfs /srv/jenkins/jobs/grml64-full_testing/workspace/grml_chroot/var/lib/dpkg tmpfs rw,noatime 0 0 | [...] | /dev/vdb /srv/pbuilder/build/cow.29884/var/cache/pbuilder/build ext4 rw,relatime,user_xattr,acl,barrier=1,data=ordered 0 0 ... and then an unrelated cowbuilder run failed to execute: | I: Mounting /var/cache/pbuilder/build | mount: wrong fs type, bad option, bad superblock on /srv/pbuilder/build, | missing codepage or helper program, or other error | In some cases useful info is found in syslog - try | dmesg | tail or so Downgrading to FAI 4.2.2 seems to fix this issue. I'm not yet sure whether this '--make-runbindable' option should be executed at all, but if that's the case it should be at least considered to not execute it as such within dirinstall. Reporting as RC bug because this FAI version shouldn't reach testing as is. regards, -mika- -- To UNSUBSCRIBE, email to debian-bugs-rc-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/2014-09-01t08-16...@devnull.michael-prokop.at
