Re: setup-translators breaks schroot chroot removal
On Wed, Oct 1, 2014 at 1:21 AM, Samuel Thibault wrote: > Gabriele Giacone, le Wed 10 Sep 2014 14:48:21 +0200, a écrit : >> How to fix it? on setup-translators or on schroot side? > > I'd say setup-translators. I'd say simply test whether the /dev and > /servers directories are translated. No idea about how to do that, fsysopts traverses firmlinks. Anyway I found out umount doesn't remove passive translators anymore hence replaced umount with settrans -apg in schroot patch. That fixes issue in question. Acceptable? http://darnassus.sceen.net/~gg0/02schroot.diff -- G..e -- To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/CABcaWC34fsJ+wX-3DMMGgG7=9jqmvho11a1xmi2nweeina9...@mail.gmail.com
Bug#763932: bindmounts-on-bindmounts mounted on first bindmount device/olddir
Source: hurd X-Debbugs-CC: debian-hurd@lists.debian.org On Wed, Sep 3, 2014 at 1:21 PM, Gabriele Giacone <1o5g4...@gmail.com> wrote: > sbuild/schroot suffer a bindmount bug we already talked about but not > tracked yet. Briefly, > mount -o bind foo0 foo1 > mount -o bind foo2 foo1/bar > proc/mounts displays: > /dev/hd0s1 foo1 > /dev/hd0s1 foo0/bar > instead of > /dev/hd0s1 foo1 > /dev/hd0s1 foo1/bar Filing bug in question. CHROOT_FILE_UNPACK_DIR is bindmounted on CHROOT_MOUNT_LOCATION. Hurd also needs bindmounting /dev and /servers on respectively $C_M_L/dev and $C_M_L/servers, but due to this bug, bindmounts-on-bindmounts result mounted on first bindmount source device i.e. CHROOT_FILE_UNPACK_DIR. e.g. I: 10mount: settrans -a /var/lib/schroot/mount/test00 /hurd/firmlink /srv/chroot/schroot-unpack/test00 I: 10mount: settrans -a /srv/schroot/mount/test00/dev /hurd/firmlink /dev I: 10mount: settrans -a /srv/schroot/mount/test00/servers /hurd/firmlink /servers produces /dev/hd0s7 /srv/schroot/mount/test00 /hurd/ext2fs writable,no-inherit-dir-group,sync=5 0 0 /dev/hd0s1 /srv/chroot/schroot-unpack/test00/dev ext2fs writable,no-inherit-dir-group,store-type=typed 0 0 /dev/hd0s1 /srv/chroot/schroot-unpack/test00/servers ext2fs writable,no-inherit-dir-group,store-type=typed 0 0 -- To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/cabcawc0tffj3p8n16zzrswtgegixkkrfmlberaaj5fnq-xn...@mail.gmail.com
Re: dd-schroot-cmd on hurd boxes
On Wed, Oct 1, 2014 at 1:33 AM, Samuel Thibault wrote: > Gabriele Giacone, le Fri 05 Sep 2014 03:11:45 +0200, a écrit : >> +# If CHROOT_TYPE is "file", _UNPACK_DIR is bindmounted on >> +# _MOUNT_LOCATION. Hurd does need to also bindmount /dev and >> +# /servers on respectively _MOUNT_LOCATION/dev and >> +# _MOUNT_LOCATION/servers, but due to a bug, currently >> +# bindmounts-on-bindmounts result mounted on first bindmount >> +# source device i.e. _UNPACK_DIR. > > Which bug is this about? Was it submitted somewhere? It's good to put > a URL instead of a vague "a bug" mention :) #763932, added to patch. -- G..e -- To UNSUBSCRIBE, email to debian-hurd-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/cabcawc2hhkq5syqkkww-lh1nvy1t705z3hojv_hsqj0j26+...@mail.gmail.com
