Hi Daniel, On Sun, 23 Sep 2018 13:21:38 +0200 Johannes Schauer <jo...@debian.org> wrote: > Quoting Daniel Dehennin (2018-09-22 19:52:55) > > I'm trying to setup an LVM schroot but I got the following error: > > > > #+begin_src > > /srv/chroot/sid-amd64-sbuild is not empty at /usr/bin/sbuild-createchroot > > line 279, <PH> line 13. > > #+end_src > > > > The problem is that sbuild-createchroot check for 3 directory entries > > without > > taking âlost+foundâ as an exception. > > the reason why sbuild requires an empty target directory, are debootstrap bugs > like these that make you loose everything in case a wrong code path is taken: > > https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=833525 > > Now imagine that you actually had some previously recovered files in > lost+found. If you are unlucky, those would then be deleted by a faulty > debootstrap. > > Why don't you just rmdir the directory? Then you also made sure that there is > indeed nothing inside that you still need. fsck will re-create it in case it > needs it.
as I thought more about this problem I now found a better solution that makes both of us happy. Consider this diff: - opendir(my $dh, $target) or die "Can't opendir($target): $!\n"; - # Attempt reading the directory thrice. If the third time succeeds, then it - # has more entries than just "." and ".." and must thus not be empty. - readdir $dh; - readdir $dh; - die "$target is not empty" if (readdir $dh); + # check if the directory is empty or contains nothing more than an + # empty lost+found directory. The latter exists on freshly created + # ext3 and ext4 partitions. + # rationale for requiring an empty directory: https://bugs.debian.org/833525 + opendir(my $dh, $target) or die "Can't opendir($target): $!"; + while (my $entry = readdir $dh) { + # skip the "." and ".." entries + next if $entry eq "."; + next if $entry eq ".."; + # if the entry is a directory named "lost+found" then skip it + # if it's empty + if ($entry eq "lost+found" and -d "$target/$entry") { + opendir(my $dh2, "$target/$entry"); + # Attempt reading the directory thrice. If the third time + # succeeds, then it has more entries than just "." and ".." + # and must thus not be empty. + readdir $dh2; + readdir $dh2; + # rationale for requiring an empty directory: + # https://bugs.debian.org/833525 + if (readdir $dh2) { + die "$target contains a non-empty lost+found directory"; + } + closedir($dh2); + } else { + die "$target is not empty"; + } + } + closedir($dh); The new version now check if either the target directory is empty, or it has only an empty lost+found directory in it. It will abort in any other case. This means that: - we still get all the safety and make sure nothing of value will accidentally get deleted - you don't need to manually rmdir a fresh lost+found directory on a new partition What do you think? Thanks! cheers, josch
signature.asc
Description: signature
