Hi, We've also encountered this problem, with kexec-tools.
Root cause aopears to be that the original upstream tarball likely has duplicate entries and the behavior in tar for handling duplicate entries has changed bewteen squeeze and wheezy. Thus trying to recreate a tarball in one environment that has been generated in the other yields a different result, and that prevents xdelta from doing its job, ending with failure. In one tar version (the one from Squeeze, I think), tar apparently doesn't do anything special to deal with duplicates - it just writes extra copies into the tarball when encountered. But the other tar version (Wheezy), tar notices these and apparently substitutes a self-referenced sym-link for the duplicate entries. (Some kind of optimization perhaps?) So the failure comes from the different tarball generation, but that ultimately stems from different behavior in handling duplicate entries in the original upstream tarball, which really should not have happened in the first place. The kexec-tools package has the same issue. Just do a tar tvf of, say, kexec-tools_2.0.3.orig.tar.gz in pool/main/k/kexec-tools, and you'll see multiple instances of at least purgatory/arch/i386/entry32-16-debug.S (along with other cases). Perhaps pristine-tar could use an enhancement to screen out duplicates from the upstream manifest, if only to prevent breakage - or perhaps to warn the user. -Mike is...@pobox.com is...@isely.net -- Mike Isely isely @ isely (dot) net PGP: 03 54 43 4D 75 E5 CC 92 71 16 01 E2 B5 F5 C1 E8 -- To UNSUBSCRIBE, email to debian-qa-packages-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: https://lists.debian.org/alpine.deb.2.00.1406101348160.15...@cnc.isely.net
