On emerge openoffice (v2.0.1) I get a failed digest verification (md5 checksum failed on verification.) Running: # cd /usr/portage/app-office/openoffice/ # ebuild openoffice-2.0.1.ebuild digest solves that problem, but then OO stalls with an error regarding a corrupt tarball. #unpacking OO_build tree #tar: skipping to next header #tar: Archive contains obsolescent base-64 headers #bzip2:data integrity error when decompressing I assume the md5 checksum failes because of the corrupted tar ball This problem was reproducible, several times I deleted the relevant tar ball(s) sources and ebuilds in the portage tree, resynced and tried to emerge again. My system is set to sync to a uk server. I run a standard system, standard safe cflags (bar -o3) on the latest kernel. Any queries please do not hesitate to contact me.
A thread discussing this can be found at http://forums.gentoo.org/viewtopic.php?p=3071564#3071564
For those still having the issue, can you post what mirror you downloaded the supposed bad files from? If you try and download from another mirror and see if you get the same message.
ftp://www.mirrorservice.org/sites/www.ibiblio.org/gentoo/distfiles/ The OOO_2_0_1-core.tar.bz2 appears to be the file thats corrupted, the one it is failing on. I don't know if there are others it hasn't checked at this point which also have a problem. I did put this link up in the URL section, I'm not sure if that's what it is supposed to be used for... Is it for my own personal page (if I had one...). Should I not do this in the future? Using http://distfiles.gentoo.org at the moment. It appears to have decompressed this tarball without complaint, and is now compiling (got to wait a few hours before I know it's a success.....),but looking good :-)
Andreas, Any ideas here?
(In reply to comment #4) > Andreas, > > Any ideas here? > If I read that bug report right, this is simply a faulty file on one mirror, not a lot I can help here, not a problem with OOo / its digest
I just downloaded all the OOO-2.0.1 files from the mirror and all md5's checked out. Might have just been that one rsync that needed to catch up on itself.