Date: Thu, 28 Jun 2007 23:04:30 +0300 From: Toomas Aas <toomas.aas@raad.tartu.ee> To: questions@freebsd.org Subject: One library fails port md5 checksum Message-ID: <4684144E.6000408@raad.tartu.ee>
next in thread | raw e-mail | index | archive | help
Hello! I'm running FreeBSD 6.2-RELEASE amd64 and have amanda-client and amanda-server installed from ports. Initially I had these ports at version 2.5.1p2 but as I was having some weird problems I decided to remove these versions and upgrade to latest available Amanda ports, 2.5.1p3. On removing the old version of amanda-client port, I got this error message: pkg_delete: '/usr/local/lib/libamandad.a' fails original MD5 checksum - not deleted. Oh my, I thought, I have a corrupt library. This is probably what caused all those weird problems I was having with Amanda. I removed /usr/local/lib/libamandad.a manually and installed the new Amanda ports (2.5.1p3). The problems that I was initially having with 2.5.1p2 seemed to be gone, but I discovered that I had given one configuration parameter wrong to the port build, so I decided to remove the amanda-server and amanda-client ports and re-install them with correct parameters. To my great surprise, on removing the new amanda-client port, I got the identical error message: pkg_delete: '/usr/local/lib/libamandad.a' fails original MD5 checksum - not deleted. I initially had these two settings in /etc/make.conf CPUTYPE=nocona CFLAGS=-O2 -pipe (This box has Intel Pentium D 945 CPU) I commented them out and removed and re-installed the Amanda ports once more. Still, the checksum that is listed in /var/db/pkg/amanda-client/+CONTENTS for /usr/local/lib/libamandad.a does not match the actual md5 checksum of the file. Having seen this, I again enabled these options in /etc/make.conf and rebuilt the Amanda ports once more. The checksum mismatch is still there. What might cause this? This kind of "random" corruption is generally attributed to hardware failure, but it seems not random enough, considering that I have now re-built Amanda 4 times and each time it's the same file that ends up having the checksum mismatch. Also, the box doesn't seem to suffer from any other problems such as random signal 11s etc. -- Toomas Aas -- ... How much net work could a network work, if a network could net work?
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4684144E.6000408>