From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.2) Gecko/20030708 Description of problem: Building this package gives variable results. Some passes it will build a correct package, another run it won't. I have yet to establish any sort of pattern, I just build it until it works. The problem remains constant whether the build is run as root or a normal user, SMP or uniprocessor. It is easy to know if a build succeeded or not, just run ldd on /usr/src/redhat/BUILD/openssl-0.9.7a/libssl.so.0.9.7 as soon as the build process gets to the phase where it runs tests. If it shows the following it isn't going to work and you can skip the long tests: libcrypto.so.4 => /lib/libcrypto.so.4 (0xb74c3000) libdl.so.2 => /lib/libdl.so.2 (0xb74c0000) libz.so.1 => /usr/lib/libz.so.1 (0xb74b2000) libc.so.6 => /lib/tls/libc.so.6 (0xb737b000) /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000) But if you get these links to kerberos it will work: libgssapi_krb5.so.2 => /usr/kerberos/lib/libgssapi_krb5.so.2 (0xb75a1000) libkrb5.so.3 => /usr/kerberos/lib/libkrb5.so.3 (0xb7543000) libcom_err.so.3 => /usr/kerberos/lib/libcom_err.so.3 (0xb7541000) libk5crypto.so.3 => /usr/kerberos/lib/libk5crypto.so.3 (0xb7531000) libresolv.so.2 => /lib/libresolv.so.2 (0xb751f000) libcrypto.so.4 => /lib/libcrypto.so.4 (0xb742e000) libdl.so.2 => /lib/libdl.so.2 (0xb742a000) libz.so.1 => /usr/lib/libz.so.1 (0xb741c000) libc.so.6 => /lib/tls/libc.so.6 (0xb72e5000) /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x80000000) Version-Release number of selected component (if applicable): openssl-0.9.7a-23 How reproducible: Sometimes Steps to Reproduce: 1. rpmbuild --rebuild openssl-0.9.7a-23.src.rpm Additional info: Full Disclosure, I'm not running on RHEL3 but am one of the rebuilders, specifically whiteboxlinux.org. This problem has been consistent since taroon-beta and since everything else builds and has stood up to a fair amount of outside testing it was time to get this bug reported. If it fails to appear under your build conditions that in itself is an important datapoint.
Doesn't seem to happen here.