we made little progress this month. One good news is that SLE-15-SP3 (and thereby Leap 15.3 core) now got patches for rpmbuild and pesign-obs-integration that allow all kmp binary packages to be verified. Those patches were part of Factory for nearly 2 years. Also a groff date patch was merged to fix corosync verifiability.
Other unverifiable SLE packages are mostly those that are known to not build reproducibly: erlang grub2 kernel-preempt libguestfs libpinyin ongres-scram perf pnetcdf postgresql-jdbc python39 qemu rabbitmq-server release-notes-* xen most causes are known, but hard to fix for me.
Only fwupd is still suffering from pesigning.
These were the SLE news. Now for the Factory status:
last month's status: https://firstname.lastname@example.org/thread/3...
Last months' reproducible builds project updates (including my work): https://reproducible-builds.org/reports/2021-02/
I uploaded https://rb.zq1.de/compare.factory-20210328/ today and rbstats are: total-packages: 13868 (-41) build-tried: 13859 (-41) build-failed: 35 (-46) build-n-a: 153 (+18) build-succeeded: 13671 (-13) build-official-failed+na: 342 (-51) build-compare-failed: 423 (-10) build-compare-succeeded: 13248 (-3) verify-failed: 493 (-4) verified-semi-reproducible: 12888 (+17) bit-by-bit-identical: 13119 (+0) not-bit-by-bit-identical: 552 (-13) not-bit-by-bit-identicalcheck: 552 (-13)
https://rb.zq1.de/compare.factory-20210328/graph.png shows the change over time
https://rb.zq1.de/compare.factory-20210328/unreproduciblerings.txt lists very unreproducible core packages (bootstrap+DVD)
Of the badly unreproducible packages, 3 were in ring0 44 were in ring1
That makes it 47/3247 => 1.45 % which is below the overall average of 423/13671 => 3.09 %
552/13671 => 4.04 % of packages are not perfectly reproducible
Ciao Bernhard M.