Hi, last month's status: https://lists.opensuse.org/opensuse-factory/2019-04/msg00414.html Last months' reproducible builds project updates (including my work): https://reproducible-builds.org/reports/2019-04/ The 2019-05 report is still a raw draft. I uploaded https://rb.zq1.de/compare.factory-20190528/ today and rbstats are: total-packages: 11911 build-tried: 11909 build-failed: 53 build-tried: 11909 build-n-a: 111 build-succeeded: 11745 build-official-failed+na: 182 build-compare-failed: 338 build-compare-succeeded: 11407 verified-semi-reproducible: 9937 verified-bit-identical: 0 bit-by-bit-identical: 11105 not-bit-by-bit-identical: 636 https://rb.zq1.de/compare.factory-20190528/graph.png shows the change over time https://rb.zq1.de/compare.factory-20190528/unreproduciblerings.txt lists very unreproducible core packages (bootstrap+DVD) Of the badly unreproducible packages, 5 were in ring0 42 were in ring1 That makes it 47/2901 => 1.62 % which is below the overall average of 338/11745 => 2.88 % 636/11745 => 5.42 % of packages are not perfectly reproducible Nearly fixed: acl https://build.opensuse.org/request/show/705693 python-marshmallow https://build.opensuse.org/request/show/705891 already fixed upstream https://github.com/marshmallow-code/marshmallow/pull/679 Progress: libreoffice Some progress with .bau files - those are now reproducible thanks to a mtime-normalization patch in the zip package. Remaining variations in .thm, .sdg, overview-summary.html and segments_3 files Newly unreproducible: containerd /usr/sbin/containerd-shim has buildid variations from parallelism similar to many other go packages go1.12 (same problems as 1.10 and 1.11 have) variations in -a files and binaries: go, trace, pprof, objdump, nm, addr2line python-numpy .pyc file variation (maybe ref counters) python-tox .pyc files contain random /tmp/ path from pip install https://bugzilla.opensuse.org/show_bug.cgi?id=1094323 affects 27+ packages atm Ciao Bernhard M.