Mailinglist Archive: opensuse-factory (377 mails)

< Previous Next >
[opensuse-factory] openSUSE reproducible builds status 2020-08
Hi,

last month's status:
https://lists.opensuse.org/opensuse-factory/2020-07/msg00417.html

Last months' reproducible builds project updates (including my work):
https://reproducible-builds.org/reports/2020-07/

I uploaded https://rb.zq1.de/compare.factory-20200830/ today
and rbstats are:
total-packages: 13244 (+111)
build-tried: 13236 (+110)
build-failed: 66 (+28)
build-n-a: 125 (-1)
build-succeeded: 13045 (+83)
build-official-failed+na: 323 (+72)
build-compare-failed: 431 (-3)
build-compare-succeeded: 12614 (+86)
verify-failed: 519 (+28)
verified-semi-reproducible: 12362 (+258)
verified-bit-identical: 0 (+0)
bit-by-bit-identical: 12488 (+106)
not-bit-by-bit-identical: 558 (-23)

https://rb.zq1.de/compare.factory-20200830/graph.png
shows the change over time

https://rb.zq1.de/compare.factory-20200830/unreproduciblerings.txt
lists very unreproducible core packages (bootstrap+DVD)

Of the badly unreproducible packages,
2 were in ring0
45 were in ring1

That makes it 47/3268 => 1.44 %
which is below the overall average of
431/13045 => 3.30 %

558/13045 => 4.28 % of packages are not perfectly reproducible


Build workers were upgraded to Leap-15.2 to have a rpm that understands
zstd.

Several small packages seem to build faster now, possibly because the
setup of the build environment is faster with zstd.
E.g. "hub" went from 181+139s to 87+64s for -j1 and -j4 builds
and "xinit" went from 64+63s to 46+51s


Notable unreproducible core packages:
cargo-c:
parallelism? + filesys in rust libgit2-sys

installation-images:
ordering issues and plenty other issues

xen:
PE timestamps in .efi files


Ciao
Bernhard M.

< Previous Next >
This Thread
  • No further messages