<div dir="ltr">Here's my report for the (short) week of June 07-11 – I was out sick on Friday after getting my vaccination. ### pypy/pypy3 ### pypy(3) / 7.3.5+dfsg-1 Made sure @tumbleweed's changes made it into Ubuntu from Debian/experimental and resolved the problems in pypy and pypy3, as discussed during my last +1 shift. All looking good, thanks Stefano! ### iotjs ### iotjs / 1.0+715-1ubuntu2 upload for python2-rm transition I removed the alternative "python" build-dependency to satisfy the "python2-rm" transition (as applied in Debian salsa, but not yet released). ### zziplib ### zziplib / 0.13.62-3.3 python2-rm ... python2 dependencies seem to be resolved upstream as of 0.13.70, needs version bump. But this is quite a version bump: Last update was a few years ago, the old Debian git repo does not exist anymore and the zzip project switched its build system from automake to cmake. I started packaging the latest 0.13.72 release on Github. A refresh of the patches and double-check of the test system still needs to be done. <a href="https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=938924">https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=938924</a> <a href="https://github.com/slyon/zziplib-debian">https://github.com/slyon/zziplib-debian</a> ### python-linecache2 ### python-linecache2 / 1.0.0-4ubuntu2 upload for python2-rm transition That's an interesting package. The Ubuntu version is ahead of Debian, but does not contain Debian's latest changes (i.e. the python2 removal). This would be a sync, but we cannot just sync it as that would break our upgrade path. So I did a "manual sync" (reverting the relevant files to 1.0.0-4) and uploaded it as the new 1.0.0-4ubuntu2 version. Once the Debian revision is bump'ed to 5 we can do an actual sync, as our Delta is only related to python2 changes, which are now dropped. The dropped py2 package did not have any reverse dependency. ### python-pbr ### python-pbr / 5.5.1-0ubuntu2 upload for python2-rm transition Dropped the python2 dependency, the python2 python-pbr binary package and adopted debian/rules accordingly. Debian is now also on 5.5.1 and already dropped the python2 support, so this package is pending a merge, but I don't know enough about that package so I'll leave that for somebody else. Another -ubuntu3 upload was needed to fix a problem with update-alternatives. ### "unkown" autopkgtests ### NO - casper on armhf vs ['linux-meta-riscv/5.11.0.1009.9+21.10.1']<div>=> retried with new linux trigger, still running. <div>NO - apparmor-profiles-extra on armhf vs ['linux-meta-riscv/5.11.0.1009.9+21.10.1'] OK - libnumbertext on armhf vs ['glibc/2.33-0ubuntu8'] OK - libnet-prometheus-perl on armhf vs ['glibc/2.33-0ubuntu8'] OK - systemd on s390x vs ['isc-dhcp/4.4.1-2.2ubuntu8'] ### nss ### nss / 2:3.63-1ubuntu1 autopkgtest regression Some investigation about dogtag-pki vs nss autopkgtest failure, building and running autopkgtests against different versions of nss to check where it started failing. This is a real regression in nss v3.63+ on s390x. I was about to open a bug report about it, but cpaelzer beat me to it: <a href="https://pad.lv/1931104">https://pad.lv/1931104</a> It might be worth doing a bitsect of nss v3.61..v3.63 to see where it starts failing, but I need access to a s390x machine to do that. ### Canonistack ### Trying to spin up a s390x instance on Canonistack to further investigate the nss regression I (once again) ran into issue creating openstack instances and decided to create a new ticket about it: cRT#131289 Cheers, Lukas </div></div></div>