+1 Maintenance Report

[Graham Inggs]

I was on +1 maintenance for the week of June 20-24.  Below are the
things I worked on.

On Monday, I found many packages had FTBFS without logs, or failed to
upload, and I retried these.

On Tuesday, the autopkgtest queues were empty so I scheduled retries
of all regressed autopkgtests (826 in total).  Once these were done, I
scheduled migration-reference/0 autopkgtests of those that were still
regressed (579 in total).  Presumably the difference being those that
passed on retry.

I found several packages that were blocked because of missing binaries
on architectures that had been dropped and filed bugs for the removal
of their binaries:
pyopencl (mesa stopped building mesa-opencl-icd on riscv64) LP: #1979297
sysbench dropped ppc64el LP: #1979299
gromacs dropped 32-bit LP: #1979302
kallisto dropped 32-bit LP: #1979304
abyss dropped 32-bit LP: #1969297 (re-opened)

dd-opentracing-cpp
==================
The version in Ubuntu FTBFS, so I sync'd the new version from Debian
which included the Ubuntu delta and a fix for the FTBFS.

sfxr-qt
=======
I fixed a FTBFS with glibc 2.34 and filed Debian bug #1013307.

chktex
======
I fixed a FTBFS when TERM is not set and filed Debian bug #1013358.

toil
====
I fixed a FTBFS with Python 3.10 as the only supported version and
filed Debian bug #1013275.

python-bayespy
==============
I cherry-picked an upstream commit to fix a FTBFS with scipy 1.8.
Since this is an orphaned package in Debian, I intended to do a QA
upload, but luckily somebody beat me to it, and I was able sync again
from Debian.

r-base
======
I merged the new upstream version from Debian unstable.

macaulay2
=========
I investigated autopkgtest flakiness on amd64 and opened an upstream
PR relaxing the timing of a test.