[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#943554: engrampa FTBFS: python segfault while running itstool



Source: engrampa
Version: 1.22.2-1
Severity: serious
Tags: ftbfs

engrampa fails to build from source in unstable using sbuild on amd64. A
build ends with:

| if test -f "${mo}"; then mo="../${mo}"; else mo="/<<PKGBUILDDIR>>/help/${mo}"; fi; \
| (cd "as/" && itstool -m "${mo}" ${d}/C/index.docbook ${d}/C/legal.xml) && \
| touch "as/as.stamp"
| /bin/bash: line 3:  6349 Segmentation fault      (core dumped) itstool -m "${mo}" ${d}/C/index.docbook ${d}/C/legal.xml
| make[3]: *** [Makefile:573: ru/ru.stamp] Error 139
| make[3]: *** Waiting for unfinished jobs....
| make[3]: Leaving directory '/<<PKGBUILDDIR>>/help'
| make[2]: *** [Makefile:506: all-recursive] Error 1
| make[2]: Leaving directory '/<<PKGBUILDDIR>>'
| make[1]: *** [Makefile:438: all] Error 2
| make[1]: Leaving directory '/<<PKGBUILDDIR>>'
| dh_auto_build: make -j8 returned exit code 2
| make: *** [debian/rules:10: build] Error 255
| dpkg-buildpackage: error: debian/rules build subprocess returned exit status 2

It is also reproducible by reproducible builds:
https://tests.reproducible-builds.org/debian/rbuild/bullseye/amd64/engrampa_1.22.2-1.rbuild.log.gz

This is itstool segfaulting. itstool is written in Python, so it
actually it is a Python 3.7 interpreter segfault. This is bad.

It is not exactly clear, what is buggy precisely. Python should never
segfault, but the segfault can be caused by an extension (such as xml2).
Very likely one of python3.7, python3-libxml2, itstool or libxml2 is
buggy. Since I have no clue which, I'm filing the bug with engrampa now
and hope that someone can reassign it to the proper package. If you
reassign, please don't forget "affects -1 + src:engrampa".

Helmut


Reply to: