Bug#1109120: libsoup3: intermittent test failure: timeout in server-test

Simon McVittie smcv at debian.org
Fri Jul 11 19:13:18 BST 2025


Source: libsoup3
Version: 3.6.4-2
Severity: important
Tags: ftbfs upstream fixed-upstream
Control: block 1035983 by -1
Control: forwarded -1 https://gitlab.gnome.org/GNOME/libsoup/-/issues/379

In some builds of libsoup3, server-test fails with a timeout, for 
example:

https://buildd.debian.org/status/fetch.php?pkg=libsoup3&arch=riscv64&ver=3.6.4-2&stamp=1737573056&raw=0
> server-test time out (After 360.0 seconds)
> 25/38 server-test              TIMEOUT         360.05s   killed by signal 15 SIGTERM
> ――――――――――――――――――――――――――――――――――――― ✀  ―――――――――――――――――――――――――――――――――――――
> stderr:
> **
> ERROR:../tests/server-test.c:851:do_fail_500_test: Unexpected status 0 (null) (expected 500 Internal Server Error)
> **
> ERROR:../tests/server-test.c:855:do_fail_500_test: 'usd.handler_called' should be TRUE
> ――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――――

One possible root cause of this (there might be more than one) has been 
fixed in upstream git, although not yet in a release; it was diagnosed 
as a deadlock caused by locking a mutex in the wrong place.

There are some CVEs that have, similarly, been fixed in upstream git but 
not in a release, so while we're fixing those CVEs *anyway*, we might as 
well try applying upstream's patch and seeing whether it helps this.

(If this test turns out to be flaky for other reasons than that one, 
then that should probably be reported separately, and developers who 
find it to be an unacceptable situation are welcome to debug further. I 
am sorry that I cannot immediately fix every bug.)

    smcv



More information about the pkg-gnome-maintainers mailing list