Bug#873839: dbus: test-relay fails on hurd-i386: Aborted
Source: dbus
Version: 1.11.16+really1.11.16-2
Severity: normal
(X-Debbugs-Cc to debian-hurd)
I recently enabled build-time tests for dbus, after fixing the home
directory issue that had previously meant they failed on all Debian
buildds. Unfortunately this led to a build failure on hurd-i386.
Because it's reasonably clear that nobody with an interest in Hurd
is currently contributing to dbus upstream, and the test results indicate
that dbus does basically work, I'm going to ignore the test failures on
!linux for the moment.
The bug I'm describing in this bug report manifests as a failure of
the test case test-relay:
> ../../../build-aux/tap-driver.sh: line 639: 11727 Aborted "$@"
> # random seed: R02Sd75850b6e61c3dffc4f87cea5555493f
> 1..3
> ok 1 /connect
> PASS: test-relay 1 /connect
> ok 2 /relay
> PASS: test-relay 2 /relay
> # Bug Reference: https://bugs.freedesktop.org/show_bug.cgi?id=34393
> ERROR: test-relay - too few tests run (expected 3, got 2)
> ERROR: test-relay - exited with status 134 (terminated by signal 6?)
The crash here must be somewhere in test_limit(), which sends 8192
messages all at once, with a low resource limit, in an attempt to
reproduce fd.o#34393, then waits for their replies. I don't know why
or where it aborts: I would have expected to see something on stderr if
it's hitting an assertion failure in either libdbus or GLib.
S
Reply to: