Summary: | dev-libs/glib-2.56.2: appinfo - Bail out! GLib-GIO:ERROR:/var/tmp/portage/dev-libs/glib-2.56.2/work/glib-2.56.2/gio/tests/appinfo.c:403:test_associations: assertion failed: (appinfo2) | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Rolf Eike Beer <eike> |
Component: | Current packages | Assignee: | Gentoo Linux Gnome Desktop Team <gnome> |
Status: | RESOLVED OBSOLETE | ||
Severity: | normal | Keywords: | TESTFAILURE |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | build.log |
Description
Rolf Eike Beer
2018-11-03 07:50:42 UTC
Do you have x11-terms/gnome-terminal installed? If so, at what version? No, this is not installed. Getting rather stumped on this. Just to make sure - does this happen sometimes with -j1 MAKEOPTS too? And what terminals do you have installed? Looks like no gnome-terminal nor xterm, but maybe you have something installed that glib code actually can use, and we have trouble with that in that different test case (than the one we are disabling if g-t nor xterm is there). I think I saw this test failure myself at some point, but can't reliably reproduce anymore (can't even remember which machine it happened with). However I am getting various other issues sometimes from that test case source. This is a chroot running inside a screen on a machine without any graphics card. If there is not accidentially an xterm or whatever merged by a previous build test then there is likely none. Btw: I can't reproduce it at the moment anymore. (In reply to Rolf Eike Beer from comment #5) > This is a chroot running inside a screen on a machine without any graphics > card. If there is not accidentially an xterm or whatever merged by a > previous build test then there is likely none. Can you check with ls /var/db/pkg/x11-terms and /var/log/emerge.log in that chroot or so, just to figure out if non-xterm non-gnome-terminal terminals could have affected this at all, even if it doesn't reproduce now? Actually I don't think we have any of the other ones really packaged anymore that it might try, besides gnome-terminal and xterm; so probably not related indeed.. Do you think we can just ignore this case for now then and remove the blocker on stabilization? Sure, given the fact that it now works for me and I have marked this already stable on sparc. From grepping my log it looks like I had eterm merged when this failed. As I'm rather stumped on this, I made it always skip the tests that fail for me too with 2.58.1 bump. Not sure if it catches your case too or not. Will revisit later, or hopefully the problem just goes away once we build with meson in the future. https://gitlab.gnome.org/GNOME/glib/issues/1601 might help with this |