Hide Forgot
Description of problem: With abrt-2.14.4-6.fc33 from https://bodhi.fedoraproject.org/updates/FEDORA-2020-7cad807210 , when I try to report a crash and send it to the retrace server, it says "preparing environment for backtrace generation" and then the whole libreport window disappears. In journal, I see: Sep 24 12:21:36 f33 abrt-applet[3947]: free(): double free detected in tcache 2 Sep 24 12:21:36 f33 audit[3948]: ANOM_ABEND auid=1000 uid=1000 gid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 pid=3948 comm="report-gtk" exe="/usr/bin/report-gtk" sig=6 res=1 Sep 24 12:21:36 f33 systemd[1]: Started Process Core Dump (PID 3964/UID 0). Sep 24 12:21:36 f33 audit[1]: SERVICE_START pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-coredump@2-3964-0 comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success' Sep 24 12:21:36 f33 systemd-coredump[3965]: [🡕] Process 3948 (report-gtk) of user 1000 dumped core. Stack trace of thread 3948: #0 0x00007f4d67d95bc5 raise (libc.so.6 + 0x3dbc5) #1 0x00007f4d67d7e8a4 abort (libc.so.6 + 0x268a4) #2 0x00007f4d67dd8127 __libc_message (libc.so.6 + 0x80127) #3 0x00007f4d67ddfe1c malloc_printerr (libc.so.6 + 0x87e1c) #4 0x00007f4d67de18ec _int_free (libc.so.6 + 0x898ec) #5 0x00007f4d67f95b99 g_free (libglib-2.0.so.0 + 0x56b99) #6 0x00007f4d68bab740 consume_event_command_output (libreport.so.1 + 0x1b740) #7 0x00005596c94e1703 consume_cmd_output (report-gtk + 0xe703) #8 0x00007f4d67f90fd7 g_main_context_dispatch (libglib-2.0.so.0 + 0x51fd7) #9 0x00007f4d67fe1ab8 g_main_context_iterate.constprop.0 (libglib-2.0.so.0 + 0xa2ab8) #10 0x00007f4d67f8e3ff g_main_context_iteration (libglib-2.0.so.0 + 0x4f3ff) #11 0x00007f4d6819e3e5 g_application_run (libgio-2.0.so.0 + 0xd53e5) #12 0x00005596c94dbc3a main (report-gtk + 0x8c3a) #13 0x00007f4d67d801a2 __libc_start_main (libc.so.6 + 0x281a2) #14 0x00005596c94dbd7e _start (report-gtk + 0x8d7e) Stack trace of thread 3951: #0 0x00007f4d67e4ea0f __poll (libc.so.6 + 0xf6a0f) #1 0x00007f4d67fe1a4e g_main_context_iterate.constprop.0 (libglib-2.0.so.0 + 0xa2a4e) #2 0x00007f4d67f906ab g_main_loop_run (libglib-2.0.so.0 + 0x516ab) #3 0x00007f4d681d16b6 gdbus_shared_thread_func.lto_priv.0 (libgio-2.0.so.0 + 0x1086b6) #4 0x00007f4d67fbdf0e g_thread_proxy (libglib-2.0.so.0 + 0x7ef0e) #5 0x00007f4d672e43f9 start_thread (libpthread.so.0 + 0x93f9) #6 0x00007f4d67e59b03 __clone (libc.so.6 + 0x101b03) Stack trace of thread 3949: #0 0x00007f4d67e4ea0f __poll (libc.so.6 + 0xf6a0f) #1 0x00007f4d67fe1a4e g_main_context_iterate.constprop.0 (libglib-2.0.so.0 + 0xa2a4e) #2 0x00007f4d67f8e3ff g_main_context_iteration (libglib-2.0.so.0 + 0x4f3ff) #3 0x00007f4d67f90031 glib_worker_main (libglib-2.0.so.0 + 0x51031) #4 0x00007f4d67fbdf0e g_thread_proxy (libglib-2.0.so.0 + 0x7ef0e) #5 0x00007f4d672e43f9 start_thread (libpthread.so.0 + 0x93f9) #6 0x00007f4d67e59b03 __clone (libc.so.6 + 0x101b03) Stack trace of thread 3952: #0 0x00007f4d67e4ea0f __poll (libc.so.6 + 0xf6a0f) #1 0x00007f4d67fe1a4e g_main_context_iterate.constprop.0 (libglib-2.0.so.0 + 0xa2a4e) #2 0x00007f4d67f8e3ff g_main_context_iteration (libglib-2.0.so.0 + 0x4f3ff) #3 0x00007f4d68bd764d dconf_gdbus_worker_thread (libdconfsettings.so + 0x664d) #4 0x00007f4d67fbdf0e g_thread_proxy (libglib-2.0.so.0 + 0x7ef0e) #5 0x00007f4d672e43f9 start_thread (libpthread.so.0 + 0x93f9) #6 0x00007f4d67e59b03 __clone (libc.so.6 + 0x101b03) Sep 24 12:21:36 f33 systemd[1]: systemd-coredump: Succeeded. Sep 24 12:21:36 f33 audit[1]: SERVICE_STOP pid=1 uid=0 auid=4294967295 ses=4294967295 subj=system_u:system_r:init_t:s0 msg='unit=systemd-coredump@2-3964-0 comm="systemd" exe="/usr/lib/systemd/systemd" hostname=? addr=? terminal=? res=success' Sep 24 12:21:37 f33 abrt-server[3972]: Package 'libreport-gtk' isn't signed with proper key Sep 24 12:21:37 f33 abrt-server[3972]: 'post-create' on '/var/spool/abrt/ccpp-2020-09-24-12:21:37.154747-3948' exited with 1 Sep 24 12:21:37 f33 abrt-server[3972]: Deleting problem directory '/var/spool/abrt/ccpp-2020-09-24-12:21:37.154747-3948' Sep 24 12:21:37 f33 abrt-applet[3895]: Fetching title for problem report failed: HTTP Error 404: NOT FOUND For some reason, the libreport crash doesn't show up in abrt itself. Version-Release number of selected component (if applicable): abrt-2.14.4-6.fc33.x86_64 abrt-addon-ccpp-2.14.4-6.fc33.x86_64 abrt-addon-kerneloops-2.14.4-6.fc33.x86_64 abrt-addon-pstoreoops-2.14.4-6.fc33.x86_64 abrt-addon-vmcore-2.14.4-6.fc33.x86_64 abrt-addon-xorg-2.14.4-6.fc33.x86_64 abrt-cli-2.14.4-6.fc33.x86_64 abrt-dbus-2.14.4-6.fc33.x86_64 abrt-desktop-2.14.4-6.fc33.x86_64 abrt-gui-2.14.4-6.fc33.x86_64 abrt-gui-libs-2.14.4-6.fc33.x86_64 abrt-java-connector-1.2.0-5.fc33.x86_64 abrt-libs-2.14.4-6.fc33.x86_64 abrt-plugin-bodhi-2.14.4-6.fc33.x86_64 abrt-retrace-client-2.14.4-6.fc33.x86_64 abrt-tui-2.14.4-6.fc33.x86_64 gnome-abrt-1.3.6-5.fc33.x86_64 libreport-2.14.0-8.fc33.x86_64 libreport-anaconda-2.14.0-8.fc33.x86_64 libreport-cli-2.14.0-8.fc33.x86_64 libreport-fedora-2.14.0-8.fc33.x86_64 libreport-filesystem-2.14.0-8.fc33.noarch libreport-gtk-2.14.0-8.fc33.x86_64 libreport-plugin-bugzilla-2.14.0-8.fc33.x86_64 libreport-plugin-kerneloops-2.14.0-8.fc33.x86_64 libreport-plugin-logger-2.14.0-8.fc33.x86_64 libreport-plugin-reportuploader-2.14.0-8.fc33.x86_64 libreport-plugin-systemd-journal-2.14.0-8.fc33.x86_64 libreport-plugin-ureport-2.14.0-8.fc33.x86_64 libreport-web-2.14.0-8.fc33.x86_64 python3-abrt-2.14.4-6.fc33.x86_64 python3-abrt-addon-2.14.4-6.fc33.noarch python3-libreport-2.14.0-8.fc33.x86_64 How reproducible: tried twice, happened twice Steps to Reproduce: 1. will_abort 2. click Report in ABRT 3. confirm sending to retrace server 4. see the window disappear after ~10 seconds Additional info: I'm not sure whether this is a libreport crash or an abrt crash, because the crash itself references report-gtk, but the journal also says "abrt-applet[3947]: free(): double free detected in tcache 2".
Created attachment 1716297 [details] full journal
Proposing as a Beta blocker, this is another episode in the "What the hell happened to ABRT in F33?" documentary. However, note that bug 1881745 seems to be fixed now, so local generation can be used, if you reject remote retracing first. For that reason, this might be deemed to be a Final blocker instead, so proposing as a Final blocker and Beta FE as well.
Do you think you could upload the coredump as well?
Created attachment 1716334 [details] coredump Here's the coredump. If this is not helpful, I can try to gather useful information from it using gdb. I'm just juggling a bit too many things at once right now :-) Let me know, thanks.
Btw, this should be trivially reproducible from a VM with F33 Workstation Beta 1.3 installed and the latest abrt bodhi update applied: https://dl.fedoraproject.org/pub/alt/stage/33_Beta-1.3/
Rejected as a Beta blocker, accepted as a Final blocker, accepted as a Beta freeze exception: https://pagure.io/fedora-qa/blocker-review/issue/123
*** Bug 1882692 has been marked as a duplicate of this bug. ***
FEDORA-2020-6fc633096b has been submitted as an update to Fedora 33. https://bodhi.fedoraproject.org/updates/FEDORA-2020-6fc633096b
FEDORA-2020-6fc633096b has been pushed to the Fedora 33 testing repository. In short time you'll be able to install the update with the following command: `sudo dnf upgrade --enablerepo=updates-testing --advisory=FEDORA-2020-6fc633096b` You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2020-6fc633096b See also https://fedoraproject.org/wiki/QA:Updates_Testing for more information on how to test updates.
With abrt-2.14.4-6.fc33.x86_64 libreport-2.14.0-11.fc33.x86_64 this problem seems to be resolved. The libreport window crash no longer happens. That doesn't mean that the retrace server works, it simply returns "Retrace job failed". But at least libreport is no longer crashing. I think we can close this once libreport-2.14.0-11.fc33.x86_64 hits stable repos.
(In reply to Kamil Páral from comment #10) > That doesn't mean that the retrace server works, it simply returns > "Retrace job failed". But at least libreport is no longer crashing. Can you report another bug for this? That should be a final blocker IMO (basic functionality failure). If we cannot fix the retrace server ASAP, then ABRT needs to stop attempting to upload core dumps to the retrace server so we stop failing by default.
FYI @mgrabovsk is investigating while the retracing always fails. He should have some news soon.
FEDORA-2020-6fc633096b was unpushed, so I'm setting this back to ASSIGNED
Actually, it looks like FEDORA-2020-6fc633096b isn't the update that includes the fix. FEDORA-2020-e3af715292 does and it has gone stable so I'm closing it per Kamil's comment #10.
(In reply to Michael Catanzaro from comment #11) > Can you report another bug for this? That should be a final blocker IMO > (basic functionality failure). If we cannot fix the retrace server ASAP, > then ABRT needs to stop attempting to upload core dumps to the retrace > server so we stop failing by default. Reported as bug 1885154.
Bug fixed, commonbugs not needed.