Bug 1762719

FWIW, this is also being discussed on the retrace-server project and also in RHEL:

https://github.com/abrt/retrace-server/issues/258
rpmlib(PayloadIsZstd) <= 5.4.18-1 is needed

https://bugzilla.redhat.com/show_bug.cgi?id=1715799
Re-enable zstd compression support in RHEL-8 rpm

*** Bug 1784752 has been marked as a duplicate of this bug. ***

*** Bug 1757146 has been marked as a duplicate of this bug. ***

Isn't the inability of abrt to report problems a major issue for bug reports in Fedora 31? How are people supposed to report all the random crashes in their installation?



$ abrt-cli
eca2539 1x systemd 2019-12-14 15:38:21
35c57d2 1x nautilus 2019-12-26 09:00:26
559f870 1x nautilus 2020-01-04 14:41:07
16353be 1x abrt 2020-01-04 14:58:20

$ abrt-cli report 559f870
Reporting problem 559f870

no actions matching this problem found for event 'collect_GConf'
no actions matching this problem found for event 'collect_vimrc_system'
no actions matching this problem found for event 'collect_vimrc_user'
('report_uReport' completed successfully)
No /home/christian/.cache/gdm/session.log
Ok to upload core dump? (It may contain sensitive data). If your answer is 'No', a stack trace will be generated locally. (It may download a huge amount of data). [y/N/f/e] y
Querying server settings
Preparing an archive to upload
Uploading 5.9 MiB
Upload successful
Retrace job started
Analyzing crash data
..........................
Preparing environment for backtrace generation
.....
Retrace job failed
Retrace failed. Try again later and if the problem persists report this issue please.
    2020-01-04 23:03:00 Analyzing crash data
 2020-01-04 23:07:34 INFO:faf.Coredump2Packages:Executing eu-unstrip
INFO:faf.Coredump2Packages:Mapping build-ids into debuginfo packages
WARNING:faf.Coredump2Packages:Debuginfo conflict: '1e0069b4c149b951f05d0a5cc7e422c776fb3e41' is provided by set([u'python3-gobject-base-debuginfo', u'python3-gobject-debuginfo'])
INFO:faf.Coredump2Packages:Getting binary packages from debuginfos

    2020-01-04 23:07:35 Preparing environment for backtrace generation
   2020-01-04 23:08:44 /usr/bin/mock init --resultdir /srv/retrace/tasks/601212613/log --configdir /srv/retrace/tasks/601212613 exitted with 30
=== OUTPUT ===
INFO: mock.py version 1.4.21 starting (python version = 3.6.8)...
Start: init plugins
INFO: selinux enabled
Finish: init plugins
INFO: Signal handler active
Start: run
Start: clean chroot
Finish: clean chroot
Start: chroot init
INFO: calling preinit hooks
INFO: enabled HW Info plugin
Mock Version: 1.4.21
INFO: Mock Version: 1.4.21
Start: yum install
ERROR: Command failed:
 # /usr/bin/yum --installroot /var/lib/mock/601212613/root/ --releasever 7.7 --skip-broken install bash cpio glibc-debuginfo abrt-addon-ccpp shadow-utils findutils gdb rpm --setopt=tsflags=nocontexts
Failed to set locale, defaulting to C
[...]
ERROR You need to update rpm to handle:
rpmlib(PayloadIsZstd) <= 5.4.18-1 is needed by xz-libs-5.2.4-6.fc31.x86_64
rpmlib(PayloadIsZstd) <= 5.4.18-1 is needed by dbus-1:1.12.16-3.fc31.x86_64
rpmlib(PayloadIsZstd) <= 5.4.18-1 is needed by abrt-addon-ccpp-2.13.0-1.fc31.x86_64
[...]

*** Bug 1788098 has been marked as a duplicate of this bug. ***

Indeed, I'm not sure how nothing has happened with this. I guess that's one way to avoid bug reports, by the simplest way to report them being broken..

Enough with the conspiracy theories. The fix is being worked on.

From a user perspective it's hard to avoid thinking things like that. The user experience of reporting bugs through ABRT is already terrible enough given that the retrace server is usually very slow, when it doesn't fail for unknown reasons, and the only other option is flooding your install with a bunch of debuginfo packages to generate a local backtrace. And in my experience, the vast majority of crashes reported to Bugzilla through ABRT end up being closed as EOL without a single developer comment at all, so one wonders why they should go through the trouble.

Having the retrace server be broken for an entire Fedora release might as well be the equivalent of giving users that want to report bugs the finger.

Any update on this? Fedora 32 is around the corner and I'm still unable to report any crashes via the Problem Reporting program. Again, isn't the inability of abrt to report problems a major issue for for Fedora? If I were to release a distribution, I'd like to know if things are crashing.

Admittedly, this bug appears to report two problems:

> Retrace job failed
> Retrace failed. Try again later and if the problem persists report this issue please.
> [...]
> Preparing environment for backtrace generation
> ERROR You need to update rpm to handle:
> rpmlib(PayloadIsZstd) <= 5.4.18-1 is needed by xz-libs-5.2.4-6.fc31.x86_64
> rpmlib(PayloadIsZstd) <= 5.4.18-1 is needed by dbus-1:1.12.16-3.fc31.x86_64

Unfortunately the retrace server problem would have to be debugged by whoever is running it (Fedora Infrastructure?). And although the rpmlib problem is being adressed in bug 1715799 and Github/abrt/retrace-server/issues/258, there hasn't been much movement there lately. Are there any details missing from these reports?

Thanks for working on this.

*** Bug 1828562 has been marked as a duplicate of this bug. ***

This is blocked by reprovision of the server and we need to deploy the new version (the new code with the fix does not support el7 and the upgrade to el8 is a bit complex). Sorry :(

Still an issue with F32; can someone bump the version number of this bug? Also, I hope the reprovisioning is progessing well, reading that ugrading from EL7 to EL8 is "complex" may not be good advertisment to the world, hm? :-)

F31 is still supported, so bumping the version makes no sense and the upgrade complexities stem from our deployment, not the operating system itself.

*** Bug 1839678 has been marked as a duplicate of this bug. ***

This is resolved in upstream, just wait to get retrace back online. Moving to Close Upstream so it does not polute with the NEW list of bugs.

Hi. How much longer?
A month later I still see:

--- Running report_uReport ---
Failed to upload uReport to the server 'https://retrace.fedoraproject.org/faf' with curl: Failed to connect to retrace.fedoraproject.org port 443: Connection timed out
Error: curl_easy_perform: Timeout was reached
('report_uReport' exited with 1)

regards, L.