Description of problem: # dnf upgrade --refresh -v --exclude=mutter Version-Release number of selected component: python3-3.8.0~rc1-1.fc32 Additional info: reporter: libreport-2.10.1 backtrace_rating: 4 cgroup: 0::/user.slice/user-1000.slice/user/gnome-terminal-server.service cmdline: /usr/bin/python3 /usr/bin/dnf upgrade --refresh -v --exclude=mutter crash_function: solv_calloc executable: /usr/bin/python3.8 journald_cursor: s=af85196e12e64a24a411e7dfc5442bbf;i=35b308;b=1427d6151e494710a82e87f4b1145fb6;m=1a0da76fc4;t=59453a571f52f;x=f73aa61a81e06cca kernel: 5.4.0-0.rc1.git1.1.fc32.x86_64 rootdir: / runlevel: N 5 type: CCpp uid: 0 Truncated backtrace: Thread no. 0 (10 frames) #2 solv_calloc at ../src/util.c:79 #3 pool_setarchpolicy at ../src/poolarch.c:114 #5 dnf_sack_set_arch(DnfSack*, gchar const*, GError**) at /usr/src/debug/libdnf-0.35.5-2.fc32.x86_64/libdnf/dnf-sack.cpp:828 #6 sack_init(_SackObject*, PyObject*, PyObject*) at /usr/src/debug/libdnf-0.35.5-2.fc32.x86_64/python/hawkey/sack-py.cpp:265 #7 wrap_init at /usr/src/debug/python3-3.8.0 #8 PyObject_Call at /usr/src/debug/python3-3.8.0 #9 do_call_core at /usr/src/debug/python3-3.8.0 #10 _PyEval_EvalFrameDefault at /usr/src/debug/python3-3.8.0 #11 _PyEval_EvalCodeWithName at /usr/src/debug/python3-3.8.0 #12 _PyFunction_Vectorcall at /usr/src/debug/python3-3.8.0
Created attachment 1623197 [details] File: backtrace
Created attachment 1623198 [details] File: core_backtrace
Created attachment 1623199 [details] File: cpuinfo
Created attachment 1623200 [details] File: dso_list
Created attachment 1623201 [details] File: environ
Created attachment 1623203 [details] File: limits
Created attachment 1623204 [details] File: maps
Created attachment 1623205 [details] File: mountinfo
Created attachment 1623206 [details] File: open_fds
Created attachment 1623207 [details] File: proc_pid_status
Created attachment 1623211 [details] dnf.log
I'm going to mark this as a duplicate. It is a bit different crash, but it seems relevant. *** This bug has been marked as a duplicate of bug 1758850 ***
I'm actually going to unmark it... the other bug is a problem with g_log handler, this one seems to be crashing in libsolv. They may be related, but lets not let this one get lost.
OK. Sorry about that. Marking as libsolv bug. Let us know if you need some Python help.
I initially saw a very similar bt in libsolv debugging #1761282 --- 2019-10-14 00:10:41.251 | Thread 1 "python3" received signal SIGSEGV, Segmentation fault. 2019-10-14 00:10:41.251 | 0x00007ffff7e8611d in _int_malloc () from /lib64/libc.so.6 2019-10-14 00:10:41.254 | #0 0x00007ffff7e8611d in _int_malloc () from /lib64/libc.so.6 2019-10-14 00:10:41.254 | #1 0x00007ffff7e889c6 in calloc () from /lib64/libc.so.6 2019-10-14 00:10:41.276 | #2 0x00007ffff6787b90 in solv_calloc () from /lib64/libsolv.so.1 2019-10-14 00:10:41.276 | #3 0x00007ffff674fa8f in pool_setarchpolicy () from /lib64/libsolv.so.1 2019-10-14 00:10:41.277 | #4 0x00007ffff6bfaf24 in dnf_sack_set_arch () from /lib64/libdnf.so.2 2019-10-14 00:10:41.299 | #5 0x00007ffff51623e5 in sack_init(_SackObject*, _object*, _object*) () from /usr/lib64/python3.7/site-packages/hawkey/_hawkey.so 2019-10-14 00:10:41.299 | #6 0x00007ffff7bb3276 in ?? () from /lib64/libpython3.7m.so.1.0 ... --- as i started installing debug libraries and re-running, the problem seemed to move. However, I strongly suggest that this suggests the issue is really something like #1758850, and not actually libsolv.
Mikhail, segfaults in memory allocation hint at corrupted heap, which could be caused by the same issue as https://bugzilla.redhat.com/show_bug.cgi?id=1758737. The fix for that should be released this week and in the meantime, you can also use our nightly repositories: https://copr.fedorainfracloud.org/coprs/rpmsoftwaremanagement/dnf-nightly/ which have the fix already. How often are you getting this or similar traceback? Please let us know after you install the fix. If you can (semi-)consistently reproduce the segfault (after installing the fixed version), running under valgrind might find the problem. Thanks!
*** Bug 1762279 has been marked as a duplicate of this bug. ***
At this stage I'm quite certain the crash was caused by the log handler issue, marking this as duplicate. The fix for it has been released. If the problems persist, feel free to reopen this. *** This bug has been marked as a duplicate of bug 1758737 ***