Bug 2230638 - [abrt] systemd-udev: udev_rule_token_free(): udevadm killed by SIGABRT
Summary: [abrt] systemd-udev: udev_rule_token_free(): udevadm killed by SIGABRT
Keywords:
Status: NEW
Alias: None
Product: Fedora
Classification: Fedora
Component: systemd
Version: 38
Hardware: x86_64
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: systemd-maint
QA Contact: Fedora Extras Quality Assurance
URL: https://retrace.fedoraproject.org/faf...
Whiteboard: abrt_hash:252aceb1f7488ea157399be22bf...
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2023-08-09 18:36 UTC by oper.kh
Modified: 2023-08-17 12:00 UTC (History)
10 users (show)

Fixed In Version:
Doc Type: If docs needed, set a value
Doc Text:
Clone Of:
Environment:
Last Closed:
Type: ---
Embargoed:


Attachments (Terms of Use)
File: proc_pid_status (1.36 KB, text/plain)
2023-08-09 18:36 UTC, oper.kh
no flags Details
File: maps (3.99 KB, text/plain)
2023-08-09 18:36 UTC, oper.kh
no flags Details
File: limits (1.29 KB, text/plain)
2023-08-09 18:36 UTC, oper.kh
no flags Details
File: environ (373 bytes, text/plain)
2023-08-09 18:36 UTC, oper.kh
no flags Details
File: open_fds (3.52 KB, text/plain)
2023-08-09 18:36 UTC, oper.kh
no flags Details
File: mountinfo (3.94 KB, text/plain)
2023-08-09 18:36 UTC, oper.kh
no flags Details
File: os_info (699 bytes, text/plain)
2023-08-09 18:37 UTC, oper.kh
no flags Details
File: cpuinfo (2.84 KB, text/plain)
2023-08-09 18:37 UTC, oper.kh
no flags Details
File: core_backtrace (5.65 KB, text/plain)
2023-08-09 18:37 UTC, oper.kh
no flags Details
File: dso_list (907 bytes, text/plain)
2023-08-09 18:37 UTC, oper.kh
no flags Details

Description oper.kh 2023-08-09 18:36:48 UTC
Description of problem:
May be problem wit laggy 6.4 kernel

Version-Release number of selected component:
systemd-udev-253.7-1.fc38

Additional info:
reporter:       libreport-2.17.11
type:           CCpp
reason:         udevadm killed by SIGABRT
journald_cursor: s=dcf9ad488c164869a043195efd0883b2;i=131b7;b=aaf388070e804ac9a8f81b166a1264f2;m=2654f77;t=60281407fd1a4;x=4c25ae96dfd87514
executable:     /usr/bin/udevadm
cmdline:        $'(udev-worker)'
cgroup:         0::/system.slice/systemd-udevd.service/udev
rootdir:        /
uid:            0
kernel:         6.2.9-300.fc38.x86_64
package:        systemd-udev-253.7-1.fc38
runlevel:       N 5
backtrace_rating: 4
crash_function: udev_rule_token_free
comment:        May be problem wit laggy 6.4 kernel

Truncated backtrace:
Thread no. 1 (19 frames)
 #8 udev_rule_token_free at ../src/udev/udev-rules.c:267
 #9 udev_rule_line_clear_tokens at ../src/udev/udev-rules.c:274
 #10 udev_rule_line_free at ../src/udev/udev-rules.c:283
 #12 udev_rule_file_free at ../src/udev/udev-rules.c:303
 #13 udev_rules_free at ../src/udev/udev-rules.c:314
 #14 manager_free at ../src/udev/udevd.c:262
 #15 manager_freep at ../src/udev/udevd.c:271
 #16 worker_main at ../src/udev/udevd.c:695
 #17 worker_spawn at ../src/udev/udevd.c:819
 #18 event_run at ../src/udev/udevd.c:878
 #19 event_queue_start at ../src/udev/udevd.c:998
 #20 on_post at ../src/udev/udevd.c:1534
 #21 source_dispatch at ../src/libsystemd/sd-event/sd-event.c:3707
 #22 sd_event_dispatch at ../src/libsystemd/sd-event/sd-event.c:4251
 #23 sd_event_run at ../src/libsystemd/sd-event/sd-event.c:4312
 #24 sd_event_loop at ../src/libsystemd/sd-event/sd-event.c:4333
 #25 main_loop at ../src/udev/udevd.c:1993
 #26 run_udevd at ../src/udev/udevd.c:2089
 #27 run at ../src/udev/udevadm.c:122

Comment 1 oper.kh 2023-08-09 18:36:52 UTC
Created attachment 1982614 [details]
File: proc_pid_status

Comment 2 oper.kh 2023-08-09 18:36:53 UTC
Created attachment 1982615 [details]
File: maps

Comment 3 oper.kh 2023-08-09 18:36:55 UTC
Created attachment 1982616 [details]
File: limits

Comment 4 oper.kh 2023-08-09 18:36:56 UTC
Created attachment 1982617 [details]
File: environ

Comment 5 oper.kh 2023-08-09 18:36:57 UTC
Created attachment 1982618 [details]
File: open_fds

Comment 6 oper.kh 2023-08-09 18:36:59 UTC
Created attachment 1982619 [details]
File: mountinfo

Comment 7 oper.kh 2023-08-09 18:37:00 UTC
Created attachment 1982620 [details]
File: os_info

Comment 8 oper.kh 2023-08-09 18:37:02 UTC
Created attachment 1982621 [details]
File: cpuinfo

Comment 9 oper.kh 2023-08-09 18:37:03 UTC
Created attachment 1982622 [details]
File: core_backtrace

Comment 10 oper.kh 2023-08-09 18:37:05 UTC
Created attachment 1982623 [details]
File: dso_list

Comment 11 oper.kh 2023-08-09 18:37:37 UTC
May be problem wit laggy 6.4 kernel


reporter:       libreport-2.17.11
type:           CCpp
reason:         udevadm killed by SIGABRT
journald_cursor: s=dcf9ad488c164869a043195efd0883b2;i=131b7;b=aaf388070e804ac9a8f81b166a1264f2;m=2654f77;t=60281407fd1a4;x=4c25ae96dfd87514
executable:     /usr/bin/udevadm
cmdline:        $'(udev-worker)'
cgroup:         0::/system.slice/systemd-udevd.service/udev
rootdir:        /
uid:            0
kernel:         6.2.9-300.fc38.x86_64
package:        systemd-udev-253.7-1.fc38
runlevel:       N 5
backtrace_rating: 4
crash_function: udev_rule_token_free
comment:        May be problem wit laggy 6.4 kernel

Comment 12 David Tardon 2023-08-16 12:34:56 UTC
We have seen a crash like this one before (e.g., bug 2141304 or bug 2144526). So far, our opinion has been that it's caused by a random memory corruption, but it's curious that it always crashes in the same function...

Comment 13 Zbigniew Jędrzejewski-Szmek 2023-08-17 07:21:20 UTC
The crash happens in cleanup, i.e. when memory is being deallocated, but most likely the issue is a memory corruption (double free, or buffer overflow, or a random memory poke, or something else), much earlier. So the fact that the trace is the same in various cases doesn't really tell us much. The fact that this has been reported a few times suggests that this is not an issue of flaky memory hardware, but rather a bug in udev code.

Comment 14 Frantisek Sumsal 2023-08-17 12:00:48 UTC
This part of the code was reworked quite in https://github.com/systemd/systemd/pull/26767 so I wonder if the issue is still present in v254. I guess only time will tell.


Note You need to log in before you can comment on or make changes to this bug.