Community Packages

Please read this before reporting a bug:
https://wiki.archlinux.org/title/Bug_reporting_guidelines

Do NOT report bugs when a package is just outdated, or it is in the AUR. Use the 'flag out of date' link on the package page, or the Mailing List.

REPEAT: Do NOT report bugs for outdated packages!
Tasklist

FS#74215 - [ncdu] SIGABRT when refreshing in some cases

Attached to Project: Community Packages
Opened by Octavia Togami (octylFractal) - Wednesday, 23 March 2022, 19:25 GMT
Last edited by Andreas Schleifer (Segaja) - Thursday, 24 March 2022, 20:08 GMT
Task Type Bug Report
Category Packages
Status Closed
Assigned To Andreas Schleifer (Segaja)
Levente Polyak (anthraxx)
Architecture All
Severity Low
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description: When I refresh the view with 'r' in some cases, ncdu crashes with SIGABRT. I grabbed a coredump using gdb, and the 'bt full' output is attached.

Specifically, ncdu crashes when I refresh when it is run as 'sudo ncdu -x /', refreshing in my home folder appears to be fine.


Additional info:
* package version(s)
ncdu 2.1-2
* config and/or log files etc.
It prints out this when it crashes:
thread 12970 panic: reached unreachable code
1] 12969 IOT instruction sudo ncdu -x /
* link to upstream bug report, if any

Steps to reproduce:
Run 'sudo ncdu -x /'. After it scans, refresh with 'r'. It should crash. Note that I couldn't get a friend to reproduce this, so there might be something specific about my filesystem that causes this.
This task depends upon

Closed by  Andreas Schleifer (Segaja)
Thursday, 24 March 2022, 20:08 GMT
Reason for closing:  Fixed
Additional comments about closing:  fixed in ncdu 2.1-3
Comment by Octavia Togami (octylFractal) - Wednesday, 23 March 2022, 19:29 GMT
Note: I can reproduce this on another one of my computers.
Comment by Andreas Schleifer (Segaja) - Wednesday, 23 March 2022, 20:03 GMT
I can confirm the crash on my local system which is ext4 with an AMD CPU.
Comment by loqs (loqs) - Wednesday, 23 March 2022, 20:17 GMT Comment by Octavia Togami (octylFractal) - Wednesday, 23 March 2022, 20:25 GMT
I did see that issue, but I am not convinced it is the same one as I couldn't find any EINVALs in the strace output, and this crashes seemingly when scanning /boot, not /proc.
Comment by loqs (loqs) - Wednesday, 23 March 2022, 20:32 GMT
Open a new bug report upstream [1] so upstream can decide if it is the same issue?

[1] https://code.blicky.net/yorhel/ncdu/issues
Comment by Andreas Schleifer (Segaja) - Wednesday, 23 March 2022, 20:41 GMT
Hm I could either add the last lines of an strace run to the 192 issue or I make a new one. I'm not sure tho.
Comment by Andreas Schleifer (Segaja) - Wednesday, 23 March 2022, 20:50 GMT Comment by Andreas Schleifer (Segaja) - Thursday, 24 March 2022, 19:12 GMT
https://code.blicky.net/yorhel/ncdu/issues/194#issuecomment-1131 upstream made a fix. I will package this as patch now until it gets released.

Loading...