Bug 447820 - Konsole lost the ability to copy after failing to save more than 10,129,500 lines of an unlimited scrollback session
Summary: Konsole lost the ability to copy after failing to save more than 10,129,500 l...
Status: ASSIGNED
Alias: None
Product: konsole
Classification: Applications
Component: general (show other bugs)
Version: 21.12.0
Platform: Compiled Sources FreeBSD
: NOR major
Target Milestone: ---
Assignee: Konsole Developer
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2022-01-02 11:01 UTC by Graham Perrin
Modified: 2022-01-28 22:47 UTC (History)
0 users

See Also:
Latest Commit:
Version Fixed In:


Attachments
Screenshot of the point where I realised that the Edit menu was no longer fully functional (353.42 KB, image/png)
2022-01-02 11:01 UTC, Graham Perrin
Details
A shot of me browsing near the tail of the session (140.18 KB, image/png)
2022-01-02 11:29 UTC, Graham Perrin
Details
Another shot of me browsing near the tail of the session (172.95 KB, image/png)
2022-01-02 11:36 UTC, Graham Perrin
Details
A shot of me browsing the tail of the session (76.93 KB, image/png)
2022-01-02 11:41 UTC, Graham Perrin
Details
Screenshot: KDE Plasma presenting the volume as Microsoft basic data (322.96 KB, image/png)
2022-01-02 11:46 UTC, Graham Perrin
Details
Screenshot: the preference for unlimited scrollback (322.89 KB, image/png)
2022-01-02 11:48 UTC, Graham Perrin
Details
A nano view of part of the saved file (228.72 KB, image/png)
2022-01-02 11:51 UTC, Graham Perrin
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Graham Perrin 2022-01-02 11:01:02 UTC
Created attachment 145039 [details]
Screenshot of the point where I realised that the Edit menu was no longer fully functional

SUMMARY

An unusually large number of lines, so I'm not entirely surprised by the failure to save, however: 

* loss of the ability to copy, in this situation, is potentially major

– it seems that 'unlimited' is not truly without limitations. 

STEPS TO REPRODUCE

1. set scrollback to unlimited

2. run a command that might (unexpectedly) generate tens of millions of lines of output

3. File menu

4. Save Output As...

OBSERVED RESULT

5. with a different tab (or different application), tail the saved file

6. the tail of what was saved does not match the tail of what is on screen

  * i.e. _silent failure_ to save the session in its entirety

7. in the tab from which _partial_ output was saved, aim to copy part of what was not saved

  * select followed by Control-Shift-C results in loss of selection and 
    progression to the next line, as if Return or Enter was keyed

  * in the Edit menu, the Copy option is greyed out.

EXPECTED RESULT

5. successful save, of the session in its entirety, and no loss of functionality

  * the ability to copy is essential.

SOFTWARE/OS VERSIONS

Operating System: FreeBSD 14.0
KDE Plasma Version: 5.23.4
KDE Frameworks Version: 5.89.0
Qt Version: 5.15.2
Kernel Version: 14.0-CURRENT (64-bit)
Graphics Platform: X11
Memory: 15.9 GiB of RAM
Graphics Processor: AMD TURKS

% pkg info -x konsole
konsole-21.12.0_1
% date ; freebsd-version -kru
Sun  2 Jan 2022 10:40:40 GMT
14.0-CURRENT
14.0-CURRENT
14.0-CURRENT
% uname -aKU
FreeBSD mowa219-gjp4-8570p-freebsd 14.0-CURRENT FreeBSD 14.0-CURRENT #118 main-n251923-4bae154fe8c: Sat Dec 25 08:03:37 GMT 2021     root@mowa219-gjp4-8570p-freebsd:/usr/obj/usr/src/amd64.amd64/sys/GENERIC-NODEBUG  amd64 1400045 1400045
% 

ADDITIONAL INFORMATION

In this case, the unexpectedly massive output resulted from a single run of: 

fsck -fy

– for what should have been an empty UFS file system on a USB flash drive. Knowing that (previously) a bug had occurred at/around the time of creation of the file system, I did expect some nonsense within the output. I did not expect tens of millions of lines.
Comment 1 Graham Perrin 2022-01-02 11:23:07 UTC
Size of the saved file: 251328965

----

% pwd
/usr/home/grahamperrin/Documents/IT/BSD/FreeBSD/Verbatim STORE N GO kernel panics
% ls -hlrt
total 7187
-rw-r--r--  1 grahamperrin  grahamperrin    56K  6 Jan  2021 outline.txt
-rw-r--r--  1 grahamperrin  grahamperrin   512B  6 Jan  2021 3
-rw-r--r--  1 root          grahamperrin   7.3G  2 Jan 08:42 verbatimstorengo-bent.img
-rw-r--r--  1 grahamperrin  grahamperrin   240M  2 Jan 09:18 2022-01-02 09-00 Konsole output.txt
-rw-r--r--  1 grahamperrin  grahamperrin   353K  2 Jan 09:21 2022-01-02 09-21-35.png
-rw-r--r--  1 grahamperrin  grahamperrin   140K  2 Jan 09:22 2022-01-02 09-22-39.png
-rw-r--r--  1 grahamperrin  grahamperrin   173K  2 Jan 09:23 2022-01-02 09-23-01.png
-rw-r--r--  1 grahamperrin  grahamperrin   207K  2 Jan 09:24 2022-01-02 09-24-05.png
-rw-r--r--  1 grahamperrin  grahamperrin   323K  2 Jan 09:25 2022-01-02 09-25-36.png
-rw-r--r--  1 grahamperrin  grahamperrin   323K  2 Jan 09:29 2022-01-02 09-29-07.png
-rw-r--r--  1 grahamperrin  grahamperrin   229K  2 Jan 09:34 2022-01-02 09-34-27 line 79 of 10129500.png
% du -h 2022-01-02\ 09-00\ Konsole\ output.txt 
 23M    2022-01-02 09-00 Konsole output.txt
% file 2022-01-02\ 09-00\ Konsole\ output.txt
2022-01-02 09-00 Konsole output.txt: ASCII text
% ls -l 2022-01-02\ 09-00\ Konsole\ output.txt
-rw-r--r--  1 grahamperrin  grahamperrin  251328965  2 Jan 09:18 2022-01-02 09-00 Konsole output.txt
% tail -n 16 2022-01-02\ 09-00\ Konsole\ output.txt
UNEXPECTED SOFT UPDATE INCONSISTENCY

CLEAR? yes

INODE CHECK-HASH FAILED I=859828  OWNER=4189496942 MODE=17701
SIZE=14014691814411263273 MTIME=??? ?? ??:?? ???? 
FIX? yes

INODE CHECK-HASH FAILED I=859829  OWNER=1528400184 MODE=67562
SIZE=13898472712188788268 MTIME=??? ?? ??:?? ???? 
FIX? yes

UNKNOWN FILE TYPE I=859829
UNEXPECTED SOFT UPDATE INCONSISTENCY

CLEAR? yes
%
Comment 2 Graham Perrin 2022-01-02 11:29:59 UTC
Created attachment 145040 [details]
A shot of me browsing near the tail of the session

At the top of this shot, the tail of the (extraordinarily long) output from the first run of: 

fsck -fy /dev/da2p1

Then, the beginning of the (pleasantly short) output from the second run of the same command.
Comment 3 Graham Perrin 2022-01-02 11:36:51 UTC
Created attachment 145042 [details]
Another shot of me browsing near the tail of the session

The second run of 
fsck -fy /dev/da2p1 
concluded with marking clean of the file system. 

Subsequent use of commands such as du and ls confirmed what I expected: 

* a file system that's essential empty

* a negligible amount of data (not from any file that was ever written to the file system) in the lost+found area.
Comment 4 Graham Perrin 2022-01-02 11:41:29 UTC
Created attachment 145043 [details]
A shot of me browsing the tail of the session

Output (minimal) from 

ls -ahlR /tmp/danger/lost+found/

Finally, unmount of /tmp/danger
Comment 5 Graham Perrin 2022-01-02 11:46:57 UTC
Created attachment 145044 [details]
Screenshot: KDE Plasma presenting the volume as Microsoft basic data

Off-topic from Konsole, but part of a coherent story: this shot, taken today, helps to remind me of the history of the disk, prior to the command (around a year ago) that began creating the UFS file system.
Comment 6 Graham Perrin 2022-01-02 11:48:49 UTC
Created attachment 145045 [details]
Screenshot: the preference for unlimited scrollback

To remind me of the wording that appears for the yellow alert for unlimited scrollback.
Comment 7 Graham Perrin 2022-01-02 11:51:02 UTC
Created attachment 145046 [details]
A nano view of part of the saved file

Line 79 of 10129500: the first line of output from the first run of 

fsck -fy /dev/da2p1
Comment 8 Bug Janitor Service 2022-01-28 22:47:02 UTC
A possibly relevant merge request was started @ https://invent.kde.org/utilities/konsole/-/merge_requests/590