Bug 5760 - telinit 5 from runlevel 3 boot aborts all tty sessions
Summary: telinit 5 from runlevel 3 boot aborts all tty sessions
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: Cauldron
Hardware: i586 Linux
Priority: Normal normal
Target Milestone: ---
Assignee: Mageia Bug Squad
QA Contact:
URL:
Whiteboard:
Keywords: PATCH
Depends on:
Blocks:
 
Reported: 2012-05-05 18:26 CEST by Felix Miata
Modified: 2012-05-06 12:15 CEST (History)
1 user (show)

See Also:
Source RPM: systemd
CVE:
Status comment:


Attachments
tgz of /root/drakx (528.55 KB, application/octet-stream)
2012-05-05 18:26 CEST, Felix Miata
Details

Description Felix Miata 2012-05-05 18:26:26 CEST
Created attachment 2188 [details]
tgz of /root/drakx

Parallel of https://bugzilla.novell.com/show_bug.cgi?id=746594 & https://bugzilla.redhat.com/show_bug.cgi?id=708537

To reproduce:
1-boot with 3 on Grub cmdline
2-login on tty(s) (I did tty2 & tty3, started mc on tty2, did 'init 5' on tty3)
3-'telinit 5'

Actual results:
1-KDM appeared on tty1
2-screen cleared to login prompt on tty2
3-screen cleared to login prompt on tty3

Expected results:
1-KDM appeared on tty7
2-mc remained running on uncleared tty2
3-shell prompt remained on uncleared tty3
Manuel Hiebel 2012-05-05 22:55:37 CEST

Keywords: (none) => PATCH
CC: (none) => mageia
Source RPM: (none) => systemd

Comment 1 Colin Guthrie 2012-05-06 10:55:45 CEST
Your expected results are wrong. Should be probably:

Expected results:
1-KDM appeared on tty1
2-mc remained running on uncleared tty2
3-shell prompt remained on uncleared tty3

But yes, ttys2 & 3 should be unaffected by this. tty1 will always be nuked due to the conflicts in the prefdm.service.
Comment 2 Colin Guthrie 2012-05-06 11:10:12 CEST
Upstream patch http://cgit.freedesktop.org/systemd/systemd/commit/?id=4771148bb92ace55eaa6759a53d04a0f2de9b0d2 included in svn.

Status: NEW => RESOLVED
Resolution: (none) => FIXED

Comment 3 Felix Miata 2012-05-06 11:32:36 CEST
(In reply to comment #1)
> Your expected results are wrong. Should be probably:
> Expected results:
> 1-KDM appeared on tty1
> 2-mc remained running on uncleared tty2
> 3-shell prompt remained on uncleared tty3
 
> But yes, ttys2 & 3 should be unaffected by this. tty1 will always be nuked due
> to the conflicts in the prefdm.service.

In Rawhide last I checked (not for some time), booting into runlevel 3 (which is my preferred init on all test systems, and without plymouth) keeps X on tty[7-10] where it/they belong. tty1 is for text boot messages, and I rarely use it except to view the last screen of boot messages (except in installations that clear it by default and I can't figure out how to prevent the clearing, such as in Mandriva/Mageia). In openSUSE I still make use of the option to not use systemd except on a very few systems purely used for testing systemd in Factory. So here at least, there should rarely if ever be a reason to nuke tty1.
Comment 4 Colin Guthrie 2012-05-06 12:15:45 CEST
I even linked to the commit message where this is clearly explained that tty1 is still nuked due to the conflict in prefdm.service.

If you want you can copy the prefdm.service unit to /etc/systemd/system and customise it to taste such that it does not conflict with getty@tty1.

So your comments about how it works in Rawhide are wrong.

This is an old discussion that was had on the mailing list a long time ago. RC stage is NOT the time to revisit this, so no further comments on this topic please. If you strongly disagree, please participate in discussion when planning mga3.

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