Bug 1045115

Summary: Keyboard option not accepted
Product: [Fedora] Fedora Reporter: Tim Niemueller <tim>
Component: anacondaAssignee: David Shea <dshea>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 20CC: anaconda-maint-list, bphinz, dshea, edgar.hoch, g.kaviyarasu, jonathan, tim, twaugh, vanmeeuwen+fedora
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: anaconda-21.26-1 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2014-03-28 14:22:22 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Attachments:
Description Flags
working kickstart
none
Failing kickstart (cleaned up)
none
anaconda.log
none
anaconda.journal.log
none
anaconda.packaging.log
none
anaconda.program.log none

Description Tim Niemueller 2013-12-19 16:01:58 UTC 
Description of problem:
When using a kickstart for installation, the installer always requires me to enter the keyboard configuration screen once. Even if I just dismiss it right away with "Done" it does not go through with the automatic installation.

Version-Release number of selected component (if applicable):
anaconda-20.25.15-1.fc20.x86_64.rpm

How reproducible:
Always

Steps to Reproduce:
1. Have something like this in the kickstart file:
keyboard --vckeymap=de-nodeadkeys --xlayouts='de (nodeadkeys)'
2. Run a kickstart installation

Actual results:
The installation is interrupted and requires manual intervention to continue.

Expected results:
The installation should be fully automatic and no manual intervention should be necessary.

Additional info:
This is the Fedora 20 x86_64 installation DVD.
Comment 1 David Shea 2013-12-19 18:12:48 UTC 
Created attachment 839150 [details]
working kickstart

I completed an install on a VM using the attached kickstart. Please attach the complete kickstart that is failing (with passwords removed, of course) along with any log files from /tmp
Comment 2 Tim Niemueller 2013-12-20 20:47:30 UTC 
Created attachment 839785 [details]
Failing kickstart (cleaned up)
Comment 3 Tim Niemueller 2013-12-20 20:48:08 UTC 
I have attached the failing kickstart. Unfortunately the logs were gone after the reboot. If you need them I can re-do the install when back in the lab in January.
Comment 4 David Shea 2014-01-02 22:09:56 UTC 
The logs should be available in the installed system under /var/log/anaconda
Comment 5 Tim Niemueller 2014-01-08 15:50:43 UTC 
Created attachment 847216 [details]
anaconda.log
Comment 6 Tim Niemueller 2014-01-08 15:51:07 UTC 
Created attachment 847218 [details]
anaconda.journal.log
Comment 7 Tim Niemueller 2014-01-08 15:51:58 UTC 
Created attachment 847219 [details]
anaconda.packaging.log
Comment 8 Tim Niemueller 2014-01-08 15:52:25 UTC 
Created attachment 847221 [details]
anaconda.program.log
Comment 9 Tim Niemueller 2014-01-08 15:52:55 UTC 
I hope these are the logs you need.
Comment 10 David Shea 2014-03-03 22:39:44 UTC 
This appears to be a problem with the VNC server. The server is not accepting the keymap option, which can also be seen in interactive mode: configuring different layouts has no effect. I even tried using the setxkbmap binary to bypass libxklavier, and still no results. The keymap is always set to us, which, since it doesn't match what was configured, is why anaconda is stopping and asking for input.
Comment 11 Tim Waugh 2014-03-04 12:10:37 UTC 
VNC keymap handling is strange, as a result of the fact that the client tells the server "keysym up/down" -- keysym, not keycode. The server adjusts the keymap dynamically to deal with this.

Maybe it would make sense for anaconda not to check this when the X server is VNC?
Comment 12 David Shea 2014-03-04 14:27:07 UTC 
Fair enough, that'd work too. Taking the bug back.
Comment 13 David Shea 2014-03-04 21:32:20 UTC 
*** Bug 1053875 has been marked as a duplicate of this bug. ***