Bug 4370

Summary: BootCD (NetInstall) doesn't support Xen virtual HW (xen-vbd.ko xen-vnif.ko / xen_blkfront xen_netfront)
Product: Mageia Reporter: Jaromír Cápík <tavvva>
Component: InstallerAssignee: Mageia Bug Squad <bugsquad>
Status: RESOLVED INVALID QA Contact:
Severity: normal    
Priority: Normal CC: marja11, tmb
Version: 1   
Target Milestone: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Source RPM: CVE:
Status comment:

Description Jaromír Cápík 2012-01-31 23:22:58 CET
Description of problem:
When I try to install Mageia as a Xen Guest, the network interface is not detected and the installation can't continue. Please, build the kernel with the mentioned Xen modules.
Comment 1 Thomas Backlund 2012-02-01 16:03:36 CET
The server kernels has xen support.

this kernel is available on boot.iso

just type: alt1

at boot prompt and try again.

CC: (none) => tmb

Comment 2 Jaromír Cápík 2012-02-04 17:37:14 CET
Hello Thomas.

Sorry. The problem seems to be in the web interface to the server console. It's written in Java and possibly using the internal Sun API. It doesn't work with OpenJDK and even with SunJDK the behaviour is unpleasant. In my case the screen is not refreshed after the power-on (for unknown reason) and the syslinux initial screen is not displayed. I got a screen refresh after the initial 15s delay and that's why I thought there's no choice. But this works well when I try that in VirtualBox. Unfortunately I had to migrate my server without longer outages and that forced me to install Debian there. Debian installer displays a graphical menu and that was displayed correctly (the refresh was probably triggered by the resolution change .... don't now exactly and can't test that since the server is already in production). Could switching from syslinux to grub help prevent that? Or ... spliting the boot CD to 2CDs for Workstation and Server? Just random thoughts ....
I believe this bug can be closed since this was somehow my fault.

Thanks.
Jaromir.
Comment 3 Marja Van Waes 2012-03-25 21:39:43 CEST
(In reply to comment #2)

> I believe this bug can be closed since this was somehow my fault.
> 

Thanks for the feedback :)

Closing

Status: NEW => RESOLVED
CC: (none) => marja11
Resolution: (none) => INVALID