Bug 12118

Summary: Stage 1 first screen is garbage - boot-nonfree 64 bit
Product: Mageia Reporter: Morgan Leijström <fri>
Component: InstallerAssignee: Thierry Vignaud <thierry.vignaud>
Status: RESOLVED DUPLICATE QA Contact:
Severity: normal    
Priority: Normal    
Version: Cauldron   
Target Milestone: ---   
Hardware: x86_64   
OS: Linux   
Whiteboard:
Source RPM: CVE:
Status comment:
Attachments: Screenshot of stage 1 initial screen

Description Morgan Leijström 2013-12-27 01:53:06 CET 
The page is completely unreadable
Pressing enter begins sucessful installation

Reproducible: 

Steps to Reproduce:
Comment 1 Morgan Leijström 2013-12-27 02:00:53 CET 
Created attachment 4668 [details]
Screenshot of stage 1 initial screen

This is from a virtual machine.
It looks the same on my Thinkpad R61
Morgan Leijström 2013-12-27 02:01:04 CET

Priority: Normal => release_blocker

Comment 2 Thierry Vignaud 2013-12-27 07:05:20 CET 
Dup

*** This bug has been marked as a duplicate of bug 11602 ***

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

Comment 3 Manuel Hiebel 2013-12-27 10:48:01 CET 
What was your iso ? is it supposed to be fixed (and was the case in my test)
Comment 4 Morgan Leijström 2013-12-27 11:31:03 CET 
Freshly fetched (from mirror) hour before I reported.
This is probably the latest, as bug 12089 is not in this one.

boot-nonfree.iso, md5sum: aec53c870b5592620921bddc792732c3
Comment 6 Thierry Vignaud 2013-12-27 12:57:35 CET 
That's the MD5 of Beta2
See http://svnweb.mageia.org/web/www/trunk/en/downloads/get/definitions.ini?view=markup&pathrev=2878#l572

It's old and has been fixed since.

Priority: release_blocker => Normal
Severity: major => normal

Comment 7 Morgan Leijström 2013-12-27 14:38:41 CET 
Argh, yes i did not notice the download page directed me to sunet.se, and i did not check the files was recent.  ( and i already knew - bug 12062  Doh.)

I use a much more updated mirror for everything i set manually and also fetched my former boot iso manually, but now i got lazy and used the download page.

So this is old, yes. Thanks for checking.


Hmmm.... this raises the question if testers and developers should always use distrib-coffee directly.   And cauldron need not be mirrored.

And/or there should be some mechanism always used that check datestamp of used mirror, alerting user.
Comment 8 Manuel Hiebel 2013-12-27 17:09:31 CET 
(In reply to Morgan Leijström from comment #7)
> Argh, yes i did not notice the download page directed me to sunet.se, and i
> did not check the files was recent.  ( and i already knew - bug 12062  Doh.)

I missed to removed the no synced mirrors for the download page, will check.
(for the main mirror api cf your bug I can do nothing)
Comment 9 James Kerr 2013-12-27 17:46:55 CET 
(In reply to Morgan Leijström from comment #7)

>if testers and developers should always use distrib-coffee directly. 

There are many mirrors that are up to date. You should check the mageia_timestamp file on the mirror from which you are downloading.

If you want to test the pre-release iso's then you need to be registered as a tester:

https://wiki.mageia.org/en/Pre-release_ISO_testing

Those who are registered as iso testers are provided with a password protected url from which pre-release versions may be downloaded.
Comment 10 Morgan Leijström 2013-12-27 22:40:49 CET 
Yes we testers could be aware to check datestamps.

But lagging mirrors delay for all users, also including security updates.
I* have also in th past had problems with (same i think) mirror for long time not updating all packages, leading to unsatisfied dependencies.
Such problems exists for normal users.
Therefor i think urpm* should check the timestamp automatically.