Bug 855349

Summary:	crash during update
Product:	[openSUSE] openSUSE 13.1	Reporter:	Arvin Schnell <aschnell>
Component:	YaST2	Assignee:	Josef Reidinger <jreidinger>
Status:	RESOLVED WONTFIX	QA Contact:	Jiri Srain <jsrain>
Severity:	Normal
Priority:	P5 - None	CC:	aschnell, lslezak
Version:	Final
Target Milestone:	---
Hardware:	x86-64
OS:	Other
Whiteboard:
Found By:	Development	Services Priority:
Business Priority:		Blocker:	---
Marketing QA Status:	---	IT Deployment:	---
Attachments:	YaST logs

Description Arvin Schnell 2013-12-13 11:37:17 UTC

I did a update from 13.1 final x86 to 13.1 final x86-64. YaST did show
two warning that the resulting system might not work but otherwise the
update started fine. Half way though it crashed.

The update was started from DVD.

Comment 1 Arvin Schnell 2013-12-13 11:37:46 UTC

Created attachment 571705 [details]
YaST logs

Comment 2 Arvin Schnell 2014-01-08 11:49:37 UTC

Backtrace in file 'signal' point to libpy2lang_ruby.so and libruby2.0.so.

Comment 3 Josef Reidinger 2014-01-08 12:05:03 UTC

lslezak - it fails somewhere in package callbacks. Do you have idea what can crash or do you already see such issue? It can also fail because ruby will switch from x86 to x86_64

Comment 4 Ladislav Slezák 2014-01-08 13:57:09 UTC

Um, I have no idea what might went wrong, but architecture change is not supported anyway so if I were you I'd not spend much time here (and close it as INVALID, sorry...).

Comment 5 Josef Reidinger 2014-01-08 13:59:25 UTC

OK, it is not supported and there is many possible issues, so closing it as invalid as there is more issues we have with yast with higher priority.

Comment 6 Arvin Schnell 2014-01-08 14:16:42 UTC

I don't see how the architecture change could cause a crash: YaST runs
from DVD completely inside a 64 bit system, whether it updates/installs
32 or 64 bit RPMs shouldn't matter.

Comment 7 Ladislav Slezák 2014-01-09 15:32:11 UTC

It depends, some code could be run from the target system (via SCR). But if it crashes in pkg callbacks then it's probably related to some other problem. I agree, it should work in this case.

Unfortunately I currently have other more urgent tasks and I cannot spend time debugging this rather unsupported scenario... I'm really sorry :-(