https://bugzilla.novell.com/show_bug.cgi?id=288730
Summary: Hardware settings: aspect ratio of screen crashes X Product: openSUSE 10.3 Version: Alpha 5 Platform: i686 OS/Version: Linux Status: NEW Severity: Normal Priority: P5 - None Component: YaST2 AssignedTo: bnc-team-screening@forge.provo.novell.com ReportedBy: danielstefanmader@web.de QAContact: jsrain@novell.com Found By: Beta-Customer
When performing the installation (DVD medium), the final step is Hardware settings. Being curious what the aspect ratio settings offers I clicked this. Since I was unsure what diagonal size my 15" laptop screen has I pressed Cancel. The Xsession crashed, restarted and offered the login screen right away. I rebooted and received a complaint about an unfinished previous installation :)
Looks like the sax2 component is not very tolerant towards errors or screens not reporting their size?
https://bugzilla.novell.com/show_bug.cgi?id=288730#c1
Matej Horvath mhorvath@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |danielstefanmader@web.de
--- Comment #1 from Matej Horvath mhorvath@novell.com 2007-07-04 07:23:23 MST --- Please attach your yast logs (http://en.opensuse.org/Bugs/YaST), /var/log/Xorg.0.log, /var/log/messages and output of 'hwinfo --monitor --gfxcard'.
https://bugzilla.novell.com/show_bug.cgi?id=288730#c2
--- Comment #2 from Daniel Mader danielstefanmader@web.de 2007-07-04 10:21:13 MST --- Created an attachment (id=149700) --> (https://bugzilla.novell.com/attachment.cgi?id=149700) /var/log/messages
linux-01q6:~ # hwinfo --monitor 29: None 00.0: 10000 Monitor [Created at fb.71] Unique ID: rdCR.EY_qmtb9YY0 Hardware Class: monitor Model: "Generic Monitor" Vendor: "Generic" Device: "Monitor" Resolution: 800x600@75Hz Driver Info #0: Max. Resolution: 800x600 Vert. Sync Range: 50-90 Hz Hor. Sync Range: 31-48 kHz Config Status: cfg=new, avail=yes, need=no, active=unknown
https://bugzilla.novell.com/show_bug.cgi?id=288730#c3
--- Comment #3 from Daniel Mader danielstefanmader@web.de 2007-07-04 10:22:30 MST --- linux-01q6:~ # hwinfo --gfxcard 13: PCI 02.0: 0300 VGA compatible controller (VGA) [Created at pci.296] UDI: /org/freedesktop/Hal/devices/pci_8086_3582 Unique ID: _Znp.WoJcbu5eHk4 SysFS ID: /devices/pci0000:00/0000:00:02.0 SysFS BusID: 0000:00:02.0 Hardware Class: graphics card Model: "Acer Incorporated [ALI] 855 GM" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x3582 "855 GM" SubVendor: pci 0x1025 "Acer Incorporated [ALI]" SubDevice: pci 0x003d Revision: 0x02 Memory Range: 0xb0000000-0xb7ffffff (rw,prefetchable) Memory Range: 0xf0000000-0xf007ffff (rw,non-prefetchable) I/O Ports: 0xe000-0xe007 (rw) IRQ: 10 (13396 events) I/O Ports: 0x3c0-0x3df (rw) Module Alias: "pci:v00008086d00003582sv00001025sd0000003Dbc03sc00i00" Driver Info #0: XFree86 v4 Server Module: i810 Driver Info #1: XFree86 v4 Server Module: i810 3D Support: yes Extensions: dri Config Status: cfg=no, avail=yes, need=no, active=unknown
14: PCI 02.1: 0380 Display controller [Created at pci.296] UDI: /org/freedesktop/Hal/devices/pci_8086_3582_0 Unique ID: ruGf.WqkU0AHb1z9 SysFS ID: /devices/pci0000:00/0000:00:02.1 SysFS BusID: 0000:00:02.1 Hardware Class: graphics card Model: "Acer Incorporated [ALI] 855 GM" Vendor: pci 0x8086 "Intel Corporation" Device: pci 0x3582 "855 GM" SubVendor: pci 0x1025 "Acer Incorporated [ALI]" SubDevice: pci 0x003d Revision: 0x02 Memory Range: 0x30000000-0x37ffffff (rw,prefetchable) Memory Range: 0x38000000-0x3807ffff (rw,non-prefetchable) Module Alias: "pci:v00008086d00003582sv00001025sd0000003Dbc03sc80i00" Driver Info #0: XFree86 v4 Server Module: i810 Driver Info #1: XFree86 v4 Server Module: i810 3D Support: yes Extensions: dri Config Status: cfg=no, avail=yes, need=no, active=unknown
https://bugzilla.novell.com/show_bug.cgi?id=288730#c4
--- Comment #4 from Daniel Mader danielstefanmader@web.de 2007-07-04 10:24:46 MST --- Created an attachment (id=149703) --> (https://bugzilla.novell.com/attachment.cgi?id=149703) /var/log/Xorg.0.log
https://bugzilla.novell.com/show_bug.cgi?id=288730#c5
--- Comment #5 from Daniel Mader danielstefanmader@web.de 2007-07-04 10:26:43 MST --- Created an attachment (id=149704) --> (https://bugzilla.novell.com/attachment.cgi?id=149704) /var/log/Xorg.0.log.old
https://bugzilla.novell.com/show_bug.cgi?id=288730#c6
--- Comment #6 from Daniel Mader danielstefanmader@web.de 2007-07-04 10:27:50 MST --- Created an attachment (id=149705) --> (https://bugzilla.novell.com/attachment.cgi?id=149705) Yast logs created with 'save_y2logs /tmp/y2logs.tgz'
https://bugzilla.novell.com/show_bug.cgi?id=288730
Matej Horvath mhorvath@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|bnc-team- |ms@novell.com |screening@forge.provo.novell| |.com | Status|NEEDINFO |NEW Info Provider|danielstefanmader@web.de |
https://bugzilla.novell.com/show_bug.cgi?id=288730#c7
Marcus Schaefer ms@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|ms@novell.com |jdsn@novell.com
--- Comment #7 from Marcus Schaefer ms@novell.com 2007-07-09 04:51:39 MST --- I don't think the X-Server crashed because there is no X-Server access while setting up aspect and ratio. The result of this setup is just a line in xorg.conf which takes into effect at next restart of the X-Server. I think it is more likely that yast was exited for whatever reason and so the running X-Server is closed and the installation is "finished"
Unfortunately the logs doesn't point me to something useful. The X-Server log is clean which means there is no X-server problem. The yast log told me:
clients/x11_proposal.ycp:65 x11 package status is: <true> clients/x11_proposal.ycp:99 Loading library cache... clients/x11_proposal.ycp:105 Reading libsax cache data... clients/x11_proposal.ycp:214 X11: DualHead Mode not supported with this NoteBook [Interpreter] x11/x11Dialog.ycp:360 Called YaST client returned: $["links":["sax-monitor", "sax-resolution", "sax-displaysize", "sax-colors", "sax-3d", "sax-ext-vga", "sax-test"], "preformatted_proposal":"<ul><li>Graphics card: Intel 855 GM</li><li>Monitor: <a href="sax-monitor">--> LCD 800X600@60HZ</a> (<a href="sax-displaysize">Aspect Unconfigured</a>) - (<a href="sax-resolution">800x600</a>)</li><li>Colordepth: <a href="sax-colors">24</a> bit (16.7 Mio. Colors)</li><li>3D Acceleration: <a href="sax-3d">Deactivated</a></li></ul>"] 2007-06-30 11:38:21 <1> linux-01q6(4228) [wfm] Y2CCWFM.cc(createInLevel):153 Parsing finished
which looks good to me in the first place. Maybe yast wasn't able to write an error to the log file... just a wild guess
assigning to new maintainer of yast2-x11
https://bugzilla.novell.com/show_bug.cgi?id=288730#c8
J. Daniel Schmidt jdsn@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |ms@novell.com
--- Comment #8 from J. Daniel Schmidt jdsn@novell.com 2007-08-28 10:04:45 MST --- @Marcus: I think I found the cause - but need your advice here.
Have a quick look at y2log (from comment #6) line 2529. It says 'No item "undef" exiting'. (The following couple of lines come from yast as it exits and reboots, see the timestamp.)
This "undef" comes from XLib.pm::getDisplaySize line 236, @result = ("undef"). The following code checks if $traversal is defined and calculates the display size. There is no else part for that if clause. So if $traversal is undefined @result remains "undef" and returns it. This is anyhow a valid return value, as the ycp expects a string, but XLib in return expects a integer when setDisplaySize is called with this string.
Do you have an advice what to do in such an else part. What should be the default here? I could check the string before calling setDisplaySize and set it to your fallback (in x11_proposal.ycp, line 285) in case it is an invalid string ("undef" is a valid string, but invalid in terms of containing only numbers). The latter is ugly because XLib.pm::getDisplaySize remains buggy - so I prefer to patch for XLib.pm.
Comment please.
https://bugzilla.novell.com/show_bug.cgi?id=288730#c9
Marcus Schaefer ms@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |ms@novell.com Status|NEEDINFO |NEW Info Provider|ms@novell.com |
--- Comment #9 from Marcus Schaefer ms@novell.com 2007-08-29 03:38:52 MST --- I would say a patch for XLib.pm::setDisplaySize() checking for valid input would be the best solution. If there are no values to set a valid display size I would avoid the $mDesktop->setDisplayRatioAndTraversal... call
https://bugzilla.novell.com/show_bug.cgi?id=288730#c10
J. Daniel Schmidt jdsn@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED
--- Comment #10 from J. Daniel Schmidt jdsn@novell.com 2007-09-01 00:58:49 MST --- I tried to add checks to all functions that pass along these strings, that want to be integers, but that broke the proposal completly.
So it remains only the check in XLib.pm. If one avoids the call of $mDesktop->setDisplayRatioAndTraversal... then the module fails to finish a correct setup, so I return to default values in this case.
fixed in yast2-x11-2.15.11
-
bugzilla_noreply＠novell.com