Re: reinstalling X.Org server
On Fri, Jan 30, 2009 at 6:01 PM, Lennart Sorensen
<lsorense@csclub.uwaterloo.ca> wrote:
> On Fri, Jan 30, 2009 at 04:29:24PM -0600, Ron Johnson wrote:
>> On 01/30/2009 04:18 PM, Francesco Pietra wrote:
>> >Is it possible to reinstall X.Org server on a multisocket dual-opteron
>> >computer running amd64 lenny?
>> >
>> >Why? Failure of a DIMM bank did no more allow to launch amd64. Removed
>> >the faulty slots, amd64 could be launched and the system seemed to be
>> >in order. Filled the empty bank with fresh DDR1 ECC, the total memory
>> >increased correspondingly. Again, it was now possible to carry out
>> >sophisticated chemical calculations.
>> >
>> >However, needing now to check 3D molecular structures, I found that
>> >"startx" does not launch X anymore, X only appearing as a flash at the
>> >bottom of the screen. I checked many X-related files against a i386
>> >lenny, ssh linked, computer (in particular /etc/X11/xorg.conf), don't
>> >detecting damaged lines. I did not carry out a checksum. Command:
>> >
>> >tail --lines 200 /var/log/Xorg.0.log|grep EE
>> >
>> >was not much informative, as shows below between === lines.
>> >
>> >I suspect that one or more X-related files were damaged as a
>> >consequence of the RAM problems above.
>> >
>> >Thanks
>> >francesco pietra.
>> >
>> >========================
>> >(II) MACH64(0): Not using default mode "1920x1440" (insufficient
>> >memory for mode)
>> >(II) MACH64(0): Not using default mode "960x720" (bad mode
>> [snip]
>> >drmOpenDevice: node name is /dev/dri/card0
>> >drmOpenDevice: open result is -1, (No such device or address)
>> >drmOpenDevice: open result is -1, (No such device or address)
>> >drmOpenDevice: Open failed
>> >
>> >Backtrace:
>> >0: X(xf86SigHandler+0x6a) [0x48dd0a]
>> >1: /lib/libc.so.6 [0x2b1686b4ef60]
>> >
>> >Fatal server error:
>> >Caught signal 11. Server aborting
>> >======================
>>
>> I'd reinstall the whole system.
>
> For an ati fglrx driver module missing?
>
> Seems like overkill.
>
> The ATI driver loves to crash/segfault if anything displeases it (like
> running 8bit colour mode, which some versions default to).
Does fglrx support the Mach64 chip?
Reply to: