Thanks for your input, Ray. On Fri, 11 Jun 2004, Ray Olszewski wrote:
> The first thing you need to pin down is whether this is an X problem or > some sort of USB problem (possibly one involving a kernel module). > > To do that, see if you can boot successfully to a command prompt. The > simplest way to do that is to use your emergency procedure to get access to > the hard disk, then rename the file /etc/rc2.d/S99xdm to something that > will not execute (I usually change it to No99xdm). Then see if the system > will boot/init successfully. The system boots successfully after having renamed S99xdm NoS99xdm. I get a console I can log into the system with and do things in text mode. Having fired up Pine, I'm writing this message using the system. When I issue startx from the command line, I get the dreaded "no signal" output on the monitor followed by blank screen and hung system. So, it does look like an X problem. > If it does, then your problem is with X, and we need to explore that more > closely. > > If not, then your problem is probably with the recent changes you made, and > we need to examine them more closely. > > (Another option is to see if you can telnet or ssh to the system after it > boots/inits but apparently "hangs". This could work if the problem is just > an unusually nasty console error, but not if it is a kernel problem. Also, > I don't know if your system is on a LAN and runs the needed daemons for > these connections.) As soon as I send off this message, I'll issue startx, let the screen go blank and system freeze, then try to telnet from another machine (I am on a LAN and a telnet server is running). > (Another option, if your lilo setup permits it, is to boot into single-user > mode. This will not run the rc2 scripts so will not start xdm.) I would very much like to be able to do this. When this problem first occured, I did a bit of web searching to see how I could make lilo do this. That initial attempt at finding information failed: do you have any pointers on doing this? I really should have that sort of "failsafe" option in my lilo menu. > I cannot find any Sid package with the name "usbitem" or (if I check "usb") > anything close to it, so I suspect another paraphrase. If so, and if the > problem seems to be with usb, please be specific about which usb packages > you've installed. I think the problem is not with usb. Nevertheless, I'll note that I installed (apt-get install) 4 usb packages: from memory these were 1) usb-utils; 2) hotplug; 3) usbmgr; and 4) usbview. Since installing those had no evident effect on my attempt to make my usb flash-reader accessible, I added a line to /etc/fstab: usbdevfs /proc/bus/usb usbdevfs defaults 0 0 (found reference to this on the web). After rebooting and having the problem, I accessed /etc/fstab and commented out that line, but I was nonetheless still unable to fully boot the system as described. > In partlcular, since you quote log output referencing keyboard problems, > might you have installed (for example) "console-keymaps-usb"? Not by intent. Could it have been installed as a dependency for one of the usb programs I installed? > Or have you installed any "usb-modules-*" package ... these install kernel > modules that could cause problems ... a kernel has no real protection > against a module bug, since modules when run become part of the kernel. Again, I did not knowingly install such a thing. Whether it got installed as a dependency of one of the other programs I installed remains a possibility. Here goes a try at telnetting into the frozen system . . . James > > At 09:01 PM 6/11/2004 +0000, James Miller wrote: > >I've somehow managed to foul up my Debian system such that it won't "go > >graphical" and then somehow freezes up. Since I can't see any screen > >output, I can't tell very well what's going on. I'm sort of mystified as > >to how and why this happened: I recently installed some USB stuff (apt-get > >install usbitem), and I just can't understand how that would foul up my > >display. I haven't installed any X related programs lately, nor have I > >fiddled with any display settings. What happens is that I boot the system > >as normal (let lilo start the default kernel), the machine goes through > >its bootup routine, displaying messages and then, at the stage where I > >expect the XDM login window to appear, my monitor goes blank with a "no > >signal" message. Any keyboard input I've tried has no effect on the > >machine: ctrl-alt-Fx won't get me any virtual terminals, ctrl-alt-bkspc > >does not kill X (is it even started?) and dump me at a command prompt, and > >ctrl-alt-del won't reboot the system. I looked at /var/log/messages after > >booting the system from a Knoppix disk (thank God for Klaus and co.!), and > >the KDE it boots into looks prfectly normal (except not as fine a > >resolution as I'd like). Some /var/log/messages output that looks > >significant is as follows: > > > >Jun 11 13:58:38 debian kernel: mtrr: 0xd0000000,0x2000000 overlaps > >existing 0xd0000000,0x1000000 > >Jun 11 13:58:38 debian kernel: [drm] Initialized radeon 1.9.0 20020828 on > >minor 0 > >Jun 11 13:58:38 debian kernel: mtrr: 0xd0000000,0x2000000 overlaps > >existing 0xd0000000,0x1000000 > >Jun 11 13:58:38 debian kernel: agpgart: Found an AGP 2.0 compliant device > >at 0000:00:00.0. > >Jun 11 13:58:38 debian kernel: agpgart: Putting AGP V2 device at > >0000:00:00.0 into 1x mode > >Jun 11 13:58:39 debian kernel: atkbd.c: Unknown key released (translated > >set 2, code 0x7a on isa0060/serio0). > >Jun 11 13:58:39 debian kernel: atkbd.c: This is an XFree86 bug. It > >shouldn't access hardware directly. > >Jun 11 13:58:39 debian kernel: atkbd.c: Unknown key released (translated > >set 2, code 0x7a on isa0060/serio0). > >Jun 11 13:58:39 debian kernel: atkbd.c: This is an XFree86 bug. It > >shouldn't access hardware directly. > > > >Not sure why this problem suddenly crops up, or what to do about it. I've > >thought of upgrading the X server as one possible solution, but don't want > >to just take wild stabs like that. So, I ask for input here. Does this > >output mean anything to anyone here? Any suggestions on what to do about > >it? I can provide more output from /var/log/messages if this is not the > >relevant part. Any help will be appreciated. For now, when I get stuck > >at this point in the boot process I can only power the machine off with > >the power switch. > > > >Thanks, James > > > - > To unsubscribe from this list: send the line "unsubscribe linux-newbie" in > the body of a message to [EMAIL PROTECTED] > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.linux-learn.org/faqs - To unsubscribe from this list: send the line "unsubscribe linux-newbie" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.linux-learn.org/faqs
