running X w/serial console (XDCMP)
I was wanting to run gdm w/XDCMP on my ultra 1 which has a serial console, I can get the GDM login screen but cannot get past that, the logs say X failed to start, when I tried to configure X it still wouldn't load saying /dev/tty0 (or something like that) did not exist. is there a way to get GDM to do this? maybe it's a config option, I don't think I need to run a local X server if I am logging into XDCMP? (local as in on the sparc). Haven't ever tried this configuration before. or if there is a way to run a local X server w/serial console I could try that too, maybe tricking it into thinking /dev/tty0 is there somehow? thanks!! nate
E450 installation problems
Hi! I have an Ultra 450 Enterprise Server and I'm trying to install Debian 3.0 for Sparc to it. Here is what I do. 1. I make an iso image of the debian/sparc boot installation cd using jigdo-lite in the windows platform. 2. I have burnt the iso image with a variety of burning programs such as Veritas, Nero and ECDC in maximum speed (that means 16x) 3. I do a Stop-A, insert the installation CD and type boot cdrom. The CD boots. 4. I have 3 disks of 17Gb each. I format the first one with 512Mb swap and the rest ext3. The two others are formatted as ext3 each. 5. I mount the swap and the / partition on the first disk. I mount the /home partition on the second disk and a /storage on the third disk. Problems : 6. On the load kernel modules menu I receive a Modules cannot loaded error or sth like that. 7. The rest of the installation continues without problems. 8. ...make system bootable and 9. ... reboot! When the system is rebooted it cannot find the SILO or anything else! And the weird is that it boots the previous OS (Solaris 9)!!?? Any ideas of what am I missing? Thanx alot for your time! Whysohappy _ MSN 8 with e-mail virus protection service: 2 months FREE* http://join.msn.com/?page=features/virus
Some issues with UltraSparc 10
Hi,
I just installed an UltraSparc 10 (256MB mem / 333MHz single processor)
with woody. All went fine so far but I found the following log entries:
Jun 4 17:53:01 ehec2 /USR/SBIN/CRON[10752]: (mail) CMD ( if [ -x
/usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
Jun 4 18:00:42 ehec2 kernel: hda: timeout waiting for DMA
Jun 4 18:00:42 ehec2 kernel: ide_dmaproc: chipset supported ide_dma_timeout
func only: 14
Jun 4 18:00:42 ehec2 kernel: hda: status error: status=0x58 { DriveReady
SeekComplete DataRequest }
Jun 4 18:00:42 ehec2 kernel: hda: drive not ready for command
Jun 4 18:00:42 ehec2 kernel: hda: status timeout: status=0xd0 { Busy }
Jun 4 18:00:42 ehec2 kernel: hda: drive not ready for command
Jun 4 18:00:42 ehec2 kernel: ide0: reset: success
Jun 4 18:02:20 ehec2 kernel: hda: timeout waiting for DMA
Jun 4 18:02:20 ehec2 kernel: ide_dmaproc: chipset supported ide_dma_timeout
func only: 14
Jun 4 18:02:20 ehec2 kernel: hda: status error: status=0x58 { DriveReady
SeekComplete DataRequest }
Jun 4 18:02:20 ehec2 kernel: hda: drive not ready for command
Jun 4 18:02:20 ehec2 kernel: hda: status timeout: status=0xd0 { Busy }
Jun 4 18:02:20 ehec2 kernel: hda: drive not ready for command
Jun 4 18:02:20 ehec2 kernel: ide0: reset: success
Jun 4 18:08:01 ehec2 /USR/SBIN/CRON[20094]: (mail) CMD ( if [ -x
/usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
Jun 4 18:23:01 ehec2 /USR/SBIN/CRON[20121]: (mail) CMD ( if [ -x
/usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
Jun 4 18:23:52 ehec2 kernel: hda: timeout waiting for DMA
Jun 4 18:23:52 ehec2 kernel: ide_dmaproc: chipset supported ide_dma_timeout
func only: 14
Jun 4 18:23:52 ehec2 kernel: hda: status error: status=0x58 { DriveReady
SeekComplete DataRequest }
Jun 4 18:23:52 ehec2 kernel: hda: drive not ready for command
Jun 4 18:23:52 ehec2 kernel: hda: status timeout: status=0xd0 { Busy }
Jun 4 18:23:52 ehec2 kernel: hda: drive not ready for command
Jun 4 18:23:52 ehec2 kernel: ide0: reset: success
Jun 4 18:32:18 ehec2 kernel: hda: timeout waiting for DMA
Jun 4 18:32:18 ehec2 kernel: ide_dmaproc: chipset supported ide_dma_timeout
func only: 14
Jun 4 18:32:18 ehec2 kernel: hda: status error: status=0x58 { DriveReady
SeekComplete DataRequest }
Jun 4 18:32:18 ehec2 kernel: hda: drive not ready for command
Jun 4 18:32:18 ehec2 kernel: hda: status timeout: status=0xd0 { Busy }
Jun 4 18:32:18 ehec2 kernel: hda: drive not ready for command
Jun 4 18:32:18 ehec2 kernel: ide0: reset: success
Jun 4 18:38:02 ehec2 /USR/SBIN/CRON[20164]: (mail) CMD ( if [ -x
/usr/sbin/exim -a -f /etc/exim/exim.conf ]; then /usr/sbin/exim -q ; fi)
These were the only hda - ide related entries over 12 hours and I wonder if this
is a reason to think about.
The next problem is more critical and I think I have to file a bug report
about it but I just want to make sure that I did not made a stupid error.
Currently the box is running the install kernel. I wanted to upgrade it to
kernel-image-2.4.19-sun4u. Please see the output in the end of the mail
for the result.
Kind regards
Andreas.
Reading Package Lists...
Building Dependency Tree...
The following NEW packages will be installed:
kernel-image-2.4.19-sun4u
0 packages upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
Need to get 0B/3818kB of archives. After unpacking 10.4MB will be used.
Selecting previously deselected package kernel-image-2.4.19-sun4u.
(Reading database ... 10612 files and directories currently installed.)
Unpacking kernel-image-2.4.19-sun4u (from
.../kernel-image-2.4.19-sun4u_26_sparc.deb) ...
Setting up kernel-image-2.4.19-sun4u (26) ...
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/arch/sparc64/solaris/solaris.o
depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/drivers/block/nbd.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/bluetooth/hci_uart.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/bluetooth/hci_usb.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/bluetooth/hci_vhci.o
depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/drivers/char/lp.o
depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/drivers/fc4/fc4.o
depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/drivers/fc4/soc.o
depmod: *** Unresolved symbols in /lib/modules/2.4.19/kernel/drivers/fc4/socal.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/ide/ataraid.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/ide/hptraid.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/ide/ide-floppy.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/ide/ide-tape.o
depmod: *** Unresolved symbols in
/lib/modules/2.4.19/kernel/drivers/ide/pdcraid.o
depmod: *** Unresolved symbols in
video drivers
Hi, I have a 7(10 possible)/1 bet at present. I'm hoping someone in the debian sparc community can narrow the odds :-) My choice of video driver for my Ultra 1 offered as follows: ati (I don't think so) fbdev (likewise) glint (no) sunbw2 (I doubt it) suncg14, 3, 6 (?) sunffb (again ?) sunleo (and again?) suntcx (???) If anyone can offer assistance I'd be very grateful. John
Re: [debian-sparc] video drivers
Hello, On 5 Jun 2003, John Hedge wrote: I have a 7(10 possible)/1 bet at present. I'm hoping someone in the debian sparc community can narrow the odds :-) My choice of video driver for my Ultra 1 offered as follows: [...] That depends on which video card is present in your Ultra1. If it's a Creator or a Creator3D, then you have to use sunffb. -- Erwann ABALEA [EMAIL PROTECTED] - RSA PGP Key ID: 0x2D0EABD5 - SJ Just do it. TM Traduction ? Bouge toi le cul. -+- RMD in http://neuneu.mine.nu : Move your ass -+-
Lite-On CD-RWs vs. Ultra 5s
And the loser appears to be -- anyone trying this combination. I installed a Lite-On 52x24x52x burner in my Ultra 5. It works fine as a CD-ROM in both IDE and SCSI emulation modes, but it causes total system lockups every time I attempt to burn with it. The last line of output from cdrecord is ide-scsi: The scsi wants to send us more data than expected - discarding data. There was a short thread back in January where Dustin Lundquist had the same problem with two different models of Lite-On CD-RWs in his U5. However, Jacob Helwig mentioned that he had a successfully running Lite-On 52x24x52x in his U10. Since U5s and U10s are essentially the same machine (same system board, at least), I'm not sure why this drive would work in one but not the other. I gave up on the Lite-On and installed my older HP 9100 8x4x32x CD-RW. I had already set up SCSI emulation, and the HP worked fine with no additional set-up. That rules out any of the typical SCSI emulation/CD-Writing-HOWTO issues. It appears that there is some conflict between cdrecord on an Ultra 5 and Lite-On drives. Any ideas what this could be, or why a U10 wouldn't have a problem but a U5 does? Could it be a firmware issue with the CD-RW? Additional info: I'm running Debian 3.0r1 stable with a custom 2.4.20 kernel, and cdrecord-1.10. All of the scsi/ide cd drivers are modules, with the appropriate ignore=hdc option in /etc/modules.conf. My U5 is 2000-vintage, 400MHz w/ latest OBP and the 375-0115 system board. The CD-RW is an LTR-52246S. Thanks for any help- Justin _ Add photos to your messages with MSN 8. Get 2 months FREE*. http://join.msn.com/?page=features/featuredemail
Re: auto power-off on a SPARC LX (was Re: Ethernet lockup)
On Wed, 4 Jun 2003, Eric Brower wrote:
The Ultra patch is not applicable. The SPARC kernel will not power-down
a system with a serial console attached; from process.c:
The SPARC LX in question is completely headless... no keyboard/monitor,
and no serial attached... only power, SCSI, and network... I am curious as
to when such changes took place in the kernel, as things were working as
expected with the 2.2.x kernels.
void machine_power_off(void)
{
#ifdef CONFIG_SUN_AUXIO
if (auxio_power_register !serial_console)
*auxio_power_register |= AUXIO_POWER_OFF;
#endif
machine_halt();
}
If this implementation bothers you, consider coding-up a sysctl to
modify this behavior.
Is this a new addition to the kernel for the 2.4.x series?
Or, would you recommend any PROM settings I could set to get the current
default behavior of my LX to really power-off when I shutdown?
---
Matthew Haas Squirrel's Law #209203923: Thou shalt ignoreth
Corning/Geneseo, NY all RAM specifications, they're noo scottish
Re: running X w/serial console (XDCMP)
Matthew French said: Check /etc/inittab. The tty's are managed from there. It may be that you have getty tty0 when you shouldn't - or the other way round. I presume you have checked that /dev/tty0 exists? The other thing that bothers me is that with X it should be using /dev/tty7. it does exist, but I think since I am using a serial console the hardware disables the port(no keyboard/mouse/monitor connected) I run my Ultra5 with a serial console, and X does not start up. But then, I have no keyboard attached, so it is not too surprising X does not start. so maybe it's just going to be getting gdm not to try to start X locally.. nate
Re: running X w/serial console (XDCMP)
Martin said: I am slightly confused at what you're trying to do. XDCMP is commonly used over IP - not serial links. Serial link may not have the bandwidth nessecary to do this properly. well what I meant is there is no keyboard/mouse on the machine at the moment, the console is on the serial port, I am connecting to gdm via XDCMP over TCP/IP (connecting works fine, I get the gdm login screen, just when I try to login, gdm cycles and the log says it cannot start X) Posting the exact error message would be a lot more useful - does it say /dev/tty0 doesn't exist or /dev/ttyS0 doesn't exist? from the gdm log: Jun 5 02:18:40 sparky gdm[4616]: gdm_slave_xioerror_handler: Fatal X error - Re starting :0 from plain ol startx: Fatal server error: xf86OpenConsole: Cannot open /dev/tty0 (No such file or directory) (I think because the hardware disables the keyboard console when no keyboard is plugged in.. there is a device file /dev/tty0 ) You will need to run an XServer on whichever machine is controlling the display hardware. You will also need the correct drivers for that hardware. yes I have that setup already..just for some reason gdm wants to run an X server on the machine gdm is on, rather then realize I have an X server on the remote side.. You don't log in to XDCMP. It is a protocol that identifies which servers are willing to accept X sessions over a network. well connecting via XDCMP, logging in via gdm(I see the login screen and can authenticate ..) Then why not network the Ultra 1 up to another machine and get comfortable using GDM/XDCMP and running remote displays in a more standard config. it is .. :/ thanks! I suspect this is a gdm configuration issue ... nate
Re: running X w/serial console (XDCMP)
nate said: I was wanting to run gdm w/XDCMP on my ultra 1 which has a serial console, I can get the GDM login screen but cannot get past that, I guess it's impossible to run gdm w/o a local X server. ever since I instaleld the software to support the local X server gdm won't load anymore. just tried removing all X-related packages and reinstalling gdm and afterstep but that didn't help ohwell. woulda been nice. not sure why/how it worked the first time with no local X server. nate
Re: SCSI ERRORS
I've seen that error on my SS10 with two disks as well. Do you have an external device (e.g. CD-ROM, Streamer) connected ? That error seemed to happen less frequently after I disconnected my external CD-ROM drive. that's the point where scsi gets a bit esoteric: a bit to long cables, a terminator that is not completely ok, some plugs that are dirty and you have exactly such problems. try to reorder the devices, use another cable and/or terminator usw. it is possible that this helps... one time i had such a problem it was the temperature of the devices and/or cables. turning some fans resolved it... gruess vom emi
if you are in iowa, you can have a nice large machine
Hi i won this machine, http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItemitem=3027472460category=11216rd=1 it weights more than i can deal with, and i am 270 miles away, the sun weighs 500 lbs. alot more than i thought. the other box weighs 750 lbs. it's bigger than it looks. So if you would like to give the box a good home. The box was working a fews back when he last used, when powered on this time it wouldn't boot and didn't have time to investigate. the interior rack is 19 so it would be a cool place to mount your rack mount hardware even if you don't use the 690mp, last i heard it is supported by debian. So email me if your interested and i will give you the sellers email address. James Dickens Miwaukee, WI
Re: if you are in iowa, you can have a nice large machine
Sh-, the rack alone is worth $1. You could sell off the car battery sized SCSI drives and recoup double your money! And the tape library machine, well, what would be really cool is if you could convert it to serve drinks rather than tapes. Talk about free beer! The total capacity of the tape library is probably 10GB or something. a On Fri, Jun 06, 2003 at 01:08:53AM -0500, jamesd wrote: Hi i won this machine, http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItemitem=3027472460category=11216rd=1 it weights more than i can deal with, and i am 270 miles away, the sun weighs 500 lbs. alot more than i thought. the other box weighs 750 lbs. it's bigger than it looks. So if you would like to give the box a good home. The box was working a fews back when he last used, when powered on this time it wouldn't boot and didn't have time to investigate. the interior rack is 19 so it would be a cool place to mount your rack mount hardware even if you don't use the 690mp, last i heard it is supported by debian. So email me if your interested and i will give you the sellers email address. James Dickens Miwaukee, WI -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
New kernel images coming to unstable
Based off of 2.4.21-rc7+patches. The package naming scheme has changed: kernel-image-2.4.21-sparc32 kernel-image-2.4.21-sparc32-smp kernel-image-2.4.21-sparc64 kernel-image-2.4.21-sparc64-smp Not to mention there are also some udeb's for work I am doing on debian-installer for sparc. I have not tested the sparc32 kernels, so feel free to send feedback about them. All kernels are built with the now default gcc-3.2.3 64-bit enabled compiler (one compiler for all kernels and userspace...yay!). Don't expect to really see these for a day or two as they make their way to the mirrors. Just a heads up. -- Debian - http://www.debian.org/ Linux 1394 - http://www.linux1394.org/ Subversion - http://subversion.tigris.org/ Deqo - http://www.deqo.com/
Re: running X w/serial console (XDCMP)
On Thu, 2003-06-05 at 18:52, nate wrote: nate said: I was wanting to run gdm w/XDCMP on my ultra 1 which has a serial console, I can get the GDM login screen but cannot get past that, I guess it's impossible to run gdm w/o a local X server. Erm, my headless server runs gdm just fine. Not surprisingly X does not start (no keyboard), but I can connect over the network without any issues. However, you will need to edit /etc/gdm/gdm.conf and do teh following: 1. set [xdcmp]:Enable to true 2. comment out all the lines under [servers] AFAIK that is all I had to do to get GDM running. HTH, - Matthew
2.4.21-rc7 sparc32 kernel debs
Hopefully people don't mind my plugging my sparc32 kernel debs again. 2.4.21-rc7 debs have been posted; UP has been tested on an ss5, SMP on an SS10. I'm uploading them now to http://osinvestor.com/sparc/debs/ and they're almost all up there. Now back to your regularly scheduled mailings, Rob Radez
Re: video drivers
- Original Message - From: John Hedge [EMAIL PROTECTED] To: Debian Sparc Mailing List debian-sparc@lists.debian.org Sent: Thursday, June 05, 2003 8:36 AM Subject: video drivers Hi, I have a 7(10 possible)/1 bet at present. I'm hoping someone in the debian sparc community can narrow the odds :-) My choice of video driver for my Ultra 1 offered as follows: ati (I don't think so) fbdev (likewise) glint (no) sunbw2 (I doubt it) suncg14, 3, 6 (?) sunffb (again ?) sunleo (and again?) suntcx (???) My Ultra 1 has a suncg6. Good luck, Steven
