via epia MD hanging on netboot

[dothedog]

Hi All,
so I have had a 710Beta system running for about a year now. I had a hard drive failure so I thought I would install 810 Alpha from scratch. Well, it took a couple of tries and now I have a Hybrid/Core running. The problem I am having is that I have 4 via epia boxes that I have been using as md's for the last year. In 810alpha, they hang at the line that says Starting chrond. I knew from my last install that I had to install the openchrome drivers but I couldn't even get to a tty. Actually the tty's are there Ctl+Alt+F2 etc. but I can't log in. It allows me to put in a user name but it doesn't prompt for password. I have also tried ssh'ing into the box but it says "no route to host", so I am assuming that sshd hasn't started.

So I tried compiling the openchrome drivers in a chroot environment (/usr/pluto/diskless/XX), using the info in the wiki http://wiki.linuxmce.org/index.php/Display_Drivers. Still no love. The one thing that I couldn't figure out was how to get the x server to use the new driver as there doesn't seem to be a xorg.conf file. I am also not sure that it is just a graphics driver problem as I can't even log in to a tty.

Another thing I noticed is that there is a Fatal ...modules.dep not found.
I fixed that by doing a depmod -a in the chroot environment.

I took a look through the logs but I didn't see anything obvious. It just looks like it stops. I do have a copy of the logs, so if there is anything I can share, I am more than happy to post them. Any and All help is much appreciated.

DoTheDog

[dothedog]

So looking at the logs+
Xorg.0.log shows it is trying to load the vesa driver and failing:

Code: [Select]

(II) Loading /usr/lib/xorg/modules/drivers//vesa_drv.so
(II) Module vesa: vendor="X.Org Foundation"
compiled for 1.5.0, module version = 1.3.0
Module class: X.Org Video Driver
ABI class: X.Org Video Driver, version 4.1
(II) LoadModule: "fbdev"

(II) Loading /usr/lib/xorg/modules/drivers//fbdev_drv.so
(II) Module fbdev: vendor="X.Org Foundation"
compiled for 1.5.0, module version = 0.4.0
ABI class: X.Org Video Driver, version 4.1
(II) VESA: driver for VESA chipsets: vesa
(II) FBDEV: driver for framebuffer: fbdev
(II) Primary Device is: ISA
(WW) Falling back to old probe method for vesa
(--) Chipset vesa found
(II) resource ranges after xf86ClaimFixedResources() call:
[0] 0 0 0x000b8000 - 0x000bffff (0x8000) MX[B]
[1] 0 0 0x000b0000 - 0x000b7fff (0x8000) MX[B]
[2] 0 0 0x000a0000 - 0x000affff (0x10000) MX[B]
[3] 0 0 0x000003c0 - 0x000003df (0x20) IX[B]
[4] 0 0 0x000003b0 - 0x000003bb (0xc) IX[B]
(WW) Falling back to old probe method for fbdev
(II) Loading sub module "fbdevhw"
(II) LoadModule: "fbdevhw"

(II) Loading /usr/lib/xorg/modules/linux//libfbdevhw.so
(II) Module fbdevhw: vendor="X.Org Foundation"
compiled for 1.5.2, module version = 0.0.2
ABI class: X.Org Video Driver, version 4.1
(II) FBDEV(1): using default device

Fatal server error:
Cannot run in framebuffer mode. Please specify busIDs        for all framebuffer devices
So how do I make it use the openchrome driver that I compiled for it, if there is no xorg.conf file?

Nothing really in syslog:

Code: [Select]

Sep  8 22:37:00 moon41 anacron[5678]: Will run job `cron.monthly' in 15 min.
Sep  8 22:37:00 moon41 kernel: [    1.242997] PCI: 0000:00:10.3 reg 20 io port: [fb00, fb1f]
Sep  8 22:37:00 moon41 NetworkManager: <info>  Trying to start the system settings daemon...
Sep  8 22:37:00 moon41 anacron[5678]: Jobs will be executed sequentially
Sep  8 22:37:00 moon41 kernel: [    1.243047] pci 0000:00:10.3: supports D1
Sep  8 22:37:00 moon41 NetworkManager: <info>  (eth0): carrier now ON (device state 1)
Sep  8 22:37:00 moon41 kernel: [    1.243054] pci 0000:00:10.3: supports D2
Sep  8 22:37:00 moon41 nm-system-settings:    SCPlugin-Ifupdown: init!
Sep  8 22:37:00 moon41 kernel: [    1.243062] pci 0000:00:10.3: PME# supported from D0 D1 D2 D3hot D3cold
Sep  8 22:37:00 moon41 nm-system-settings:    SCPlugin-Ifupdown: update_system_hostname
Sep  8 22:37:01 moon41 kernel: [    1.243132] pci 0000:00:10.3: PME# disabled
Sep  8 22:37:01 moon41 nm-system-settings:    SCPluginIfupdown: guessed connection type (eth0) = 802-3-ethernet
Sep  8 22:37:01 moon41 kernel: [    1.243229] PCI: 0000:00:10.4 reg 10 32bit mmio: [fdffe000, fdffe0ff]
Sep  8 22:37:01 moon41 kernel: [    1.243306] pci 0000:00:10.4: supports D1
Sep  8 22:37:01 moon41 kernel: [    1.243312] pci 0000:00:10.4: supports D2
Sep  8 22:37:01 moon41 kernel: [    1.243321] pci 0000:00:10.4: PME# supported from D0 D1 D2 D3hot D3cold
Sep  8 22:37:01 moon41 nm-system-settings:    SCPlugin-Ifupdown: update_connection_setting_from_if_block: name:eth0, type:802-3-ethernet, autoconnect:0, id:Ifupdown (eth0), uuid: 681b428f-0000-0000-0000-000039000000
Sep  8 22:37:02 moon41 kernel: [    1.243390] pci 0000:00:10.4: PME# disabled
Sep  8 22:37:02 moon41 nm-system-settings:    SCPlugin-Ifupdown: addresses count: 1
Sep  8 22:37:02 moon41 kernel: [    1.243543] HPET not enabled in BIOS. You might try hpet=force boot option
Sep  8 22:37:02 moon41 nm-system-settings:    SCPlugin-Ifupdown: No dns-nameserver configured in /etc/network/interfaces
Sep  8 22:37:02 moon41 NetworkManager: <info>  (eth0): device state change: 1 -> 2

nor in dmesg:

Code: [Select]

[   17.349995] input: No brand KVM as /devices/pci0000:00/0000:00:10.1/usb2/2-1/2-1:1.1/input/input2
[   17.355208] input,hidraw1: USB HID v1.10 Mouse [No brand KVM] on usb-0000:00:10.1-1
[   17.355507] usbcore: registered new interface driver usbhid
[   17.355575] usbhid: v2.6:USB HID core driver
[   17.639363] RPC: Registered udp transport module.
[   17.639440] RPC: Registered tcp transport module.
[   17.740923] NET: Registered protocol family 17
[   17.743613] eth0: link up, 100Mbps, full-duplex, lpa 0xC1E1
[   17.751166] Marking TSC unstable due to TSC halts in idle
[   30.212339] udevd version 124 started
[   33.964400] pci_hotplug: PCI Hot Plug PCI Core version: 0.5
[   34.253223] Linux agpgart interface v0.103
[   34.316968] shpchp: Standard Hot Plug PCI Controller Driver version: 0.4
[   34.403106] agpgart: Detected VIA VT3314 chipset
[   34.416303] agpgart-via 0000:00:00.0: AGP aperture is 128M @ 0xe8000000
[   36.422417] input: Power Button (FF) as /devices/LNXSYSTM:00/LNXPWRBN:00/input/input3
[   36.436038] ACPI: Power Button (FF) [PWRF]
[   36.436421] input: Power Button (CM) as /devices/LNXSYSTM:00/device:00/PNP0C0C:00/input/input4
[   36.448092] ACPI: Power Button (CM) [PWRB]
[   36.784684] input: PC Speaker as /devices/platform/pcspkr/input/input5
[   40.227702] parport_pc 00:0a: reported by Plug and Play ACPI
[   40.227827] parport0: PC-style at 0x378, irq 7 [PCSPP,TRISTATE]
[   44.869624] ACPI: PCI Interrupt Link [ALKC] enabled at IRQ 22
[   44.869717] VIA 82xx Audio 0000:00:11.5: PCI INT C -> Link[ALKC] -> GSI 22 (level, low) -> IRQ 22
[   44.869972] VIA 82xx Audio 0000:00:11.5: setting latency timer to 64
[   89.642862] NET: Registered protocol family 10
[   89.644428] lo: Disabled Privacy Extensions
[   90.949783] warning: `ntpd' uses 32-bit capabilities (legacy support in use)

Help? Is there anywhere else I should be looking to try to figure this out?

DoTheDog

[dothedog]

Maybe found another hint. in auth.log I get:

Code: [Select]

Sep  8 22:36:34 moon41 sshd[4911]: Server listening on :: port 22.
Sep  8 22:36:34 moon41 sshd[4911]: Server listening on 0.0.0.0 port 22.

So it looks like it is not getting an ip address? Could it be related to this in syslog?

Code: [Select]

Sep  8 22:37:02 moon41 nm-system-settings:    SCPlugin-Ifupdown: No dns-nameserver configured in /etc/network/interfaces
Sep  8 22:37:02 moon41 NetworkManager: <info>  (eth0): device state change: 1 -> 2
DoTheDog

[dothedog]

OK, so I added a line
dns-nameserver 192.168.80.1
to /usr/pluto/diskless/41/etc/network/interfaces
and the boot got past "Starting....chrond"
but it hangs on (Sorry I'm doing this from memory as I am not right in front of it now) something to the effect
Lock pluto.conf stale
removing lock pluto.conf
retrying lock pluto.conf

If I cd /usr/pluto/diskless/41/usr/pluto/locks. it is an empty directory.

Although I do now have a new line in auth.log:

Code: [Select]

Sep  9 07:19:49 moon41 CRON[6129]: pam_unix(cron:session): session opened for user root by (uid=0)

However, I still can't log in to a tty and it is still hung.

Any Help is appreciated.

Thanks,
DoTheDog

[lemming86_au]

I'm having the same problem, I cant get any MDs to boot, only one is via based.
I posted here a while back (http://forum.linuxmce.org/index.php?topic=8425.msg56274#msg56274) & have worked with some people in chat, but haven't fixed it
So I would be interested to know if you find a solution

[totallymaxed]

I'm having the same problem, I cant get any MDs to boot, only one is via based.
I posted here a while back (http://forum.linuxmce.org/index.php?topic=8425.msg56274#msg56274) & have worked with some people in chat, but haven't fixed it
So I would be interested to know if you find a solution

As far as I know none of the various Via hardware has cleanly installed on 0810 or 904 Kubuntu let alone LinuxMCE. So I can be of much help apart from say we have given up putting any effort into Via based hardware that uses Via GPU's... I wish you luck getting them back to life but I would recommend not being too hopeful at all :-(

All the best

Andrew

[lemming86_au]

Well that explain one of my machines, only 3 to sort out.
Its a pity as I have quite a few Via machines lying around doing nothing.
Thanks Andrew

[dothedog]

I've done a bit of googling and found that most epia systems on 8.10 will boot but they are having difficulty with getting 3d and XvMC acceleration going. I really don't need that as I am mainly using it for music. Any ideas why this won't boot?

[dothedog]

One other thing I noticed is as it is booting is says something about a bad option for mount.nfs. Also, there is nothing in the logs section on the web admin page. Could it be losing its mounts? I am getting the exact same behavior on an old KT3 Ultra/geforce 5500/e1000 setup that worked in 710 as well. So I am wondering if it is an issue with the 810 alpha? I am more than happy to troubleshoot and if I get it working I'll write it up.  Any help?

[dothedog]

Not sure if this will solve it for via epia however, I did get atom 330 motherboards booting. I posted the solution here:
http://forum.linuxmce.org/index.php?topic=8425.msg61276#msg61276

Dothedog

[MarcoZan]

Hi Dothedog
a very trivial one: did you try to boot your Via Epia from a Kubuntu 8.10 live CD and see what happens?
Maybe you can get some more info in this way.

Just my 0,02
Regards
Marco

[dothedog]

MarcoZan,
I never did get the epia's working, so I swapped them out for atom 330 boards. When I got the same boot problem, and after a lot of googling, I finally figured out that NetworkManager was misbehaving with static ips in intrepid. The solution was to remove NetworkManager from the rc.d, the soution is posted in the link in my previous post. I believe, but haven't confirmed, (I pulled out my epia mobos and replaced them with the atom 330 boards) that the same solution would work on the epia's as well. But your suggestion is a good one that I never tried. Might have saved me a few bucks on new Mobos

DoTheDog

[MarcoZan]

Just out of curiosity ...

which kind of VIA Epia mobos did you have?

Was the mobos replacement straighforward, or did you have to change also case/power supply?

I also have a very old VIA Epia M10000 based MD, and I was wondering how much could cost to upgrade that MD to an Atom based one...

Thanks and regards
Marco

[dothedog]

MarcoZan,
I was using a few different epia boards, I think I had a C1000, 600, 800. In any event, I was using a Pico Power supply and m200 case from here: http://www.mini-box.com/site/index.html

I ended up with an atom 330 (dual core, i945 because I wasn't using it for video, rtl8111c) Intel BOXD945GCLF2D, with 2 GB of ram. The only thing I needed to change was a 4 pin mobo 12 volt adapter for the power supply. The same one worked just needed the 4 pin connector, I got that at mini-box.com as well.

Hope that helps.

Rob

[dothedog]

It looks like the killing network-manager allows via epia mds to work. I haven't gotten the openchrome driver working yet (I'm using vesa) but it boots, and I  can get the audio to work. For the solution go to the link in post above here: http://forum.linuxmce.org/index.php?topic=8425.msg61276#msg61276