Re: [Soekris] OpenBSD installation problem
Hi Aniss,
in your previous post you got this:
boot
booting hd0a:/bsd: 6563548+1052072=0x7435f8
entry point at 0x200120
Followed by {void}. Note that here boot.conf has been parsed and the
kernel image has been loaded. This would in most cases indicate a system
booting with the wrong output parameters.
But you're right, it can also point at a system that simply hangs.
Interesting question is why it then doesn't if you enter the boot
parameters manually. Perhaps better consult the OpenBSD misc list.
I hope I understand you well, posting the output of a bsd.rd boot as
well doesn't make things clearer.. (that's a totally different story).
Warning about the misc list in case you're not familiar with it: always
post at least a full dmesg, and show lots of respect for the project -
they bark a lot there but they bite as well:)
Good luck,
Bill
On Thu, 2009-12-31 at 13:52 +, aniss siyouti wrote:
hi,
yes i can log via SSH but just when i hit in the command prompt
boot the instruction of the boot.conf file, if i do nothing the
system hangs .
comBIOS ver. 1.32i 20071005 Copyright (C) 2000-2007 Soekris Engineering.
net5501
0512 Mbyte MemoryCPU Geode LX 500 Mhz
Pri Sla SMC512AFB6E LBA 1014-16-63 511 Mbyte
Slot Vend Dev ClassRev Cmd Stat CL LT HT Base1Base2 Int
---
0:01:2 1022 2082 1010 0006 0220 08 00 00 A000 10
0:06:0 1106 3053 0296 0117 0210 08 40 00 E101 A0004000 11
0:07:0 1106 3053 0296 0117 0210 08 40 00 E201 A0004100 05
0:08:0 1106 3053 0296 0117 0210 08 40 00 E301 A0004200 09
0:09:0 1106 3053 0296 0117 0210 08 40 00 E401 A0004300 12
0:20:0 1022 2090 06010003 0009 02A0 08 40 80 6001 6101
0:20:2 1022 209A 01018001 0005 02A0 08 00 00
0:21:0 1022 2094 0C031002 0006 0230 08 00 80 A0005000 15
0:21:1 1022 2095 0C032002 0006 0230 08 00 00 A0006000 15
1 Seconds to automatic boot. Press Ctrl-P for entering Monitor.
Using drive 0, partition 3.
Loading...
probing: pc0 com0 com1 pci mem[639K 511M a20=on]
disk: hd0+
OpenBSD/i386 BOOT 3.02
switching console to com0
OpenBSD/i386 BOOT 3.02
boot stty com0 19200
com0: 19200 baud
boot set tty com0
switching console to com0
OpenBSD/i386 BOOT 3.02
boot boot bsd.rd
booting hd0a:bsd.rd: 5651156+913072 [52+211008+196339]=0x6a6260
entry point at 0x200120
Copyright (c) 1982, 1986, 1989, 1991, 1993
The Regents of the University of California. All rights reserved.
Copyright (c) 1995-2009 OpenBSD. All rights reserved. http://www.OpenBSD.org
OpenBSD 4.6 (RAMDISK_CD) #53: Thu Jul 9 21:41:35 MDT 2009
dera...@i386.openbsd.org:/usr/src/sys/arch/i386/compile/RAMDISK_CD
cpu0: Geode(TM) Integrated Processor by AMD PCS (AuthenticAMD 586-class)
500 M
.
bios0 at mainbus0: AT/286+ BIOS, date 20/71/05, BIOS32 rev. 0 @ 0xfac40
.
com0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo
com0: console
com1 at isa0 port 0x2f8/8 irq 3: ns16550a, 16 byte fifo
...
erase ^?, werase ^W, kill ^U, intr ^C, status ^T
Welcome to the OpenBSD/i386 4.6 installation program.
(I)nstall, (U)pgrade or (S)hell? S
# ls
.profile etc install.sub sbin usr
bin install mnt tmp var
dev install.md mnt2 upgrade
#
2009/12/29, Bill Maas b...@stsx.org:
Hi Aniss,
I don't think the order of the stty and set command matters a lot.
Just don't place the boot command in top, or they never will be
executed. Did you activate serial output in /etc/ttys? That's the only
other trivial thing I can come up with. A non-trivial thing would be the
boot loader not being able to correctly parse /etc/boot.conf for some
reason and load the kernel using default settings, but so far I've never
seen that happen, and you probably would get an error message in that
case. It MUST be the kernel serial port setup somewhere, since
everything else goes according to plan. Can you log in over SSH?
Bill
On Tue, 2009-12-29 at 14:26 +0100, aniss siyouti wrote:
Hi Bill ,
I checked the order of the boot commands in boot.conf and i noticed
that is different from the order in the document
my boot.conf :
set tty com0
stty com0 19200
boot tftp:/bsd.rd
the correct order :
stty com0 19200
set tty com0
boot tftp:/bsd.rd
i tried to reprogram my soekris box using pxebooting but my system
hang again after the same message,I’m apparently unlucky;
aniss
2009/12/28 Bill Maas b...@stsx.org
Hi Aniss,
I sent the previous message prematurely, so let's finish it.
The text
comes from the doc I pointed you to. Your Soekris box is
probably
booting just fine, you just don't get to witness it;).
Bill
Re: [Soekris] OpenBSD installation problem
Hi Aniss, Yes. See http://www.stsx.org/openbsd/obsd-bootsoekris.html I wrote it, in what seems previous life lately. Don't know whether it's entirely up2date, but I expect so - with OpenBSD things don't change a lot at this level. Bill On Mon, 2009-12-14 at 22:35 +, aniss siyouti wrote: One other question,is it possible to install the file sets of openbsd without connecting the soekris to internet.Can i transfer these file from my computer to the soekris using tftp. aniss ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Cloning a bootable CF
Hi Marco, On Tue, 2009-11-24 at 10:17 -0600, Darrick Hartman wrote: Marco A. Calamari wrote: On Tue, 2009-11-24 at 06:40 -0600, Darrick Hartman wrote: Marco A. Calamari wrote: On Tue, 2009-11-24 at 10:51 +0100, Flemming Jacobsen wrote: Hey Why don't you just clone it using dd: Insert old card dd if=/dev/DEVICE of=/tmp/soekris.image bs=64k Insert new card dd of=/dev/DEVICE if=/tmp/soekris.image bs=64k The original CF is 512Mb and the new is 4Gb. I don't know if that works, but for sure would waste 3.5Gb Many thanks. Marco What's the file system? If it's ext2, you can resize the partition after you copy the data using dd as described above. What you are sayng is not clear to me. If you dd a partitioned, bootable device, let's say /dev/sda, you copy also the boot block and the partition table to the image dd'ing the image on the new CF you have the original partition table, so the dimension of the partition is the same and the volume dimension are wrong. Am I missing something? Many thanks to all for your time and help. Marco Correct, but after it's on the new larger device, you can expand the size of the partition IF the file system supports that operation. ext2 does support growing the file system. Something like gparted will work nicely for you if you have a Linux desktop. Otherwise, grab a bootable CD that has gparted as one of the available utilities. It seems to me that in that case you might as well do it properly and mke2fs, mount, rsync the content and install the boot loader, all of which can easily be done from a shell script if you need to automate the process. A fresh installation of the OS on the Soekris box will give you the cleanest and most reliable installation if the CF is being moved from one type of box to another, if that's not too much trouble. It may take up less time than finding and fixing compatibility bugs afterwards. Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Loading standard CentOS
Hi Yakout, On Thu, 2009-11-05 at 09:00 +1300, Yakout Esmat wrote: Hi All, We build and distribute Asterisk-based PBXes, I guess like many others. What we are looking for is to standardize our hardware platform, which makes backup/restore, replacement etcmuch easier than using normal PC/Server hardware. My questions: 1- Can we install standard CentOS 5.x on the Soekris 5501? No, according to http://lists.soekris.com/pipermail/soekris-tech/2007-May/012031.html But that message may obviously be outdated. 2- Can we boot from an external USB DVD just like we can with a standard PC? No, you have the choice between CF, HDD and network. CF is quite convenient, actually. 3- Is there any intention to build boards with faster CPU, like 1Ghz for example? Only Soren can answer that question. Bill Thanks for your help. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Tcl hate, was: Writing a BIOS config from cmd line
Hi Andy, On Wed, 2009-10-21 at 19:25 -0500, Andy Michaels wrote: Off-topic, but why all the hatred for Tcl? Ignorance breeds contempt;) Bill -a Tcl hacker *sniff* On Wed, Oct 21, 2009 at 4:18 PM, Bill Maas b...@stsx.org wrote: Hi Graham, how about using different machines for different purposes? As an alternative to burning a few boards for the sake of the experiment or having pairs of boxes configuring each other. Not really sure what you're planning to accomplish.. By the way, 'expect' is The Horror as some of the others pointed out, it used to be needed for talking to modems from Linux, thank heavens those days are gone.. And I'm not at all convinced about how reliable repetitive unattended flash burning will prove in the long term. Bill On Wed, 2009-10-21 at 07:12 +1100, Graham Menhennitt wrote: Erik M. Cummings wrote: That's what I have put together right now and it works for my test bench. Works okay, but doesn't scale at all. I would need a whole other PC in the field with each of these in order to do that. Erik On Oct 19, 2009, at 6:49 PM, Graham Menhennitt wrote: Erik M. Cummings wrote: I need to be able to save and write BIOS configs from the command line. Could you use some kind of expect (http://sourceforge.net/projects/expect/) script over the console serial port to do this. Something like: send \r sleep 1 send +++ sleep 1 send \r expect Soekris send cmoswrite whatever Ok, I misunderstood what you meant by command line. Perhaps you could explain it better. For starters, what operating system? I imagine that the CMOS is mapped into the o/s memory somewhere and so would be readable and possibly writable. Graham ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Writing a BIOS config from cmd line
Hi Graham, how about using different machines for different purposes? As an alternative to burning a few boards for the sake of the experiment or having pairs of boxes configuring each other. Not really sure what you're planning to accomplish.. By the way, 'expect' is The Horror as some of the others pointed out, it used to be needed for talking to modems from Linux, thank heavens those days are gone.. And I'm not at all convinced about how reliable repetitive unattended flash burning will prove in the long term. Bill On Wed, 2009-10-21 at 07:12 +1100, Graham Menhennitt wrote: Erik M. Cummings wrote: That's what I have put together right now and it works for my test bench. Works okay, but doesn't scale at all. I would need a whole other PC in the field with each of these in order to do that. Erik On Oct 19, 2009, at 6:49 PM, Graham Menhennitt wrote: Erik M. Cummings wrote: I need to be able to save and write BIOS configs from the command line. Could you use some kind of expect (http://sourceforge.net/projects/expect/) script over the console serial port to do this. Something like: send \r sleep 1 send +++ sleep 1 send \r expect Soekris send cmoswrite whatever Ok, I misunderstood what you meant by command line. Perhaps you could explain it better. For starters, what operating system? I imagine that the CMOS is mapped into the o/s memory somewhere and so would be readable and possibly writable. Graham ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Question with net5501
Hi Manh Do, on OpenBSD (a NetBSD descendant, as you probably know), the message device not configured means as much as no matching kernel driver present. Since vr* should be supported, but apparently doesn't work, this is probably one for the NetBSD list. Bill On Sat, 2009-09-05 at 10:57 -0700, Manh Do wrote: Hi, I'm trying to rebuild the NetBSD kernel so I enable the 'vr* at pci? dev ? function ?' under the 'Network Interfaces' to accomodate the VIA Rhine III VT6105M Ethernet Controller. However at bootup time, I got the following console messages: Vendor 0x1106 product 0x3053 (ethernet network, revision 0x96) at pci0 dev 6 function 0 not configured Vendor 0x1106 product 0x3053 (ethernet network, revision 0x96) at pci0 dev 7 function 0 not configured Vendor 0x1106 product 0x3053 (ethernet network, revision 0x96) at pci0 dev 8 function 0 not configured Vendor 0x1106 product 0x3053 (ethernet network, revision 0x96) at pci0 dev 9 function 0 not configured Apparently, these are indeed the 4 ethernet ports on the net5501 board. How I can fix these errors so the ethernet ports are properly recognized and configured correctly? Thanks, Manh Do ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Ready to burn Flash Card Images?
Hi Mads, take a look at http://www.stsx.org/debian/ibs/index.html It will take care of the haggis of potential PXE/DHCP/TFTP errors for you, so you can focus on the OS you want to install. Debian lenny installs perfectly, for Soekris-specific details about Ubuntu see the Soekris wiki. Bill On Mon, 2009-08-03 at 23:17 +0200, Mads Hjorth wrote: Hi, I am trying to get my Soekris 5501 back up and running. I had it running nicely with a OpenBSD but ran into problems when I found that FUSE (http://fuse.sourceforge.net/) did not run under OpenBSD. I decided to give latest Ubuntu Server 9.4 a try. I did not manage to get it installed via PXE and my first try into debootstrap did not get my all the way. From the plenty of pages with people giving guide lines on how to install various OS on Soekris it seems like there most be plenty of Compact Flash Cards out there with running installations. My question... Is there a repository of ready to go installation of some basic OS for Soekris? more specific Ubuntu 9.4 on a 5501? Regards, Mads ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Installing Fedora 11 on Soekris 5501-70
Hi Casey, take a look at this: http://www.myfaq.com.cn/2005September/2005-09-13/200325.html Maybe RedHat/Fedora isn't such a great choice for a Soekris after all.. Since you are not that familiar with computer architecture or Linux I'd recommend you to go with the mainstream and use Debian or Ubuntu instead - you'll get tons more support. Still, getting any Linux to run on a Soekris box requires some familiarity with what's under the hood, at least if you like to know what you're doing. I think the Installing Debian Linux (4.0r1 Etch) from Windows mini-HOWTO on the Soekris Wiki is best suited for your purpose, and it's probably safe to substitute Etch with Lenny. Once you have gotten familiar with that, you can always try installing a less commonly supported distro. Main issues: boot loader, disk geometry, kernel modules, and of course the Soekris comBIOS. Good luck, Bill On Mon, 2009-07-20 at 12:21 -0400, Catherine Charlotte Keys wrote: Hi, I'm trying to run any Red Hat distribution on a Soekris net5501-70 from Compact Flash. Using liveusb-creator from Fedora I've been able to get Fedora running, but there is no persistent storage. If I install to the Compact Flash directly from a live CD the net5501-70 can see the OS, but it cannot read the image. If I use a kick start file the OS will install, but it will not run. The start up process just hangs after the auto boot. I've tried to follow the directions for installing Red Hat from the Soekris support page, but I'm just not that familiar with computer architecture or Linux. Does anyone know how to install a Red Hat distribution of Linux on the net5501-70? Best, Casey ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] voyage linux installation
Hi John, On Tue, 2009-07-21 at 12:21 -0500, John G. Heim wrote: I would like to try installing voyage linux on my 4801. But I don't understand the installation instructions. -- begin quote -- Download the Voyage Linux software package from http://www.voyage.hk/download/voyage/ to a Linux machine. extract the software tarball: ^^ tar --numeric-owner -jxf voyage-version.tar.bz2 as root, run the installation script: cd voyage-version ./usr/local/sbin/voyage.update -- end quote -- It seems there must be a step missing. How can I run the installation script unless I've already got linux on my machine? I don't run Voyage, but from the installation instructions I understand that you are expected to build the CF image on a separate machine (PC, laptop) running Linux, then flash it onto CF and then transfer the CF card to the Soekris box. The latter step being a highly manual one;) Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Readdressing the cooling question....
Hi, On Fri, 2009-07-17 at 06:17 +, Poul-Henning Kamp wrote: In message 4a5ff2bc.9010...@foofus.com, AMuse writes: Naturally with that kind of abuse I'm wondering if I should put extra cooling onto the soekris box -- and what's the best way to do so? One of the first things to do, is to choose a box which has a surface which is bright metal, to minimize heating from the sunlight. It doesn't matter much which metal, as long as the reflectivity is very high. I in my uneducated state of mind (as far as hardware matters are concerned) would in that case consider mounting a loose reflective surface onto, or around, the case, if necessary thermally insulated from the case itself and with enough space in between to provide some airflow. I guess that's what they call a heat shield:). At least it saves having to re-paint the case (paint doesn't necessarily hold well in sunlight either). Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris Booting Problem
Hi Kenneth, On Thu, 2009-07-16 at 10:07 -0400, Hendrickson, Kenneth wrote: --- Wim Vandeputte wrote: show the screen capture of your ComBIOS booting --- I said: I will try to publish the list in about 10-12 hours from now. Is there a way to get cu to capture all IO to a file?? I don't know how to do that. I don't see it in the man page. Even though the guru's won't like it, imho this is the simplest and neatest way to do it: r...@host:~# cu -l /dev/tty00 -s 19200 | tee soekris.out See http://www.stsx.org/openbsd/obsd-bootsoekris.html for a (tiny) bit more text concerning this issue. Bill Ken Hendrickson ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris Booting Problem
Hi Kenneth, On Thu, 2009-07-16 at 09:48 -0400, Hendrickson, Kenneth wrote: joel jaeggli [mailto:joe...@bogus.com] wrote: I have a couple systems that required updates to 1.33c in order to support the mass-storage devices we choose to use. I upgraded my bios to 1.33c, and it still doesn't work. When the bios starts, I *do* get the message that the drive exists, and I get the LBA c/h/s numbers printed. However, the motherboard won't boot from the drive. The bios claims there are no bootable devices. But, once again, after I get OpenBSD booted (via PXEboot), the disk works just fine. This indicates that the disk controller, the disk, and the cables are all good. So this looks to me like a bug in the bios. Or it might be a bug in the OpenBSD install code. Or there might be a problem in the way I laid out my disk slices which is incompatible with the comBios. Some of this limitation is due to nature of low volume embedded systems, by their nature they don't recieve as much testing with different components as high-volume systems do. people who productize around them tend to use the same components for a really long time. It would be nice to make two lists on the Soekris Wiki: (1) known good devices, and (2) known non-working devices. Maybe make that the Soekris Web Site. The Wiki is independent and unofficial, as it should be. I.e. the Wiki can't and doesn't have to take any responsibility for issues like these. Bill Søren: can you please help by publishing relevant snippets of the bios code for booting from ide/sata drives? Thanks, Ken Hendrickson ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501, Odd errors on boot
Hi Max, On Thu, 2009-07-09 at 22:02 -0800, Max Lou wrote: Hello, I have a net5501 I just finished building a fresh 2.6.29 kernel and configured serial output via lilo.conf: serial=0,19200n8 append=console=tty0 console=ttyS0,19200 Two problems: One, once the kernel loads the characters seem to be printing twice (the soekris combios displays fine, it is only when the kernel loads that the problem exists). See http://lists.soekris.com/pipermail/soekris-tech/2002-February/36.html [...] Then I get a bunch of messages which do not occur when booting off the CF card on a regular desktop system (see below) Does anyone know what this means? I tried googling, but there is so much displayed here I dont even know where to start. To me this looks like the regular kernel output, possibly a bit more verbose than what you would see on your desktop system (which may hide the kernel output behind a splash screen altogether). Bill [0.00] Linux version 2.6.29 (root@(none)) (gcc version 4.3.2 (CRUX) (GCC POST: 012345689b ^^^ A copy/paste thing I assume? ) ) #1 SMP Thu Jul 9 21:35:50 UTC 2009 [0.00] KERNEL supported cpus:yright (C) 2000-2008 Soekris Engineer [0.00] Intel GenuineIntel [0.00] AMD AuthenticAMDt 995-128-63 4013 Mbyte CP [0.00] NSC Geode by NSC [0.00] Cyrix CyrixInsteadRev Cmd Stat CL LT HT Base1Ba [0.00] Centaur CentaurHauls [0.00] Transmeta GenuineTMx86 [0.00] Transmeta TransmetaCPU 0:01:2 1022 2082 10100 [0.00] UMC UMC UMC UMC000 10 [0.00] PAT not supported by CPU. 0:06:0 1106 3053 0296 0117 0210 0 [0.00] BIOS-provided 0:20:2 1022 209A 01018001 0005 02A0 08 00 00 000 [0.00] BIOS-e820: fff0 - 0001 (reserved)2 2094 0C031002 0006 0230 08 00 80 A0005000 15 [0.00] DMI not present or invalid. 0:21:1 1022 2095 0C032002 0006 0230 0 [0.00] last_pfn = 0x2 max_arch_pfn = 0x10 1 Seconds to automa ^^^ Same... [...] ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Console port doesn't show anything
Hi Matt, On Mon, 2009-06-08 at 03:37 -0500, Matt Lawrence wrote: On Sun, 7 Jun 2009, Bill Maas wrote: (b) set up an NFS boot server, boot the box off the NFS root, and again try SSH. In this case reconfiguration will only require a reboot of the Soekris box. The drawback is that you may get to see numerous nondescript error messages from NFS before you get it all right, each of which may take a day or more to fix (really!). How about configuring to PXEboot, which uses DHCP tftp to grab a boot image, to boot something like RIPlinux. Adding in trying to use NFS for a root filesystem really isn't necessary and a lot more work. Just run straight out of the initrd. Hadn't even thought of that because I never do it that way, but definitely worth trying. Still, nothing beats an NFS boot when it comes to on-the-spot configuration/updating. It's also quite useful for quickly grabbing a dmesg and other config info off a box. And once set up properly (...), it tends to work fine, most of the time;) Bill -- Matt It's not what I know that counts. It's what I can remember in time to use. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Console port doesn't show anything
(Pressed Reply i.o. Reply to All - it happens.. Here's a copy for the list) Hi Cesar, On Fri, 2009-06-05 at 14:44 -0500, Cesar Bonilla wrote: Thanks a lot Bill, The cable worked right If that implies but I still can't see any output, it looks like the UART is broken. The good news is that the LED activity on the box _suggests_ that otherwise the board still _may_ be in good health. You can do 2 things: (a) configure a CF card on a PC, perform a blind boot and hope and pray that you may be able to ssh into the system. The main drawback is that for each configuration error you will have to re-flash the CF. And unless you know the CHS values, you will need to use a boot loader that ignores the fs, e.g. lilo. (b) set up an NFS boot server, boot the box off the NFS root, and again try SSH. In this case reconfiguration will only require a reboot of the Soekris box. The drawback is that you may get to see numerous nondescript error messages from NFS before you get it all right, each of which may take a day or more to fix (really!). I recently finished a piece of software (shell code) recently which sets up a Debian boot server with NFS support inside a chrooted environment: http://www.stsx.org/debian/ibs/index.html It's finished for now, publishable, and _almost_ marketable: the only thing left to do is finish the high-level Operation Manual, which is intended to demonstrate how easy it is to set up and operate the kit. But the manpages should be quite readable as well, and if time is an issue, in most cases you can skip/skim the blah and look at the examples, which mostly speak for themselves. I'll see if I can get myself to giving the OM a push in the right direction the coming days. I hope it may be of use to you and others. Anyway good luck with the boxen, Bill Regards Cesar, Bill Maas escribió: Hi Cesar, make sure you're using a correct serial null modem cable: http://www.camiresearch.com/Data_Com_Basics/RS232_standard.html If both boxes are properly connected, you should at least get garbage from the comBIOS output, even if the serial parameters are wrong. Check for compliance with the DB9 Null Modem Cable pin config, even if the cable was sold as such (there are crappy cables sold out there). Bill On Thu, 2009-06-04 at 09:58 -0500, Cesar Bonilla wrote: Hi Everybody, I have a Net5501 with a USB Memory. I connect to Console port but I can't see anything! Power led and Error led turn on. Then Disk led turns on for a while, after then Error led and Disk leds turn off. My settings of Serial port are: Baud speed: 19200 Data bits: 8 Stop bits: 1 Parity: none Flow control: none Thanks, -- Cordial Saludo, Ing. César Augusto Bonilla A. Ingeniero de Nuevos Productos HIT COMUNICACIONES Bogotá, Colombia Tel: +57 1 7424000 ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech __ No virus found in this incoming message. Checked by AVG - www.avg.com Version: 8.5.339 / Virus Database: 270.12.53/2156 - Release Date: 06/05/09 06:24:00 -- Cordial Saludo, Ing. César Augusto Bonilla A. Ingeniero de Nuevos Productos HIT COMUNICACIONES Bogotá, Colombia Tel: +57 1 7424000 __ Este mensaje contiene información confidencial para uso exclusivo del destinatario. Si Ud. es una persona distinta al destinatario, no debe distribuir, leer o copiar este correo electrónico. Sírvase notificar inmediatamente a la persona que envió este mensaje si Ud. ha recibido el mismo por error bórrelo inmediatamente de su sistema. Muchas Gracias. This message contains confidential information and it is intended only for the individual named. If you are not the named addressee you should not distribute, read or copy this e-mail. Please notify the sender immediately by e-mail if you have received this message by mistake delete this e-mail from your system. Thank you. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Net5501 and Crux
Hi, On Thu, 2009-04-30 at 21:58 -0800, TechKid PC wrote: I finally have the necessary cables and am able to access the comBIOS via a USB-Serial-Null-Mode-Female-Gender-Changer adapter chain. [...] Some experimenting raises second thoughts on the messed up comBIOS issue. [Bill wrote back:] To answer your question properly: no, there doesn't seem to be a way to reset the comBIOS to factory defaults on a Soekris box. It wouldn't help much in the case of a wrong baud rate setting either, because the baud rate would still be a valid one (or else it would be rejected on the first place, I assume) Confirmed for baud rates like 300 and 1200. Not for 115200 however. In fact the net5501 supports that speed: set ConSpeed=115200 show ConSpeed = 115200 Result: r...@hibernia:~# cu -l /dev/ttyUSB0 -s 115200 Connected. POST: 012345689bcefghips1234ajklnopqr,,,tvwxy Providing support for an undocumented serial transfer rate (or not documenting a supported rate - I looked into both the manual and the changelog, but I couldn't find it mentioned anywhere) is surely an effective way to confuse users. and the problem could therefore be delegated to the terminal configuration. I don't see how one could mess up the comBIOS settings in such a way that it would no longer be accessible. Except if the box can go into a supported invalid baud rate. Hard to imagine how this could happen involuntarily, possibly through thoughtless automatism. Maybe a fine excuse to finally update/create the net5501 doc. By the way, assuming that the UART supports 8 speeds (probably), if 6 speeds are this one is also supported, which is the 8th one? I hope I'm not missing something fundamental here.. Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Net5501 and Crux
Hi Tom, On Thu, 2009-04-30 at 21:58 -0800, TechKid PC wrote: I finally have the necessary cables and am able to access the comBIOS via a USB-Serial-Null-Mode-Female-Gender-Changer adapter chain. With that you should have dealt with about 80% of standard possible trouble. Is there anything special that I need to do in the bios, or to the OS installed on the cf card for it to boot? I setup Crux on a MicroSD card (via a MicroSD-SD-CF-IDE adapter chain) and it runs fine and boots on a dell optiplex desktop. When I insert the card in the soekris CF slot and turn it on, the comBIOS does not recognize any bootable device. I tried setting the partition value to 1 via the set command and it made no difference. Any thoughts? Could it be a compatibility issue with the microsd card? (Its a 2gb kingston). I have not tried a CF card yet but will run out and buy one if I need to. Although I want to get it running on the microsd card if at all possible. If the Soekris site doesn't state that MicroSD cards are supported, the odds are against you.. (this is a case of non-standard possible trouble) Sandisk CFs are practically guaranteed to work under all circumstances. Most other should do if no 2.5 HDD is attached, just stay away from Kingston it seems. I use DaneElec XS cards, great value for modest amount of money. Having a CF available for emergencies is always worth the trouble. On a second note, I dont see any way to hard reset the bios settings that are changed via the set command. What recovery options are available if one sets an incorrect option in the combios and disables serial output (or something similar) making it impossible to manage the device? There is no removable battery and no reset jumper I can find. Section 2.5 of the Soekris net4801 manual, preliminary version 0.01;). Bill Thanks, Tom - Original Message - From: Bill Maas b...@stsx.org Sent: Monday, April 13, 2009 1:24 AM Subject: Re: [Soekris] Net5501 and Crux Access to the Soekris comBIOS is not just useful for watching boot messages. It's an essential tool in case of trouble. See the Soekris documentation and the Soekris Wiki: http://wiki.soekris.info/Main_Page ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Net5501 and Crux
On Fri, 2009-05-01 at 03:28 -0800, TechKid PC wrote: From: Bill Maas b...@stsx.org Sent: Friday, May 01, 2009 1:22 AM On Thu, 2009-04-30 at 21:58 -0800, TechKid PC wrote: I finally have the necessary cables and am able to access the comBIOS via a USB-Serial-Null-Mode-Female-Gender-Changer adapter chain. With that you should have dealt with about 80% of standard possible trouble. I know its a pretty convoluted setup. :-P But it is the *only* thing that worked as my laptop does not include a serial port. I am able to use the comBIOS without any problems, so it seems to be working fine. If the Soekris site doesn't state that MicroSD cards are supported, the odds are against you.. (this is a case of non-standard possible trouble) That could very well be the case. The comBIOS however does see the card and list it when I first turn it on with the proper size. Out of curiosity, do you know why some CF brands work better then others when it comes to compabilitiy? I could see if a manufacturer did not include support for a certain call or function, but if the card is booting just fine off of a desktop via a CF-IDE adapter how could it not work with a soekris box? (Asuming the disk geometry is not the issue, which mention at the bottom of this message) Initialization and timing issues which are well beyond my knowledge, but I've been following this list for a while;). Search for soekris boot problem and you'll find plenty of background info. In general, the assumption that if stuff works fine in one context it should also work in another (apparently) similar context doesn't seem to hold well if computer hardware (or software, for that matter) is involved. To answer your question properly: no, there doesn't seem to be a way to reset the comBIOS to factory defaults on a Soekris box. It wouldn't help much in the case of a wrong baud rate setting either, because the baud rate would still be a valid one (or else it would be rejected in the first place, I assume) and the problem could therefore be delegated to the terminal configuration. I don't see how one could mess up the comBIOS settings in such a way that it would no longer be accessible. Unless that ConLock option does something funny with the UART. Bill On a second note, I dont see any way to hard reset the bios settings that are changed via the set command. What recovery options are available if one sets an incorrect option in the combios and disables serial output (or something similar) making it impossible to manage the device? There is no removable battery and no reset jumper I can find. Section 2.5 of the Soekris net4801 manual, preliminary version 0.01;). I read that section in the manual, as well as the in the wiki under: http://wiki.soekris.info/What_do_all_those_BIOS_settings_do They just list what the options are for, not how to reset any of them if you are unable to see any output. From: 686f6c6d 686f6...@googlemail.com Sent: Friday, May 01, 2009 12:56 AM I am no expert, but other than setting a unusual baudrate (which you would have to guess then to make it work again) I see no way to achieve such a mess-up. at least, I have not yet managed to do that. Ah ok that makes sense and could be an answer to my question about not seeing a reset option for the bios. If there are no settings that could be changed which could lock out the user then a reset feature is not needed. Ive seen many people overclock a cpu/ram, set a password and forget it, etc. in their destktop PC's bios and the only way to fix it is the reset jumper. I was worried something similar could happen with the Soekris box. I assume that the conMute setting simply prevents output from being sent to the serial port, but still accepts commands? (Such as set ConMute=Disabled) Im' not sure from your explanation if you did this installation on a PC or your Soekris. If it was on a PC, did you give fdisk (maybe before, maybe during install) the correct disk geometry values as described in http://wiki.soekris.info/What_is_this_disk_geometry_thing That could very well be the case. I didnt even think of checking the geometry values. Ill give it a try and if it does not work I will pickup a sandisk CF card. Thanks for the replies, Tom ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4801 start up issue
Hi Martin, On Mon, 2009-04-27 at 20:36 +0200, Martin Arendtsen wrote: I'm having some fun with a net4801. After a failed attempt to install freebsd on the net4801 (attached is a fujitsu harddrive) I wanted to give it a go once again. But now when the net4801 boots it simply don't show any boot. Instead it only show the following line repeatetly. ^BOOT:23064381,0,0,0,21 It repeats the message with 10 sec interval. Try a more powerful power supply. A Soekris box momentarily draws a relative large amount of power after it has been powered up for a few seconds, approx. 9W. The harddrive just might be a little too much for the psu to handle at that point. If that's the case, you get these repeated retries. By the way, I think that Soekris should publish boot error codes, if there are any (it does look like there are a few here). Bill I simply can't get it to respond to anything. Pressing a key doesn't give anything. Resetting it doesn't help either. Now I'm lost and need some help. Any ideas /Martin ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Problems booting from transcend 2GB CF in net4801 after bios upgrade
On Fri, 2009-04-24 at 03:32 -0500, James R. Van Artsdalen wrote: Bill Maas wrote: geometry issues usually don't result in No boot device errors, they just cause the boot loader to fail. After all, sector 0 is sector 0, no matter what geometry is used. AFAIK. Except, of course, for CHS, where sector 0 is sector 1... Mmmm... of course:) Bill I agree that No Boot device is unlikely to be a geometry issue. If the BIOS POST output is inspected it will show if the device is found at all. If the Soekris BIOS LBA functions works it shouldn't be that hard to modify the nanobsd build script to be able to use GPT partitioning instead of MBR, which would permanently solve geometry problems. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Problems booting from transcend 2GB CF in net4801 after bios upgrade
Hi Rick, geometry issues usually don't result in No boot device errors, they just cause the boot loader to fail. After all, sector 0 is sector 0, no matter what geometry is used. AFAIK. Bill On Thu, 2009-04-23 at 15:18 +0200, Rick van der Zwet wrote: After a bios upgrade from (1.27a - 1.33) my Trancend CompactFlash 2GB 120x is no longer bootable, stating the error: No Boot device available, enter monitor. Downgrading to 1.31 or 1.30 did not fix it either. Trying back to 1.28 or 1.27a resulted in the error: Error, Old revision of BIOS image. My best guess is a geometry as I did not build the NanoBSD image with the exact right geometry specifications, but could not find much in the changelog with regards why did suddenly could cause failure. Any advice on how to revert to 1.27a or to get it working at 1.33? /Rick ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] comBIOS conlock problem
Hi Hansen, try this: # apt-get install cu # chown uucp /dev/ttyS0 # cu -l /dev/ttyS0 -s 19200 You can probably imagine that the values for the local serial port (/dev/ttyS0) and the transfer rate (19200) are just examples. You will also probably have to run the chown command after each reboot on Ubuntu. The penalty for not doing so will be: cu: open (/dev/ttyS0): Permission denied cu: /dev/ttyS0: Line in use The advantage of using cu(1) is that you can rule out various quirks and omissions in the configuration of high-level programs like minicom themselves. It's only when you need to use stty(1) for changing terminal settings that you might want to go back to one of those UI programs, but the standard terminal settings on Ubuntu should be quite reasonable in most cases. Please send the terminal output if this still doesn't work. You can capture cu's output by simple shell redirection: # cu -l /dev/ttyS0 -s 19200 | tee soekris.out There are more fancy ways to do this, but they're not much more practical. Try removing the control characters inserted by the shell (^M etc) before you paste the input file's contents into your message. Look here for directions on how to do this in vi[m]: http://todayslearnings.blogspot.com/2009/02/remove-control-m-chars-in-unix-by-using.html Please don't be offended if that's something you've done a thousand times before, it's just that you're not giving a lot of details, so I can't tell from here. Another thing you could try is to use a strict 3-wire null-modem cable without any non-standard connections between the pins on the same connector. I had trouble with an off-the-shelf null-modem cable for STB's which turned out to be wired up in mysterious ways, and caused the RTS/CTS hardware flow control lines to lock up the connection. Bill On Sun, 2009-04-19 at 21:15 -0600, RB wrote: On Sun, Apr 19, 2009 at 19:26, Hansen Cahyono hcahy...@gmail.com wrote: Hi Bill, I tried using HyperTerm from WinXP also tried minicom and gtkterm from Linux (Ubuntu) Still can't communicate with the soekris box appropriately. FWIW, I've had diminishing fortune on Linux using anything but picocom. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] comBIOS conlock problem
Hi Hansen, What OS are you running to talk with the Soekris box? Unless it's MSDOS or a derivative, stepping through the range of supported transfer rates from the terminal should do. Unless the feature itself is broken, of course. Bill On Sun, 2009-04-19 at 13:16 +0700, Hansen Cahyono wrote: Any suggestion on this? How can I fix this ? TIA, Hansen On Tue, Apr 14, 2009 at 9:36 PM, Hansen Cahyono hcahy...@gmail.com wrote: Hi all, I got a 4501 with 1.33 bios. I played around with the bios param and set conlock to disable now I cant seem to get the correct baudrate to communicate with the soekris. How can I reverse the conlock to enable again ? Thanks, Hansen ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Net5501 and Crux
Hi TechKid PC, On Sun, 2009-04-12 at 17:57 -0800, TechKid PC wrote: From: Christopher R. Hertel c...@ubiqx.mn.org That's why it's called comBIOS. The Soekris BIOS sends its output to the serial port. Ah thanks that explains it. That was the only way I assumed would prove useful for trouble shooting any boot errors but all the documentation I have found online did not mention it operated in that manner. Access to the Soekris comBIOS is not just useful for watching boot messages. It's an essential tool in case of trouble. See the Soekris documentation and the Soekris Wiki: http://wiki.soekris.info/Main_Page Last question I hope: Do you have any links that explains how to initiate a connection to the combios from another computer? I assume I use telnet or hyper terminal? What options should be used? Attach a terminal emulator (cu, minicom, hyperterminal, whatever..) to the serial line in question. The Wiki site should hold some info on how it's done for the more common terminal emulators. The most important thing, as Paul Bartell pointed out, is to have a null modem cable config. I use a null modem cable attached to a standard Serial-USB adapter for that. Have fun pressing Control-P;) Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Net5501 and Crux
Hi, Setting up serial support over USB has got really simple nowadays, as I recently discovered myself: http://www.linux-usb.org/USB-guide/x356.html Depending on how your kernel is configured, you might also have to modprobe one or more modules (module names seem to change over time/kernel versions, so I'll leave those for what they are). Adapters are quite cheap, and available throughout the Nets. With SSH only you (a) cannot reach the comBIOS, (b) won't be able to see vital messages at boot and (c) are in trouble if you find out you just uploaded a faulty ethX/sshd_config configuration file for example;). When working with Soekris boxes, the serial cable is you escape hatch: better have one at hand! Bill On Sun, 2009-04-12 at 01:43 -0800, TechKid PC wrote: Hello everyone I just signed up to the list and am having some difficulties with my Net5501 box. Ive had this unit for over 4 months now and still cannot get it up and running. Im using Crux 2.5 i586 with the 2.6.29 kernel. I am currently using a MicroSD-SD-CF II adapter. (Im trying to stick with MicroSD cards. I love the little guys and have tons of them). I setup a fully working Crux install on a dell optiplex test system using a CF-IDE adapter. I compiled in almost all the drivers the kernel offered in menuconfig. (I was worried I might of missed something so I went for broke.) I have not installed any third party modules or modifications to the kernel source, I also disabled module loading in the kernel for security. When I boot up the soekris box, the power and error lights are illuminated and the disk activity light blinks twice. Then the eth0 light (which is plugged in) turns on. About 5 seconds later the error light turns off and I have a solid power and eth0 (with occasional blinking). I do not have a serial cable. The way I have currently been testing the device is by simply watching the disk activity light. I have a friend who has the 4501 and he informed me it should blink for 30-60 seconds when booting. I set a static IP on eth0 to 192.168.2.10 so I can ping the device once it boots from another network computer. If I get a response I know it booted correctly and was able to load the basic required drivers. I would ideally like to configure everything via ssh and not have to worry about a serial cable. Once I get it booted and a response from eth0 I can simply login and make any necessary software installations and configurations. Any ideas on what steps I can take to get this up and running? I was hoping I would have something to show after 4 months and 30+ hours. :-( Thanks, Tom ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris] NET-4801 to NET-5501 reiserfs /dev/hda3 not working
Hi, Alt: try mounting the partition as root fs from a kernel/initrd image loaded over TFTP. On my net5501 using a 2.6.27 kernel it's /dev/sdaX, and not /dev/hdaX, indeed. Bill On Sat, 2009-04-11 at 09:18 -0700, William Estrada wrote: Does the box boot from the Flash? Can you install Grub on the flash to look at what drives are been seen? Date: Fri, 10 Apr 2009 06:08:53 -0700 From: AA6QN aa...@aa6qn.sytes.net Subject: [Soekris] NET-4801 to NET-5501 reiserfs /dev/hda3 not working What is interesting is that the initial LiveCD build that is installed with the Stage3 tarball (I think its 2.6.19-gentoo-r5) will see the /dev/hda3 but does not load the natsemi ethernet interfaces. I did boot to this kernel and tried to rebuild the latest 2.6.28 kernel using genkernel in the 5501 environment (the build did complete) but still no /dev/hda3 while trying to reboot the latest kernel. Putting the hd back into the 4801 and it boots just fine to the latest Gentoo kernel. Ideas ? Thank you in advance, JohnF ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] cannot install debian on net5501
On Tue, 2009-02-17 at 08:36 +0100, Bill Maas wrote: Hi Philippe, On Sun, 2009-02-15 at 09:00 +0400, philippe monroux wrote: De (from) (von) b...@stsx.org : On Sat, 2009-02-14 at 19:43 +0100, 686f6c6d wrote: I do not fully understand the remote rootfs/nfs method you used (if you did, but I assume it from the discussion). Could you, for completeness, document what you did? If you have an NFS root available, you can boot the Soekris box from it and install lilo onto the CF. Requires lilo on the NFS root + a chrooted apt-get install lilo onto the CF's root partition + a 'lilo -C' against the root partition. Make sure both lilo's are the same version. I understand that but it's not my installation method. Sorry but thanks. Using a Debian installer image you can install Debian directly onto the CF after a TFTP boot, but I think you (Philippe) already did that. Sure I do, Doesn't Debian have an option for installing lilo i.o. GRUB? no...only grub (is lilo old ?) No, lilo is different. It finds the kernel by means of a CHS value stored in the MBR (assuming that the BL lives there..). GRUB first finds the first sector of partition to boot from through CHS. This contains [part of] the secondary BL, which retrieves the kernel image through the filesystem. That's why you can boot arbitrary images with GRUB but not with lilo. But it's also the reason why more can go wrong with GRUB, e.g. with less reliable CF cards. See the soekris-tech archives. To my knowledge you won't find any 'lilo won't boot my Soekris box' postings in the archives. So GRUB and lilo are both useful depending on the context they are used in. But GNU/Linux people just like to obsolete stuff:). In this case pointing out the fact that these are different tools for different purposes would have been more appropriate. By the way, I think that if you lower the debconf priority level before installing Debian (append priority=high .. in the PXELinux config) ^^ Copy/paste error - low of course.. Bill the installer will ask if you want to install lilo. Bill Can't remember, but I suspect it is possible (it isn't with a standard Ubuntu, but that's Ubuntu..). A colleague of mine once stated that he would never trust anything that called itself Grand and Unified, he probably was right;) Perhaps the solution is to install grub on the 1rst partition (that is not on the mbr). I'll try this ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] cannot install debian on net5501
Hi Philippe, On Sun, 2009-02-15 at 09:00 +0400, philippe monroux wrote: De (from) (von) b...@stsx.org : On Sat, 2009-02-14 at 19:43 +0100, 686f6c6d wrote: I do not fully understand the remote rootfs/nfs method you used (if you did, but I assume it from the discussion). Could you, for completeness, document what you did? If you have an NFS root available, you can boot the Soekris box from it and install lilo onto the CF. Requires lilo on the NFS root + a chrooted apt-get install lilo onto the CF's root partition + a 'lilo -C' against the root partition. Make sure both lilo's are the same version. I understand that but it's not my installation method. Sorry but thanks. Using a Debian installer image you can install Debian directly onto the CF after a TFTP boot, but I think you (Philippe) already did that. Sure I do, Doesn't Debian have an option for installing lilo i.o. GRUB? no...only grub (is lilo old ?) No, lilo is different. It finds the kernel by means of a CHS value stored in the MBR (assuming that the BL lives there..). GRUB first finds the first sector of partition to boot from through CHS. This contains [part of] the secondary BL, which retrieves the kernel image through the filesystem. That's why you can boot arbitrary images with GRUB but not with lilo. But it's also the reason why more can go wrong with GRUB, e.g. with less reliable CF cards. See the soekris-tech archives. To my knowledge you won't find any 'lilo won't boot my Soekris box' postings in the archives. So GRUB and lilo are both useful depending on the context they are used in. But GNU/Linux people just like to obsolete stuff:). In this case pointing out the fact that these are different tools for different purposes would have been more appropriate. By the way, I think that if you lower the debconf priority level before installing Debian (append priority=high .. in the PXELinux config) the installer will ask if you want to install lilo. Bill Can't remember, but I suspect it is possible (it isn't with a standard Ubuntu, but that's Ubuntu..). A colleague of mine once stated that he would never trust anything that called itself Grand and Unified, he probably was right;) Perhaps the solution is to install grub on the 1rst partition (that is not on the mbr). I'll try this ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] cannot install debian on net5501
Hi, On Sat, 2009-02-14 at 19:43 +0100, 686f6c6d wrote: I do not fully understand the remote rootfs/nfs method you used (if you did, but I assume it from the discussion). Could you, for completeness, document what you did? If you have an NFS root available, you can boot the Soekris box from it and install lilo onto the CF. Requires lilo on the NFS root + a chrooted apt-get install lilo onto the CF's root partition + a 'lilo -C' against the root partition. Make sure both lilo's are the same version. Using a Debian installer image you can install Debian directly onto the CF after a TFTP boot, but I think you (Philippe) already did that. Doesn't Debian have an option for installing lilo i.o. GRUB? Can't remember, but I suspect it is possible (it isn't with a standard Ubuntu, but that's Ubuntu..). A colleague of mine once stated that he would never trust anything that called itself Grand and Unified, he probably was right;) Bill - If you created the root filesystem on another machine and not on the soekris, did you create it with the correct cylinder/header/sector values? (see http://wiki.soekris.info/What_is_this_disk_geometry_thing) - Is the partition bootable? ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] cannot install debian on net5501
Hi Philippe, from support.microsoft.com (uhuh): If the client cannot connect, you may receive an error message similar to: PXE-T01: File not found PXE-E3B: TFTP Error - File not Found Note _connect_: the error message is deceptive, at least in this case (it might also occur if the file really isn't there or isn't readable I I guess - I'm not sure if it would be exactly the same then). Since this can't be a cable issue (got an IP address) it's probably a firewall issue. Other sources of trouble are: - hosts.allow - hosts.allow + inetd/tcpd (if I recall well tcpd only reads the old hosts.allow(5) version, not the hosts_options(5) variant - always good for a few hours of frustration) - file permissions and ownership in [/var/lib]/tftpboot - with tftpd -s : pxelinux.0 is a symbolic link pointing outside the TFTP boot environment. That's a very silly one, but it has happened to me.. That's quite a lot to take care of at the same time, especially if the system is already configured for other roles (which might require TCP wrappers for example). Fix (hopefully): Try again with tftpd as standalone daemon and hosts.allow/deny and firewall disabled - if that's legally possible - and if it works gradually throw the various security features back in again. Otherwise check file permissions etc. Bill On Mon, 2009-02-09 at 13:56 +0400, philippe monroux wrote: Hi all, I've followed : http://wiki.soekris.info/Installing_Debian_Linux_(4.0r1_Etch) http://www.debian-administration.org/articles/478 but I get: == comBIOS ver. 1.33 20070103 Copyright (C) 2000-2007 Soekris Engineering. net5501 0512 Mbyte MemoryCPU Geode LX 500 Mhz Pri Mas CF 128MBLBA 490-16-32 125 Mbyte Slot Vend Dev ClassRev Cmd Stat CL LT HT Base1Base2 Int --- 0:01:2 1022 2082 1010 0006 0220 08 00 00 A000 10 0:06:0 1106 3053 0296 0117 0210 08 40 00 E101 A0004000 11 0:07:0 1106 3053 0296 0117 0210 08 40 00 E201 A0004100 05 0:08:0 1106 3053 0296 0117 0210 08 40 00 E301 A0004200 09 0:09:0 1106 3053 0296 0117 0210 08 40 00 E401 A0004300 12 0:20:0 1022 2090 06010003 0009 02A0 08 40 80 6001 6101 0:20:2 1022 209A 01018001 0005 02A0 08 00 00 0:21:0 1022 2094 0C031002 0006 0230 08 00 80 A0005000 15 0:21:1 1022 2095 0C032002 0006 0230 08 00 00 A0006000 15 5 Seconds to automatic boot. Press Ctrl-P for entering Monitor. comBIOS Monitor. Press ? for help. boot f0 Intel UNDI, PXE-2.0 (build 082) Copyright (C) 1997,1998,1999 Intel Corporation VIA Rhine III Management Adapter v2.43 (2005/12/15) CLIENT MAC ADDR: 00 00 24 CB 54 20 CLIENT IP: 10.0.0.4 MASK: 255.255.255.0 DHCP IP: 10.0.0.1 GATEWAY IP: 10.0.0.138 TFTP. PXE-T01: File not found PXE-E3B: TFTP Error - File Not found PXE-M0F: Exiting Intel PXE ROM. No Boot device available, enter monitor. comBIOS Monitor. Press ? for help. == I've gogled but nothing. A lot of thanks ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] reading USB Keyboard on net4801 / OpenBSD 4.4
On Mon, 2008-12-22 at 10:11 +0800, David Schulz wrote:
On Wed, Dec 17, 2008 at 08:25:19PM -0500, Jed Clear wrote:
On Tue Dec 16 22:49 , David Schulz sent:
I recently purchased 2 net4801 that now run OpenBSD 4.4 Release.
I attached a USB Keyboard to the USB Port.
In my dmesg, the USB Keyboard gets attached to /dev/wskbd1
I want to automatically run a Program at boot that logs and processes
Keyboard
Input. How can i get to the Keyboard Input made on a USB Keyboard that i
attached to my Soekris USB Port? Means: How can i capture whatever is
entered
on the USB Keyboard, so i can use it in my program?
Here is a hypothetical Perl Program that takes Input and writes it to a
file.
I am aware that noone will see the print Statements when there is no
screen,
they are just for examples sake when i run this from a ssh session:
#!/usr/bin/perl
use strict;
use warnings;
open(my $fh, '', '/var/log/reader.log') or die $!;
while ('TRUE') {
print Enter ID :;
chomp(my $id = STDIN);
print $fh You entered $id\n;
if ($id eq 'exit') {
print Exiting\n;
close($fh);
exit;
}
}
The above will not work for me, because when the Machine is running
unattended, there is noone who logged in, the Program wouldn't know where
to
take the input from. I need to somehow make my Program read directly
from the Keyboard Device Entry, and take its Input from there.
You're on the right track. You need to be reading from that keyboard, not
standard in. My Perl is very rusty, but you need to either remap STDIN to
/dev/wskbd1, or just open the keyboard device as a file and read from that
file
handle. I'd probably go with the latter. Either way I think you need
Perl's
open function. If you don't have a Camel or Llama handy, google for
perldocs.
The other problem is how to run it unattended. Unix gives you many ways to
do
this, including: cron, boot RC scripts, /etc/inittab, and /dev/ttys. The
latter
has traditionally dealt with terminals. I'm more familiar with FreeBSD and
SunOS, so YMMV with OpenBSD as to the exact file used for those.
Can anyone
help me out, it would be great. I hope i have been concise but clear with
my
Problem.
Works for me. A lot better than one frequently gets on other forums.
-Jed
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech
!DSPAM:4949a729675011436264198!
Hi,
there are a couple ways to start the program unattended, my main problem is
still the 'binding' to the keyboard device. In my example Code i am trying to
basically open the /dev/wskbd1 as a file and read from it, but i only get
garbled characters. The reason would be that the Keyboard doesnt send a
literal A when i hit the letter 'A', but instead some ascii code that has to
be translated by the Operating System before it is output on my screen.
If anyone can share his ideas on how to read from a USB Keyboard that is
attached to the Soekris-4801 USB Port, please share.
Thanks and best regards,
David
Hi David,
I'm not a Perl programmer, but shouldn't the links below at least give
you something to start with?
(terminal handling / raw input mode)
http://docstore.mik.ua/orelly/perl/cookbook/ch15_07.htm
(input char mapping)
http://perldoc.perl.org/functions/chr.html
Hope that these are not telling you things you already knew. In any
case, keymap would be the first thing I'd look for here.
Bill
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech
[Soekris] Instant Boot Server
Hi, This might be of interest to the Debianist amongst us: I've been working on a shell script for setting up an on-demand boot server for the last couple of weeks. For fun an non-profit and stuff.. And because I needed one;). The result is here: http://www.stsx.org/debian/ibs/ The NFS root part is still a bit probelmatic, as the NFS kernel server refuses to give up /proc after the service has supposedly been brought down. Otherwise it seems to work well over here. Any contributions to the code, documentation or whatever would be welcome. Please don't hesitate to inform me about bugs, even if the tarball refuses to unpack to start with (it happens..). I think I'll sourceforge it this week. In the meantime, feel free to shamelessly promote it if you find it useful. Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501 boot from cdrom
Hi Ebbe, On Sun, 2008-12-07 at 20:45 +0100, Ebbe Hjorth wrote: - Message from [EMAIL PROTECTED] - Date: Thu, 4 Dec 2008 17:50:35 +0100 (CET) From: Ebbe Hjorth [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] Subject: Re: [Soekris] 5501 boot from cdrom To: soekris-tech@lists.soekris.com Hi, I tested booting from a CD and it won't work, the Net5501 doesn't support that, no matter if the drive is connected via USB, SATA or DIE. Robin So there is only the netboot way? Thats not for a newbie like me ;) The simplest way is usually to plug your disk/CF into a desktop machine and do a minimal install there, then enable serial port or ssh access and put the thing back into the Soekris box. I have done that - And it seems to be working okay, i just removed the flash card, and inserted a sata disk, just to have more speed here in the test period. in /etc/fstab i changed the 'dax' devices to 'da1' and the soekris is booting, but instead of the login prompt, it just writes: snip Starting sshd. Starting cron. Local package initialization:. Starting background file system checks in 60 seconds. Sun Dec 7 19:31:16 CET 2008 DàDàDàDà Its the last line i dont understand, where is my login prompt? ;) With OpenBSD, and using cu(1) for monitoring the client, I get weird characters like these at this point, even if both systems are otherwise set up correctly. And since everything else works well, I just ignore these. With both Debian and OpenBSD (on any side of the serial cable) I need to press Enter at the monitor once to get the login prompt. I don't know what causes this, but I can live with it. Bill Another thing, my brand new sata disk, clicks once i a while, and i have never heard that before on my other hardware, is that a soekris feature? ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501 boot from cdrom
On Mon, 2008-12-08 at 10:57 +0100, Ebbe Hjorth wrote: Hi Ebbe, On Sun, 2008-12-07 at 20:45 +0100, Ebbe Hjorth wrote: - Message from [EMAIL PROTECTED] - Date: Thu, 4 Dec 2008 17:50:35 +0100 (CET) From: Ebbe Hjorth [EMAIL PROTECTED] Reply-To: [EMAIL PROTECTED] Subject: Re: [Soekris] 5501 boot from cdrom To: soekris-tech@lists.soekris.com Hi, I tested booting from a CD and it won't work, the Net5501 doesn't support that, no matter if the drive is connected via USB, SATA or DIE. Robin So there is only the netboot way? Thats not for a newbie like me ;) The simplest way is usually to plug your disk/CF into a desktop machine and do a minimal install there, then enable serial port or ssh access and put the thing back into the Soekris box. I have done that - And it seems to be working okay, i just removed the flash card, and inserted a sata disk, just to have more speed here in the test period. in /etc/fstab i changed the 'dax' devices to 'da1' and the soekris is booting, but instead of the login prompt, it just writes: snip Starting sshd. Starting cron. Local package initialization:. Starting background file system checks in 60 seconds. Sun Dec 7 19:31:16 CET 2008 DÃ DÃ DÃ DÃ Its the last line i dont understand, where is my login prompt? ;) With OpenBSD, and using cu(1) for monitoring the client, I get weird characters like these at this point, even if both systems are otherwise set up correctly. And since everything else works well, I just ignore these. With both Debian and OpenBSD (on any side of the serial cable) I need to press Enter at the monitor once to get the login prompt. I don't know what causes this, but I can live with it. Okay, but trust me - I tried pressing every button i got ;) In your case, the weird characters string looks like a sequence of 8-bit terminal control chars being sent to an uninitialized terminal. Maybe a lead? Either that or the serial console isn't set up properly after all. Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501 boot from cdrom
On Thu, 2008-12-04 at 18:15 +, Andrew Thomas wrote: PXE boot isn't really for the faint hearted. For a start you will need a Linux PC with a DHCP server and a proper PXE configuration/image. You can copy files straight to the SD card - but watch out for file types (especially if you use a Windows machine to do it). One thing I haven't tried (as I use PXE) is inserting a HD in to a PC, installing Linux on it and then putting that HD in the 5501 box (assuming you want to keep it in there). Alas, those days seem to be over: I found machine-dependencies recently after sticking a CF with Debian etch on it, installed on a net4801, into a net 4501. Something to do with that incredibly unfollowable udev stuff. Incredibly frustrating as well, I really don't want to manually edit some file with a name like z25_persistent-net.rules, just to be able to move a disk. Or maybe I missed something.. Excuse me for being pretty off-topic. Bill I think I'll have to write a basic tutorial for Debian PXE installations. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Robin Kipp Sent: 04 December 2008 17:44 To: [EMAIL PROTECTED]; soekris-tech@lists.soekris.com Subject: Re: [Soekris] 5501 boot from cdrom Hi, So there is only the netboot way? Thats not for a newbie like me ;) Well, most people say it isn't that hard and there are some tutorials on the Soekris wiki. What you can also do is install whatever os you want on the Compact Flash card using a normal computer and card reader, but I think you'd then have to modify the fstab file... Robin ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4511 and POST's steps
Hi Alexander, to be honest, the -03- looks like a secret error code to me. Maybe Mr. Kristensen would like to shed some light on this? Probably not. Anyway, it's been reported before, but it didn't show up in my underpowered-PSU case. It looks like your only option is to ship the boxes back to Soekris Dot Com. Bill On Thu, 2008-11-27 at 15:02 +0200, Alexander Pogrebny wrote: Hi Bill. Yes, unfortunate, boot sequence completely stops after this message. This is not the power problem, because we tried lots of PSU. And we tried power on this boards without any radio or CF cards. We also tried to solder flash with BIOS (also 1.33) from another board that works fine - we got the same result. So, problem is not in BIOS. I just don't understand, why manufacturers can't tell where we should look the problem. They should know, where the problem if POST stops at this position. -- Alexander Pogrebny engineer Unidata, Ukraine 2008/11/26 Bill Maas [EMAIL PROTECTED]: Hi Alexander, does the boot sequence completely stop after this message or does it go into an endless reboot sequence?. In the latter case you may need a more powerful PSU. I had a similar problem recently with a net5501 and I found that while its nominal power usage with a [more or less] idle OS and no extra cards attached is 5.7W-6.3W, right after poweron there's a temporary surge towards almost 9W. All works well here after swichting to a 12V 1.25A CUI Inc. EPA-151DA-12 PSU obtained from kd85.com. Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4511 and POST's steps
Hi Alexander, does the boot sequence completely stop after this message or does it go into an endless reboot sequence?. In the latter case you may need a more powerful PSU. I had a similar problem recently with a net5501 and I found that while its nominal power usage with a [more or less] idle OS and no extra cards attached is 5.7W-6.3W, right after poweron there's a temporary surge towards almost 9W. All works well here after swichting to a 12V 1.25A CUI Inc. EPA-151DA-12 PSU obtained from kd85.com. Bill On Mon, 2008-11-24 at 21:58 +0200, Alexander Pogrebny wrote: Hi! Eight boards from our last order (hundred of net4511) are broken. All boards was with BIOS v1.33. They's POST stops at: POST: 012345689bcefgh*-03- Maybe someone can tell, where we should find the problem. We just want to try repair them for getting them faster to work. -- Alexander Pogrebny engineer Unidata, Ukraine ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Installing Debian Linux from a CD
Hi Robin, there's another complication with the kernel+initrd+rdev approach: IDE != FDD.. So I doubt whether the comBIOS will load the kernel without a partition table in sector 0. In fact, I've seen [IBM] harddrives ceasing to be recognized as such after zeroing out the boot sector. Still, how hard would it be to install a Debian onto a second disk? As I understood from your previous messages, the Debian Installer offers brltty support. Or boot from some live CD. But to take the problem back to its origin: your claim was that a _Windows_ (let's keep that clear!) TFTP boot doesn't work for a net5501 with comBIOS V1.33c. And that you don't want to downgrade to 1.33b, which did work, because there is some claim out there about a comBIOS upgrade possibly resulting in hardware damage. All that the Soekris download page states on this issue is that Please note: It is not recommended to downgrade a BIOS to an older version than the one a board is supplied with. So my burning question to this list is: in what way could downgrading the comBIOS to the lowest version supported by a particular board type do any damage to the hardware or otherwise? Because I don't see it. Back with Debian again: in addition to the other image types proposed: maybe a Debian businesscard install image is just what you were looking for: http://www.debian.org/CD/netinst/ Bill On Sat, 2008-11-22 at 00:50 +0100, Robin Kipp wrote: Hi Bill, that's an interesting idea, hehe... :-) The only problem is this also requires a running Linux I could use to copy the kernel onto the CF card which I don't have :-( Robin -- Debian Package Management System: The Horror. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Installing Debian Linux from a CD
On Fri, 2008-11-21 at 23:35 +0100, Bill Maas wrote: Hi Robin, for the really desperate there is yet another escape, which I've tried with floppies with success, but never with a CF card: - dd a kernel onto the boot device starting at sector 0 - dd an initrd image right after it using the seek or skip option (this involes keeping track of the kernel size in sectors!) Error: kernel size in bytes. No big deal really.. - set the root device offset in the kernel with rdev(8). From the manual: Its use is strongly discouraged. Use a boot loader like SysLinux or LILO instead. But those aren't available, are they?. Agreed, it's dirty, hackish. low-level stuff, and you'll have to figure out the details yourself, but in your case it may be just the tool you need. I'm not even sure if the Soekris comBIOS will be able to load the Linux kernel just like that (with floppies + a PC it always worked). If it does, and if you use the Debian Installer image as the initrd image, you should be able to install Debian GNU/Linux as if nothing weird had ever happened;). Bill On Thu, 2008-11-20 at 21:04 +0100, Robin Kipp wrote: Hi Bill, Good idea... The only problem is I don't have a Linux system I could use to copy Debian on the CF card :-( Robin -- Debian Package Management System: The Horror. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] debian bootloader on a 4501
Hi Esteban, I'd start with dropping the separate /boot and see if it works, or making /boot, say, 20MB to begin with. RedHat's installer used to have unsane requirements for /boot, it wanted at least 50MB while only 35MB was going to be used (exact numbers may be slightly off, they're just for the idea). But that's only a suggestion.. Bill On Thu, 2008-11-13 at 18:49 +0100, Esteban Ribičić wrote: hello, i pxe'booted a 4501 (Bios 1.32) with a 2gb CF and installed a normal debian etch installation (/boot -10mgs, boot flag .. /var and /) ... at the end it directly installed grub and failed without much of a message. i tried to use the menu to install lilo (as suggested on the wiki) and also failed... not sure how to get out of this trouble or what i did wrong (except of just pressing enter!!) any suggestion? i poked arund forums and couldnt find much .. just a few notes and dead links :( oh, i'm running a windows laptop, so if ineed to mount hte CF to adjust a few files i'm a bit screwed...heheeh for a change! hehe salut!! ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- Debian Package Management System: The Horror. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] debian bootloader on a 4501
[Follow-up to my previous message] Maybe best start with lowering the Debconf priority in the kernel commandline and see if you get some error message, for example (with PXELinux): LABEL blah kernel blah append priority=low By the way, and slightly on-topic: I found that the order in which options are passed on the Linux kernel commandline matter when trying to boot a net4801 from a Debian NFS root: NFS Error: short read: 24 28 Diag: Wrong kernel boot options order in pxelinux config Fix: Change the order of kernel boot options in /tftpboot/pxelinux.cfg/file: #append root=/dev/nfs initrd=initrd.img console=ttyS0,19200n8 nfsroot=192.168.1.6:/mnt/deb ip=dhcp rw append console=ttyS0,19200n8 initrd=initrd.img root=/dev/nfs nfsroot=192.168.1.6:/mnt/deb ip=dhcp rw Maybe this is a well-known fact, but it definitely wasn't to me, and I don't see why it should matter. Bill On Thu, 2008-11-13 at 18:49 +0100, Esteban Ribičić wrote: hello, i pxe'booted a 4501 (Bios 1.32) with a 2gb CF and installed a normal debian etch installation (/boot -10mgs, boot flag .. /var and /) ... at the end it directly installed grub and failed without much of a message. i tried to use the menu to install lilo (as suggested on the wiki) and also failed... not sure how to get out of this trouble or what i did wrong (except of just pressing enter!!) any suggestion? i poked arund forums and couldnt find much .. just a few notes and dead links :( oh, i'm running a windows laptop, so if ineed to mount hte CF to adjust a few files i'm a bit screwed...heheeh for a change! hehe salut!! ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- Debian Package Management System: The Horror. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] console hang/pxebooting OpenBSD 4.4
On Tue, 2008-11-04 at 11:01 -0500, jmc wrote: --- David Alexander [Mon, Nov 03, 2008 at 10:04:53PM -0700]: --- On Mon, 3 Nov 2008 14:03:52 -0500, jmc [EMAIL PROTECTED] wrote: at any rate, i copied the pxeboot and bsd.rd images into my tftp root. in the Soekris boot monitor, i type `boot f0' and since i don't have a boot.conf in place in my tftp root, i set up the console tty by hand. namely: set tty com0 stty com0 19200 and then just a `boot bsd.rd' my console (using cu(1) on a newly upgraded 4.4 box) just hangs thusly: http://www.cosimano.us/~jmc/hang.png I just installed 4.4 onto a 5501 last weekend and I saw something very similar. In my case I had a boot.conf that ran the same three commands you're doing manually, but I would see the same garbage after switching console and then a few seconds later the boot would continue. The solution was obvious once I thought about it. Change the sequence of your commands to set the baud rate BEFORE you switch to com0. Run the commands in this order: stty com0 19200 set tty com0 boot bsd.rd Hmm.. interesting. Here's from a doc I wrote: set tty com0 stty com0 19200 boot tftp:/bsd.rd Order matters here! Unlike its name suggests, boot.conf is a command file and not a configuration file. I was referring to setting the com port options before issuing the kernel boot command, but I never gave the set tty/stty command order a second thought. And the way it is shown in this example it's listed in almost every xBSD doc I've seen (the Neophasis folks had it right, though). Time for a few updates.. Bill this was exactly the fix that worked for me. i've now created a boot.conf and used the commands in this order. thanks, David---and all others, for the help. --john ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Using the Serial Port For Other Purposes
Hi Hal, I tried messing with the serial port used for the system console at runtime and I lived to regret it;). You should be able use com1 for input, see here http://wiki.soekris.info/5501_Interfaces for details. Bill On Sat, 2008-10-25 at 01:01 -0400, Hal Vaughan wrote: I'm using a Net5501. I have a HD Radio that is controlled through a program I wrote that can talk to it over a regular serial port or through a USB port if I use a USB/serial adaptor. Right now I'm using it with the USB adaptor but I'd prefer to use the serial port. I've noticed, though, when I try to access it through /dev/ttyS0 I don't get consistent results, which I would suspect is due to that port also being used for the console. Do I need to do anything special if I want to use the on board serial port for something other than the null-modem cable? Preferably I'd like to be able to have something in my boot procedure that changes this over so if I have to, I can always reboot with the null modem cable on instead and could use the console directly if I wanted to. I'm using Debian Etch on this system. Thanks! Hal ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Voyage Linux problem
Hi Robin, maybe you should consider sticking with a well-documented mainstream OS distro for the time being, i.o. trying out some specialized piece of kit. Because A: where are you going to find a CF card with only 64 or 128 MB anyway? and B: I'm not sure why each and every single application needs its own Linux distribution these days -- its becoming ridiculous IMHO (publishing a set of config instructions or some configuration script would probably do in most cases). Performing a Debian TFTP install on a Soekris box is a piece of cake. If you master apt-get and/or dselect and are willing to spend some time searching the docs/net/howtos for setting things up, you will find your box up and running in no time, and you will learn a lot about your system in the process. With plain old Debian, it's all been done before, and you will have to look hard for something that isn't documented or answered on some Debian forum or mailing list. Have fun! Bill On Fri, 2008-10-17 at 17:26 +0200, Robin Kipp wrote: Hey, Thanks a lot for this commands!!! Where can I actually find them and some docs on the Voyage Linux website? I've been looking through the wiki, but the most useful I could find there for now was the install guide. Btw, from experience, how close do you Voyage users think this distry gets to the real Debian Etch? Robin ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Radio Shack K Modular Connector
On Sat, 2008-10-11 at 22:15 -0400, Brian A. Seklecki wrote: Nice. Yea -- I wasn't sure if there was an industry standard for DC power connector form factors. Given the nature of the industry, probably many;) Bill JameCo.com is probably the definative source (the website doesn't do it service -- you have to get the printed catalog) ~BAS On Sat, 11 Oct 2008, Darrick Hartman wrote: Brian, I do use the PCI slot in most cases and use the following power supply with good success: http://www.logicsupply.com/products/pw_ac_60w There are many places that sell similar external blocks like that in that same price range. Rat Shack can be helpful in a pinch, but definitely over priced. Darrick Brian A. Seklecki wrote: All: Also, if you forget to order a power supply and don't feel like waiting, the K size modular connector from Radio Shack modular DC power transforms seems to work. They max at 1.5amp/1500mA, but w/o minipci, USB, sata, it doesn't seem to draw the full 3 AMPs Just be prepared to grab your ankles on the price. Not sure how that place stays in business. l8* -lava (Brian A. Seklecki - Pittsburgh, PA, USA) http://www.spiritual-machines.org/ Guilty? Yeah. But he knows it. I mean, you're guilty. You just don't know it. So who's really in jail? ~Maynard James Keenan ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] CompactFlash cards and Soekris boards
Hi Oleg, On Fri, 2008-10-10 at 10:10 -0700, Oleg Smolsky wrote: Bill Maas wrote: On Thu, 2008-10-09 at 16:15 -0700, Oleg Smolsky wrote: Hey there, I'm thinking of getting the new net5501 board and would like to find out the common way people use CF cards. I assume the board can boot from a CF card, and my question is whether can I create an ext3fs partition on the CF, mount it on / and use it as if I had a normal IDE/SATA drive? Or do I have to boot from a filesystem image in read-only mode? Yes and no (in that order). To the Soekris comBIOS the CF is just an IDE drive, and you can format/partition/mount it any way you like. If you don't need to store excessive amounts of data and you can live with not-very-hot write speeds during installations and updates/upgrades, a CF card is in fact a much more practical thing than a real HDD, if only because you can get about 12 of them for the price of 1 HDD;). Right, thank you for the reply - the main point is that it looks like an IDE device and can be partitioned like one. This is going to be a replacement router (iptables, openvpn) and I want to be able to run apt-get install ... on the box. The Debian install will be configured with no swap. The next thing that I need to figure out is how to get /var into its own (cramfs?) filesystem and flush it to disk when the box reboots. Has anyone tried this type of thing? With OpenBSD that's very easy, in fact it's practically pre-done for you;). For Linux, this might point you in the right direction: http://www.linuxdevices.com/articles/AT4540125636.html I have good experiences with rsync for flushing /var's contents at shutdown, but you will probably have to set that up yourself. Here's an (OpenBSD) example: http://www.stsx.org/software/openbsd/mfsmount.txt That's a shell script by the way, it just has a .txt suffix to avoid browser annoyance. Here's the story: http://www.stsx.org/software/openbsd/README.mfsmount.txt Good Luck, Bill Also, re: net5501 board - does it work with Debian's stock linux-image-2.6-686 (lenny) image? Is there any merit in building a more-specific AMD Geode kernel? Thanks again, Oleg. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] CompactFlash cards and Soekris boards
Hi Richard, I was talking about write speeds, not about CF card lifespan. And yes, of course it also depends on what people are going to push into the DB. I wouldn't set up a CVS server on a CF card myself, just because in general those cards are still quite slow, the affordable ones at least. Bill On Fri, 2008-10-10 at 07:05 -0700, Richard Gooch wrote: Bill Maas writes: It depends on the application you have in mind. For a router, the CF is perfect, for a DB backend server it most likely isn't. Maintaining a swap partition is also not a viable option with a CF card. This isn't actually true. Even the older CF cards have good enough wear levelling. I've got a mail server that's nearly 4 years old. It runs off of CF. It hasn't missed a beat. By the time it dies I'll probably be able to buy a new CF card with 100x the capacity, a better wear-levelling algorith, more bandwidth and it will cost peanuts. Why do people still believe this FUD? Regards, Richard -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501 getting stuck on coldboot
Hi all, is this the both the error and disk LEDs stay lit up issue? I had it myself again just an hour ago. It happens every now and then with no obvious reason after a power failure, with CF mounted internally and no additional HDDs or other exotics attached. Most of the time everything works fine, but this thingie does make the board unreliable. Bill On Tue, 2008-10-07 at 09:49 -0400, Mike Tancsa wrote: At 07:59 AM 10/7/2008, Sebastian Benoit wrote: Hi, I have a soekris 5501 with comBIOS ver. 1.33 that hangs when cold-booting. I have tried with 2 different flash cards an a hard drive. another CF card worked ok. I gather it is recommended to update the BIOS to version 1.33c? Contact Soekris for an RMA. We went through this issue with several boards. We just got 4 replacement units yesterday that supposedly are from a new batch that fix this issue once and for all. All 4 were good and we have yet to see the issue resurface. ---Mike ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Talk me out of buying a net4801
On Fri, 2008-09-26 at 09:44 +0100, Adam Retter wrote: If you hacker side is emerging then does it really have to be pfSense or m0n0wall? Why not go the whole hog and do your own setup. Word. Bill;) I have a net 4801 that I have had running for a few years now with OpenBSD on it and PF. There are blenty of PF tutorials and documents out there for people who have some time to spare. 2008/9/26 Chris Boot [EMAIL PROTECTED]: Jim, If you're looking to run pfSense I've found it doesn't work so well anymore on a net4801, you need something beefier to get anywhere with it. pfSense is less designed for embedded hardware than it is for full-blown PCs, while m0n0wall is still aimed at the smaller devices (and even that is starting to have trouble with less-powerful machines like the 4801). I'd really recommend a net5501 for any new projects, despite the added cost, size, and no doubt power consumption... Chris Jim Arnold wrote: Thanks in advance for humoring me with my unimportant story and questions... For the last few years a FreeBSD box running IPF has been protecting my little home cable modem network. Back then I was having fun hacking around teaching myself Unix. Once I finally migrated to OSX from System 9 I stopped playing around with Unix except for my little firewall and occasional peeks under OS X's hood. A recent storm took out my power supply on that firewall box. Unfortunately it was a proprietary power supply. I liberated a FreeBSD box out of the closet loaded with FreeBSD 4.11 and tried to duplicate my old firewall. Despite extensive documentation of my old system files I could not get this new box to work as my firewall. While searching around for a solutions I came across m0n0wall and PFSense. Getting them to install and work was very simple, especially PFSense, which was installed on a hard drive. Next I read about using a CF card instead of a hard drive. A few days later my IDE/CF adapter arrived and in minutes I had PFSense running off the compactflash card. My old Geek Hacker self was re-emerging. :) Then I read about the Soekris boxes. My router/firewall is in the basement so noise is not an issue. Now that I have the CF installed a power supply failure is the only weak link in the box. But since this was a home-built box it has a standard PS which will be easy to replace if necessary. The only reason I can come up with to plunk down $243 for a 4801, power supply and shipping, is to save power. I calculated my current router/firewall expends about 200 watts to run. That costs me $110 a year to power If my calculations are correct. If the 4801 takes 10 watts to power that would cost $5.50 a year. Is this a good enough rationalization? What is a reasonable lifespan of a Soekris box? ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] SanDisk 64 MB Compact Flash Cards
Hi, 256 MB generally appears to be the smallest size available for retail, see for example: http://www.flash-memory-store.com/256mb-compact-flash.html But here they even sell 32 MB: http://www.memorysuppliers.com/sandisk1.html?gclid=CN-CwMvmxpUCFRKS1QodhDG0iw [I just can't help wondering where those cards come from..] By the way, assuming that you are going to load the image into RAM and run everything from there: you don't need to use a Sandisk or any other expensive state-of-the-art industrial-grade brand CF card. Read speed is generally not a huge issue at boot time. I've tried out various cheapos and I've been using Dane-Elec cards for over a year now with frequent data updates and no problems whatsoever. Write speed is acceptable. Prices are modest. Not all CF brands appear to work well on a Soekris box, you will have to check the archives for trouble with specific brands. I don't use 2.5 drives, so I don't know how well the CF cards I use work together with those. As for the overkill: try find a brand new 10 GB hardisk.. Bill On Sat, 2008-09-06 at 00:33 +0700, Soekris Maillist wrote: Anyone know where to buy these? The smallest ones I can find here in Thailand seems to be 2 GB, which is extreme overkill for a 16 MB image. Thanks in advance. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Development suggestions
Hi Thomas, if you need to make lots of frequent little changes for testing, using an NFS root might be more practical. With Linux it's fairly easy (PXELinux + kernel boot options), with OpenBSD it's a bit trickier (see Embedded OpenBSD at www.kernel-panic.it). With other OSes I don't know, but there are likely to be numerous docs about the topic out there. Bill On Sun, 2008-08-31 at 18:03 +0700, Soekris Maillist wrote: Hi everyone, I know there are many ways to develop for the Soekris boxes, but I am interested in hearing why you have chosen your method. I find my own method a little slow, since it require a lot of swapping of CF cards. Once a new build is completed in my chroot environment I make the initramfs, copy it to the CF card along with the kernel. Move the CF card to the Soekris box for testing. Since the root environment is created from initramfs I can't really do any changes/modifications on the Soekris box so it is a rinse and repeat method. Can anyone suggest alternate methods of testing builds, that would be a little more streamlined? I apologize if this is considered a n00b question. Thanks in advance, Thomas ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4801 Bios update
Hi Peter, GRUB seeing a corrupted FS in stage 2 sounds more like disk geometry trouble (see the Soekris Wiki pages). The xmodem upload issue is almost certainly unrelated, since nothing is written to disk. You might want to give cu/lrzsz a try for that one. Bill On Sun, 2008-08-24 at 15:03 +0200, Peter J. Holzer wrote: Background: I'm trying to install Debian Etch on a net4801. The installation works, but grub aborts with an Error 16 (filesystem corrupt) while loading stage 2. In a previous thread it was suggested that this might be a problem with ComBIOS version 1.29 and that upgrading to 1.30 (or newer) might help.[1] So I grabbed b4801_133.bin from http://www.soekris.com/downloads.htm, booted the net4801 into the monitor, issued a download command, and selected send file - xmodem in minicom. The result looked like this: +---[xmodem upload - Press CTRL-C to quit]+ |Sending b4801_133.bin, 608 blocks: Give your local XMODEM rec| |eive command now.| |Xmodem sectors/kbytes sent: 0/ 0kRetry 0: NAK on sector| |Retry 0: NAK on sector | |Retry 0: NAK on sector | ... |Retry 0: NAK on sector | |Retry 0: NAK on sector | |Retry 0: Retry Count Exceeded| | | |Transfer incomplete | | | | READY: press any key to continue... | +-+ The serial port on the PC is a real serial port (no USB to RS-232 adapter). Software flow control was off. I tried it with and without hardware flow control (though at 19200 bit/s I didn't expect flow control to be necessary) - same result. I also tried a block size of 1k instead of 128 Bytes - same result. Finally, I tried it with Hyperterm on a Windows PC. Didn't work either, but the error messages were a bit different (something like unsolicited nak received (translated from German)). hp [1] The OP in this thread got it to work with LILO instead of grub - that's what I'll try next, but I think upgrading the BIOS should work, too. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] List is quiet
On Sat, 2008-08-23 at 20:41 -0700, Daniel E. Hassler wrote: The word harvester might be more appropriate. Sadly there are folks out there harvesting email addresses and sending SPAM :( IMHO the word scraper does not only sound cooler, but it also has the appropriate negative ring to it in this case. Harvesting is a friendly business. Scraping on the the other hand implies considerable effort (as der Mouse pointed out), usually with no consideration for, or even at the expense of, others, and ultimately oneself. As in scraping together a huge fortune, with Dickens' Scrooge as the prototype scraper. Hacking hotmail dot com to obtain a million e-mail addresses at once is one thing, scouring unassuming mailing lists like this one for a few hundred e-mail addresses another, and definitely pathetic. I'm very much in favour of lifting the word Scraper into the official IT jargon, if it will make those who scrape - if only for a moment - a bit sad about themselves (it probably won't, though, because they're used to feeling that way). Bill der Mouse wrote: Some @(*#$ idiot decided to write his own scraper and generated [so] many errors that the log partition ran full. As English is not my mother tounge, I even ran scraper through a dictionary. [...] Here, scraper means an automated piece of software to gather up something from somewhere - usually, email addresses from webpages. (Why the name, I'm not sure. Perhaps derived from scrape together, meaning approximately gather, but with a connotation of difficulty; perhaps derived from the image of scraping something off something else, the way you might scrape frost off a window, or dried food off a stove.) /~\ The ASCII der Mouse \ / Ribbon Campaign X Against HTML[EMAIL PROTECTED] / \ Email! 7D C8 61 52 5D E7 2D 39 4E F1 31 3E E8 B3 27 4B ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] List is quiet
On Mon, 2008-08-25 at 10:47 +, Poul-Henning Kamp wrote: In message [EMAIL PROTECTED], Bill Maas writes: I'm very much in favour of lifting the word Scraper into the official IT jargon, [...] You'd be years late for that :-) Believe it or not, but I got my first computer in July 1999. A Philips 8088 PC running DOS 5. :-) Bill -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
[Soekris] Mailing list receive order
Hello, by the way, is there some special reason why I receive postings from this list in a random order and with variable and sometimes considerable delays (days)? Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Net4801 BIOS System Parameters Documentation
Hi Lars, take a look at the comBIOS changelog on the downloads page at soekris.com: http://www.soekris.com/software/changelog.txt It should contain basic but usable info about new comBIOS features integrated after the manuals were published. Bill On Fri, 2008-08-22 at 10:32 +0300, Lars Noodén wrote: I'm looking at a Net4801 and see that there are several BIOS system parameters not listed in the manual [1] section 2.5 BIOS System Parameters The seem not to be covered elsewhere in the manual. Specifically, ConMute FastBoot BootPartition BootDrive Reset is also missing from § 2.5, but described in § 3.8. What do the other parameters above do and where are they described? -Lars [1] http://www.soekris.com/manuals/net4801_manual.pdf ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] NET4801 - ata0: FAILURE - non aligned DMA transfer attempted
Hi Thomas, On Sun, 2008-08-10 at 22:39 +0200, Thomas Elsgaard wrote: Hi Guys I am trying to install FreeBSD 6.2 via PXE boot on a 4801, and the PXE boot goes well, but when i am starting to install the OS, i am getting following error from the installer: Unable to find device node for /dev/ad0s1b in /dev And following is printed to screen: ata0: FAILURE - non aligned DMA transfer attempted ad0: setting up DMA failed ata0: FAILURE - non aligned DMA transfer attempted ad0: setting up DMA failed ata0: FAILURE - non aligned DMA transfer attempted ad0: setting up DMA failed The disk is an Hitachi Travelstar, model: HTE726060M9AT00 I know for the sure that the disk is working OK So what can be wrong? http://lists.soekris.com/pipermail/soekris-tech/2005-December/009733.html http://lists.soekris.com/pipermail/soekris-tech/2007-July/012532.html Bill ///Thomas ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] pxe boot problems 4501 to FreeBSD 6.2 box
Hello tokind, On Wed, 2008-08-06 at 16:23 -0700, tokind wrote: I tried this and got no further. I decided to focus on the 0.0.0.0 business and added a option next-server 10.222.0.1; to my dhcpd conf. This did the trick. Apparently some ROM boot clients resort to 0.0.0.0 and connect to nothing if the next-server option is not used. This has been reported (on this list) with OpenBSD as well, which uses the same PXE/DHCP client implementation I believe (I think OpenBSD borrowed the FreeBSD software in this case). Not quite the first thing that comes to mind in these cases, but looking at your original posting again, the 0.0.0.0's are pretty obvious. This is not the behaviour for next-server prescribed by the PXE standard - it should default to the gateway IP if not specified. Someone should really report this as a bug some day.. Bill So, I am much closer. Now I see the Boot Installer option screen and get on down the road to Trying to mount root from nfs: NFS ROOT: 10.222.0.1:/cdrom/boot sis0: link state changed to UP and that is where it grinds to a halt. I did the gzip -d on the tftpboot/boot/boot/mfsroot.gz, which is the only place I could find the mfsroot. I also corrected my path accordingly. Here is the thing - I am booting to the CD. I cannot unzip the mfsroot on the CD. jkv wrote: I actually had the very same problem last week, as far as i remember you have to gzip -d boot/mfsroot on the machine who are servering files(NFS or TFTP) to your soekris during install. Regards, jkv ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] pxe boot problems 4501 to FreeBSD 6.2 box
Hi tokind, On Fri, 2008-08-01 at 13:43 -0700, tokind wrote: I built a FreeBSD 6.2 server according to the instructions in Michael R. Brumm's article @ http://www.michaelbrumm.com/how-to-install-freebsd-pxe.html I am connecting the net4501 directly to the nic on the server, so there is no router involved. I set the router address same as the host address. The only thing different from the instructions was in the section on Configure the TFTP Directory section: There is no /cdrom/kernel directory to copy. I used /cdrom/boot/kernel instead. It was the only kernel directory I could find. When I attempt to boot from the net4501 console I get a Client IP 10.22.0.100, and a DHCP IP 10.222.0.1 (which is the IP of my server). Then I get many attempts to pxe boot like so: FreeBSD/i386 bootstrap loader, Revision 1.1 [EMAIL PROTECTED], Day, Date, etc. pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 start not found pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 | pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 pxe_open: server addr: 0.0.0.0 pxe_open: server path: /cdrom pxe_open: gateway ip: 10.222.0.1 can't load 'kernal' ^ ? Typo in the boot config file? Bill I've gone over the NFS and RPC configuration and the build instructions several times. Not sure what to look at next. -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Hardware recommendations
Hi Mario, you might consider the modular approach: Soekris net for routing/firewall and VPN, and one or two bigger boxes for the rest. Maintaining two or more boxes in different roles tends to be a lot less stressful than having to maintain a single one in a multi-role config. You'll have to dig deep for any benchmarks, but in a gateway role a Soekris net5501 should do pretty well in most cases. Bill On Fri, 2008-08-01 at 13:50 +0200, [EMAIL PROTECTED] wrote: Hello List, we are currently looking for some powerfull Hardware. It must have: - Intel Pentium Mobile (= 1,6GHz) - 2,5 Hard Disk (Slot) - Min 1GB Ram - Min 3 Interfaces - VGA is optional I think i have seen such a 19 Soekris Box on the Linux days 2007. Can anyone point me to some products?! We primarily need it for Proxy, Mailserver, VPN and Netfilter. Thanks, Mario ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] NET5530 - Information about Led
On Fri, 2008-07-25 at 07:48 +, Poul-Henning Kamp wrote: In message [EMAIL PROTECTED], Massimiliano Cianelli writes : I'v recently bought a Net5530, I'm trying to find how to command the 3 led on the front of panel. I pressume you mean a 5501 in a case :-) To be more precise, I want to use the 3 led for indicate the status, eg: Power led - Blink - System is online and is working (blink made via a script that do echo 1\0 /proc/blabla) The power-led is not controllable, it is directly wired to the power. Error Led - hum.. Internet connection down (ok it is just an example :P) This one is hooked up to GPIO pin 6. Disk Led - Due that it don't look at the CF R\W, use it for.. an example, show the traffic of the VPN interface. This is wired directly to the disk activity hardware pin. Is possibile to make that? In sort: No. Of course you could rip the LEDs off the board and stick your own ones into the holes on the front of the case, if you don't think you need tham. That's my uneducated guess, at least (I wouldn't take going into discussion with Poul-Henning lightly on this topic;) . But then you might as well drill a couple of new holes.. Bill -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] COM2 Problem with OpenBSD4.3 on NET4501
On Wed, 2008-07-23 at 10:04 +, Stuart Henderson wrote: On 2008-07-23, Bill Maas [EMAIL PROTECTED] wrote: RTS/CTS is enabled by default, at least on a net5501 (I had trouble with an oddly wired cable a while ago). afaik, that would only affect the boot loader. for OpenBSD, use a boot loader dated after 2008/04/20, it was rewritten to access the hardware directly rather than via BIOS. It would also affect a program that wants to talk with the comBIOS, e.g. for selecting a boot device with no direct interaction. The issue would still be with the cable rather than with the software. The cable that gave me trouble with the net5501 at boot always worked with OpenBSD from the boot loader and onwards, even with older OpenBSD versions. The boot loader and /bsd apparently turned off RTS/CTS. So the comBIOS ssems to provide adequate support for UART configuration, it is just not visible to the user, except for the baud rate. In my case a ConHandshakeDisable option would have been useful, it would at least have lead me immediately to the source of the trouble. Maybe something for b5501_134.bin? Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] NET5501 Boot Problem
Hi Eternity, On Fri, 2008-07-18 at 00:49 +0200, Eternity wrote: Hi! Yes, I thought of that also, but I have used the same image on several net5501 that is working just fine, so why not these? Also, if it was a disk geometry problem - why would it work after a warm reboot, but not a cold one? For me, it seems like either some problem with the net5501 itself or that this series of flashcards are causing something to happen. You're right, I think that the GRUB error code for disk geometry trouble (which it sees as a corrupted fs) is a different one. It's reported on several occasions in the archives. Maybe LILO or syslinux can solve this, but I would like to find out why this happens. I am not so glad in having several images to deal with when preparing the boxes.. With those at least you never have to worry about DG trouble. Anyway, I'm biased on this issue, I didn't like the way in which the otherwise excellent Lilo was obsoleted in some GNU/Linux realms. I would use Lilo in all cases, if only because I can't remember ever having seen a posting about Lilo unable to boot on this list. GRUB on the other hand is a regular visitor..;) Bill Chris On Thu, 17 Jul 2008, Bill Maas wrote: Hi Eternity, In addition to everything proposed so far: take a look at disk geometry issues on the Soekris Wiki. Those may also cause problems with GRUB. If that is the problem, just boot with Lilo or Syslinux (which is the smallest boot loader for Linux I think, an at least the easiest to configure). Bill On Thu, 2008-07-17 at 17:35 +0200, Eternity wrote: Hi! I have a problem with a whole bunch of NET5501:s. I have read some posts with similar problems, but not exactly this, I think, so I post my problem: On a Cold boot my CF Card is detected: Pri Mas CF CARD 1GB LBA Xlt 983-32-63 990 Mbyte Then at boot: GRUB Loading stage1.5. GRUB loading, please wait... Error 25 --- If I press ctrl-p, and at once in combios makes a reboot everything works OK, and GRUB can load correctly. I have used lots of net5501 boxes with this distribution (same image file I write to the CF card) without any trouble at all. I have tried upgrading BIOS to 1.33c version, but no change. Any ideas? Could it be the flash card? Or some defect on these batch of boards? Regards, Chris ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] NET5501 Boot Problem
Hi Eternity, In addition to everything proposed so far: take a look at disk geometry issues on the Soekris Wiki. Those may also cause problems with GRUB. If that is the problem, just boot with Lilo or Syslinux (which is the smallest boot loader for Linux I think, an at least the easiest to configure). Bill On Thu, 2008-07-17 at 17:35 +0200, Eternity wrote: Hi! I have a problem with a whole bunch of NET5501:s. I have read some posts with similar problems, but not exactly this, I think, so I post my problem: On a Cold boot my CF Card is detected: Pri Mas CF CARD 1GB LBA Xlt 983-32-63 990 Mbyte Then at boot: GRUB Loading stage1.5. GRUB loading, please wait... Error 25 --- If I press ctrl-p, and at once in combios makes a reboot everything works OK, and GRUB can load correctly. I have used lots of net5501 boxes with this distribution (same image file I write to the CF card) without any trouble at all. I have tried upgrading BIOS to 1.33c version, but no change. Any ideas? Could it be the flash card? Or some defect on these batch of boards? Regards, Chris ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- There is nothing to worry about - unknown ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Squashed versions of FreeBSD
Hi Crist, On Tue, 2008-07-08 at 11:41 -0700, Crist J. Clark wrote: On Fri, Jul 04, 2008 at 11:28:15AM +0200, Arnt Gulbrandsen wrote: Crist J. Clark writes: My only real worry is that I've not broken out all of the stuff from the /var that I should and I may get dead spots on my CF from too Set up an experimental box, mount everything except /var R/O, leave it running for a few days and see what happens? A properly set up system (e.g. OpenBSD) would not give any errors, but yes, there are always a few hardheaded programs that want to write to /usr/progname/etc and the like... many write-cycles earlier than I need to, but at $40, it's a risk I can take. I'd be curious about any advice on that front. $40 isn't much. The sudden and unexpected downtime if/when the box breaks may be a great deal worse. Depends how simple and cheap it will be to get it back up. I may be wrong, but I didn't think the failure mode would be a sudden and unexpected downtime. I had assumed I'd start to see read/write errors like you see when a HDD slowly gets bad sectors. That's exactly what I've seen with a TakeMS cheaper-then-cheap CF card with no wear-leveling, going bad. It took about 6 weeks going from 1 recoverable read error every 2 days to a torrent of error messages ( 100/24 hrs), so keeping an eye on the logs should do. Of course with a strong YMMV reservation, depending on the card you are using. Bill However, rethinking that, I'm not sure what to expect. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
[Soekris] MFS filling up on net4521
Hi, I'm not very familiar with memory issues let alone OpenBSD's MFS memory allocation/release algorithms, but the following strikes me as being a bit odd: net5501: Filesystem 1K-blocks Used Avail Capacity Mounted on /dev/wd0a 98348628666264765031%/ mfs:265097855 2913 455039%/var net4801: Filesystem 1K-blocks Used Avail Capacity Mounted on /dev/wd0a 98348628700264731031%/ mfs:184707855 3271 419244%/var net4521: Filesystem 1K-blocks Used Avail Capacity Mounted on /dev/wd0a 496894424180 4787090%/ mfs:221037855 5485 197873%/var All boxes have an MFS partition mounted onto /var. Amount of RAM varies from (by heart - I'm not sitting next to these boxes right now) 128MB (net4521) to 512MB (net 5501). The 4521 has the least amount of memory, that's for sure. What happens is that the disk usage on the MFS partition stabilizes at around 40% when I use a log rotating script. But with the same script, the MU keeps growing on the 4521 towards 100%. A significant part of it does not seem to be accounted for as actual disk usage: df -h: Filesystem 1K-blocks Used Avail Capacity Mounted on mfs:22103 7.7M 5.4M 1.9M74%/var du -sh /var: 1.8M/var/ After a reboot it's back to a sane value (~20M). Leaks? Allocation issues? As said, the net4521 is lowest on RAM. Or am I staring the Stupid and Obvious right in the eye here? Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] cloning CF cards on OpenBSD 4.3
Hi Lars, for a 1-time clone to an identical CF dd will probably do, just check that it is indeed bootable after the dd, or run installboot(8) afterwards. For periodical cloning (i.e. mirroring) I'd use rsync, which is very conservative when it comes to writes (no, let's not start that CF w-cycle discussion again - it may also save lots of time;). In that case the second CF needs to have an fs and to be mounted, and there may be issues with /var (not necessarily - PID files in /var/run are removed at boot time, for example). You'll need to test and possibly refine the procedure (rsync supports pattern-based file exclusion, for example). In my experience, except possibly for the trailing slash quirk (make sure that your informed about that one, see rsync(1):USAGE), it's the best tool available for jobs like this. Hope that this message contains at least a few things you didn't already know, Bill On Sun, 2008-05-11 at 16:46 +0300, Lars Noodén wrote: I've got an OpenBSD 4.3 system on 1GB CF that I would like to clone to another CF. What are some good/easy methods? dd? Regards, -Lars ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] BIOS 1.33 and boot FreeBSD issues (was net5501-60 CF not detected on cold boot)
Hi Igor, On Mon, 2008-03-10 at 20:51 +0100, Igor Sobrado wrote: [...] 3. Mhz should be really MHz, this typo hurts someone with a background in fundamental physics. ;-) 4. May I suggest renaming CpuSpeed to CPUspeed? (just as in BIOSentry); CPU is an acronym. This is a matter of convention. Preserving the uppercase characters in acronyms when MixedCase names are used, generally leads to more inconsistency and confusion than it solves.. In some enviroments it might even have to be cpuSpeed, depending on how it was used. If done consistently, squashing acronyms this way tends to provide the cleanest overall presentation (and being consistent at least also happens to enhance parser-friendliness). BIOSentry seems to be the offending one here (but I haven't looked at where or how it is used, so I might be wrong). Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Is it my imagination...
Hi nicodache, On Wed, 2008-02-27 at 21:17 +0100, nicodache wrote: I wonder what all that changes for Wim, who was kind of the official reseller for soekris in EU, or at least in be, nl and fr :) I guess no-one can forbid you or anyone else to keep ordering from www.kd85.com. And after all, the fact that they send _paper_ manuals with shipped boards/boxes (if available - at least they feel strongly about the need for such paraphernalia) indicates a level of service that isn't necessarily met elsewhere;). Bill Wim, what do you think of all that ? no problem for you business and suppose hope :) Cheers, On Wed, Feb 27, 2008 at 9:06 PM, Torben Nielsen [EMAIL PROTECTED] wrote: Hi Various observations on the new sites: * The engine for the new us and eu webshops seems a lot like the one cortexsystems.dk has used for a while. *The soekris.eu shop company has the same postal address as cortexsystems.dk *The 5501-70 is 40,75 euros (incl. vat) cheaper on soekris.eu than on cortexsystems.dk. Regards, Torben ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4801 problem pxeboot'ing OpenBSD 4.2 from FreeBSD 6.3
Hi Adam, On Wed, 2008-02-20 at 23:08 +, Adam Retter wrote: [...] CLIENT MAC ADDR: 00 00 24 C4 2C 6C CLIENT IP: 172.16.16.252 MASK: 255.255.255.240 DHCP IP: 172.16.16.245 GATEWAY IP: 172.16.16.254 172.16.16.245 172.16.16.254 Maybe the first IP is what you meant, but it might as well indicate a typo in the next-server field in dhcpd.conf(!). probing: pc0 com0 com1 pci pxe![2.1] mem[639K 127M a20=on] disk: hd0+ net: mac 00:00:24:c4:2c:6c, ip 172.16.16.252, server 0.0.0.0 I would expect a local IP here for server. This would match next-server in dhcpd.conf on OpenBSD. Apparantly downloading the boot loader succeeds (as if next-server doesn't apply yet), but the boot loader apparently fails to get the TFTP server's correct address. I'm not sure about this, I didn't write dhcpd or pxeboot;), but it looks to me as if that's what's to be happening here. Bill OpenBSD/i386 PXEBOOT 2.02 booting tftp:bsd.rd: | booting tftp:bsd.rd: 4733076read text: Unknown error: code 60 failed(60). will try /bsd boot booting tftp:/bsd: open tftp:/bsd: No such file or directory failed(2). will try /bsd Turning timeout off. boot ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris PXE boot
Hi Voipcrazy, http://www.modemsite.com/56k/x2-hyperterm.asp Search characters twice. It's called a manual;). Bill On Wed, 2008-02-20 at 17:12 +0100, voip crazy wrote: Thanks for your answers, but my problem persists, I think PXE boot works on my soekris, look thos tftpd log: Feb 20 16:51:17 localhost dhcpd: DHCPDISCOVER from 00:00:24:c9:5a:14 via eth1 Feb 20 16:51:17 localhost dhcpd: DHCPOFFER on 192.168.30.63 to 00:00:24:c9:5a:14 via eth1 Feb 20 16:51:18 localhost dhcpd: DHCPREQUEST for 192.168.30.63 (192.168.30.253) from 00:00:24:c9:5a:14 via eth1 Feb 20 16:51:18 localhost dhcpd: DHCPACK on 192.168.30.63 to 00:00:24:c9:5a:14 via eth1 Feb 20 16:51:18 localhost in.tftpd[3379]: RRQ from 192.168.30.63 filename pxelinux.0 Feb 20 16:51:18 localhost in.tftpd[3379]: tftp: client does not accept options Feb 20 16:51:18 localhost in.tftpd[3380]: RRQ from 192.168.30.63 filename pxelinux.0 Feb 20 16:51:25 localhost in.tftpd[3381]: RRQ from 192.168.30.63 filename pxelinux.cfg/01-00-00-24-c9-5a-14 Feb 20 16:51:25 localhost in.tftpd[3381]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:25 localhost in.tftpd[3382]: RRQ from 192.168.30.63 filename pxelinux.cfg/C0A81E3F Feb 20 16:51:25 localhost in.tftpd[3382]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:26 localhost in.tftpd[3383]: RRQ from 192.168.30.63 filename pxelinux.cfg/C0A81E3 Feb 20 16:51:26 localhost in.tftpd[3383]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:26 localhost in.tftpd[3384]: RRQ from 192.168.30.63 filename pxelinux.cfg/C0A81E Feb 20 16:51:26 localhost in.tftpd[3384]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:27 localhost in.tftpd[3385]: RRQ from 192.168.30.63 filename pxelinux.cfg/C0A81 Feb 20 16:51:27 localhost in.tftpd[3385]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:28 localhost in.tftpd[3386]: RRQ from 192.168.30.63 filename pxelinux.cfg/C0A8 Feb 20 16:51:28 localhost in.tftpd[3386]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:28 localhost in.tftpd[3387]: RRQ from 192.168.30.63 filename pxelinux.cfg/C0A Feb 20 16:51:28 localhost in.tftpd[3387]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:29 localhost in.tftpd[3388]: RRQ from 192.168.30.63 filename pxelinux.cfg/C0 Feb 20 16:51:29 localhost in.tftpd[3388]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:29 localhost in.tftpd[3389]: RRQ from 192.168.30.63 filename pxelinux.cfg/C Feb 20 16:51:29 localhost in.tftpd[3389]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:30 localhost in.tftpd[3390]: RRQ from 192.168.30.63 filename pxelinux.cfg/default Feb 20 16:51:48 localhost in.tftpd[3676]: RRQ from 192.168.30.63 filename debian-installer/i386/boot-screens/boot.txt Feb 20 16:51:58 localhost in.tftpd[3692]: RRQ from 192.168.30.63 filename linux Feb 20 16:51:58 localhost in.tftpd[3692]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:58 localhost in.tftpd[3693]: RRQ from 192.168.30.63 filename linux.cbt Feb 20 16:51:58 localhost in.tftpd[3693]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:58 localhost in.tftpd[3694]: RRQ from 192.168.30.63 filename linux.0 Feb 20 16:51:58 localhost in.tftpd[3694]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:58 localhost in.tftpd[3695]: RRQ from 192.168.30.63 filename linux.com Feb 20 16:51:58 localhost in.tftpd[3695]: sending NAK (1, File not found) to 192.168.30.63 Feb 20 16:51:58 localhost in.tftpd[3696]: RRQ from 192.168.30.63 filename linux.c32 Feb 20 16:51:58 localhost in.tftpd[3696]: sending NAK (1, File not found) to 192.168.30.63 But what I get on my soekris console is that CCoouulldd nnoo tt ffiinndd kk eerrnneell iimm aaggee:: lliinn uuxx bbtt:: I type linux as kernel image and it does not boot The caracters are writing twice and it cannot find the kernel image version My TFTP_Root/pxelinux.cfg/default have a line like this SERIAL 0,19200 for console connection And the default image to boot is configured like that in the same file kernel linux append vga=normal initrd=initrd.gz console=ttyS0,19200n8 ramdisk_size=14332 root=/dev/rd/0 rw -- Why is this happening? How sould I adjust my Hyperteminal? Thanks Best Regards. VoipCrazy ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris PXE boot
Hi Brian, On Thu, 2008-02-21 at 13:43 +, Brian Candler wrote: On Thu, Feb 21, 2008 at 10:50:00AM +0100, Bill Maas wrote: Hi Voipcrazy, http://www.modemsite.com/56k/x2-hyperterm.asp Search characters twice. It's called a manual;). It's nothing to do with hyperterminal or half-duplex settings. I'm very sorry.. But what I get on my soekris console is that CCoouulldd nnoo tt ffiinndd kk eerrnneell iimm aaggee:: lliinn uuxx bbtt:: (1) The character duplication is because pxelinux is outputting both to the COM port and the VGA BIOS. The Soekris redirects the VGA BIOS characters to the COM port. Hence you get them twice. (2) The 16-column line wrap is a bug in the Soekris VGA BIOS emulation. You can patch and rebuild pxelinux.0 to avoid this, by stopping pxelinux writing to the VGA BIOS. Details at the end of http://archive.netbsd.se/?ml=soekris-techa=2007-11m=5685776 and the patch, with instructions on how to rebuild pxelinux.0 using nasm, is higher up. If I remember rightly, this patch will also prevent the character duplication. Just for reference, here's a working pxelinux.cfg/default: --- serial 0 19200 0x303 DEFAULT linux label linux kernel openwrt-x86-2.6-vmlinuz append initrd=openwrt-x86-2.6-rootfs.cpio.gz init=/etc/preinit console=tty0 console=ttyS0,19200n8 --- Might do the trick.. Bill And here's a working grub/menu.lst (with settings to allow the unit to boot even when no serial cable is connected) serial --unit=0 --speed=19200 --word=8 --parity=no --stop=1 terminal --timeout=0 --dumb serial hiddenmenu default 0 timeout 0 title OpenWrt root(hd0,0) kernel /boot/vmlinuz init=/etc/preinit console=tty0 console=ttyS0,19200n8 reboot=bios initrd /boot/rootfs.cpio.gz boot However, fixing the character output problems is mostly cosmetic. It won't solve the problem of the missing kernel. B. -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4801 problem pxeboot'ing OpenBSD 4.2 from FreeBSD 6.3
Hi Adam, [apologies for off-list reply, Reply To All is only a Shift away here] On Thu, 2008-02-21 at 10:51 +, Adam Retter wrote: On Thu, Feb 21, 2008 at 9:28 AM, Bill Maas [EMAIL PROTECTED] wrote: Hi Adam, On Wed, 2008-02-20 at 23:08 +, Adam Retter wrote: [...] CLIENT MAC ADDR: 00 00 24 C4 2C 6C CLIENT IP: 172.16.16.252 MASK: 255.255.255.240 DHCP IP: 172.16.16.245 GATEWAY IP: 172.16.16.254 172.16.16.245 172.16.16.254 No this is correct, my router (imaginary, as it will be my net4801 eventually) is on 172.16.16.254 and the FreeBSD machine issuing the DHCP lease is on 172.16.16.245 Maybe the first IP is what you meant, but it might as well indicate a typo in the next-server field in dhcpd.conf(!). not sure i have a next-server field in my dhcpd.conf, I am at work at the moment and will check this when I get home... probing: pc0 com0 com1 pci pxe![2.1] mem[639K 127M a20=on] disk: hd0+ net: mac 00:00:24:c4:2c:6c, ip 172.16.16.252, server 0.0.0.0 I would expect a local IP here for server. This would match next-server in dhcpd.conf on OpenBSD. Apparantly downloading the boot loader succeeds (as if next-server doesn't apply yet), but the boot loader apparently fails to get the TFTP server's correct address. Ah so next-server could definitely be the culprit if I am missing it, will take a look tonight. Thanks very much for your help so far :-) No, I was being overly imaginative and suspicious about default program behaviour. The two IPs just looked suspiciously similar to me. The messages for trivial errors are in fact quite clear (just generated a few): Wrong next-server (PXE ROM): PXE-E11: ARP timeout PXE-E11: ARP timeout PXE-E38: TFTP cannot open connection PXE-M0F: Exiting BootManage PXE ROM. Unreadable bsd.rd (pxeboot): booting tftp:/bsd.rd: open tftp:/bsd.rd: Operation not permitted failed(1). will try /bsd boot booting tftp:/bsd: open tftp:/bsd: No such file or directory failed(2). will try /bsd Turning timeout off. Without a next-server the DHCP IP will be used as it should, but in my output I do see it after the server keyword, and not 0.0.0.0 (which might indicate ARP trouble, and many error code 60 reports are about ARP). But the ARP/next-server theory doesn't hold as well, since pxeboot can (a) determine bsd.rd's size, and (b) determine that bsd isn't there. It doesn't look like there is a trivial answer here. Bill I'm not sure about this, I didn't write dhcpd or pxeboot;), but it looks to me as if that's what's to be happening here. Bill OpenBSD/i386 PXEBOOT 2.02 booting tftp:bsd.rd: | booting tftp:bsd.rd: 4733076read text: Unknown error: code 60 failed(60). will try /bsd boot booting tftp:/bsd: open tftp:/bsd: No such file or directory failed(2). will try /bsd Turning timeout off. boot ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4801 problem pxeboot'ing OpenBSD 4.2 from FreeBSD 6.3
Hi Adam, From http://www.onlamp.com/pub/a/bsd/2003/06/05/FreeBSD_Basics.html: You'll note that FreeBSD supports both IPv4 and IPv6, so its inetd is capable of listening for both types of requests. Also note that TFTP uses UDP as its transport. This means it is not as reliable as FTP (which uses TCP). It also means that TFTP supports broadcasts, meaning you don't have to configure the TFTP client with the IP address of a particular TFTP server. From http://www.javvin.com/protocolTFTP.html: Each data packet contains one block of data, and must be acknowledged by an acknowledgment packet before the next packet can be sent. A data packet of less than 512 bytes signals termination of a transfer. If a packet gets lost in the network, the intended recipient will timeout [...] Together with server 0.0.0.0 this might hold a clue. Looks like requests are getting through, but ACKs aren't. Were there changes on the FreeBSD (or OpenBSD) front since you last succesfully did this? Bill On Thu, 2008-02-21 at 10:51 +, Adam Retter wrote: On Thu, Feb 21, 2008 at 9:28 AM, Bill Maas [EMAIL PROTECTED] wrote: Hi Adam, On Wed, 2008-02-20 at 23:08 +, Adam Retter wrote: [...] CLIENT MAC ADDR: 00 00 24 C4 2C 6C CLIENT IP: 172.16.16.252 MASK: 255.255.255.240 DHCP IP: 172.16.16.245 GATEWAY IP: 172.16.16.254 172.16.16.245 172.16.16.254 No this is correct, my router (imaginary, as it will be my net4801 eventually) is on 172.16.16.254 and the FreeBSD machine issuing the DHCP lease is on 172.16.16.245 Maybe the first IP is what you meant, but it might as well indicate a typo in the next-server field in dhcpd.conf(!). not sure i have a next-server field in my dhcpd.conf, I am at work at the moment and will check this when I get home... probing: pc0 com0 com1 pci pxe![2.1] mem[639K 127M a20=on] disk: hd0+ net: mac 00:00:24:c4:2c:6c, ip 172.16.16.252, server 0.0.0.0 I would expect a local IP here for server. This would match next-server in dhcpd.conf on OpenBSD. Apparantly downloading the boot loader succeeds (as if next-server doesn't apply yet), but the boot loader apparently fails to get the TFTP server's correct address. Ah so next-server could definitely be the culprit if I am missing it, will take a look tonight. Thanks very much for your help so far :-) I'm not sure about this, I didn't write dhcpd or pxeboot;), but it looks to me as if that's what's to be happening here. Bill OpenBSD/i386 PXEBOOT 2.02 booting tftp:bsd.rd: | booting tftp:bsd.rd: 4733076read text: Unknown error: code 60 failed(60). will try /bsd boot booting tftp:/bsd: open tftp:/bsd: No such file or directory failed(2). will try /bsd Turning timeout off. boot ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net4801 problem pxeboot'ing OpenBSD 4.2 from FreeBSD 6.3
On Thu, 2008-02-21 at 19:22 +, Adam Retter wrote: Bill, Thanks very very much :-) All it needed was the next-server option, then it booted up perfectly - im now sending this via. my new OpenBSD 4.2 net4801 router/firewall. I guess something must have changed on either the OpenBSD or FreeBSD side, I think FreeBSD switched to isc-dhcpd-3 and im not sure whether the dhcpd of pxeboot in OpenBSD has changed, but I know that I didnt used to have to use next-server to get it to work. Thanks again. Im overjoyed :-) Glad to hear that. Your output might still be of interest to FreeBSD's dhcpd3 and/or OpenBSD's pxeboot maintainers, since according to dhcpd(8) V3.0.4 on Ubuntu Linux: If no next-server parameter applies to a given client, the DHCP server’s IP address is used. Which obviously didn't entirely go according to plan here.. The bug is most likely in pxeboot, as both dhcpd and tftpd appear to be doing what they should do and pxeboot's all-zeroes looked suspicious to begin with. Bill On Thu, Feb 21, 2008 at 3:00 PM, Bill Maas [EMAIL PROTECTED] wrote: Hi Adam, [apologies for off-list reply, Reply To All is only a Shift away here] On Thu, 2008-02-21 at 10:51 +, Adam Retter wrote: On Thu, Feb 21, 2008 at 9:28 AM, Bill Maas [EMAIL PROTECTED] wrote: Hi Adam, On Wed, 2008-02-20 at 23:08 +, Adam Retter wrote: [...] CLIENT MAC ADDR: 00 00 24 C4 2C 6C CLIENT IP: 172.16.16.252 MASK: 255.255.255.240 DHCP IP: 172.16.16.245 GATEWAY IP: 172.16.16.254 172.16.16.245 172.16.16.254 No this is correct, my router (imaginary, as it will be my net4801 eventually) is on 172.16.16.254 and the FreeBSD machine issuing the DHCP lease is on 172.16.16.245 Maybe the first IP is what you meant, but it might as well indicate a typo in the next-server field in dhcpd.conf(!). not sure i have a next-server field in my dhcpd.conf, I am at work at the moment and will check this when I get home... probing: pc0 com0 com1 pci pxe![2.1] mem[639K 127M a20=on] disk: hd0+ net: mac 00:00:24:c4:2c:6c, ip 172.16.16.252, server 0.0.0.0 I would expect a local IP here for server. This would match next-server in dhcpd.conf on OpenBSD. Apparantly downloading the boot loader succeeds (as if next-server doesn't apply yet), but the boot loader apparently fails to get the TFTP server's correct address. Ah so next-server could definitely be the culprit if I am missing it, will take a look tonight. Thanks very much for your help so far :-) No, I was being overly imaginative and suspicious about default program behaviour. The two IPs just looked suspiciously similar to me. The messages for trivial errors are in fact quite clear (just generated a few): Wrong next-server (PXE ROM): PXE-E11: ARP timeout PXE-E11: ARP timeout PXE-E38: TFTP cannot open connection PXE-M0F: Exiting BootManage PXE ROM. Unreadable bsd.rd (pxeboot): booting tftp:/bsd.rd: open tftp:/bsd.rd: Operation not permitted failed(1). will try /bsd boot booting tftp:/bsd: open tftp:/bsd: No such file or directory failed(2). will try /bsd Turning timeout off. Without a next-server the DHCP IP will be used as it should, but in my output I do see it after the server keyword, and not 0.0.0.0 (which might indicate ARP trouble, and many error code 60 reports are about ARP). But the ARP/next-server theory doesn't hold as well, since pxeboot can (a) determine bsd.rd's size, and (b) determine that bsd isn't there. It doesn't look like there is a trivial answer here. Bill I'm not sure about this, I didn't write dhcpd or pxeboot;), but it looks to me as if that's what's to be happening here. Bill OpenBSD/i386 PXEBOOT 2.02 booting tftp:bsd.rd: | booting tftp:bsd.rd: 4733076read text: Unknown error: code 60 failed(60). will try /bsd boot booting tftp:/bsd: open tftp:/bsd: No such file or directory failed(2). will try /bsd Turning timeout off. boot ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501 interface information updated on the Wiki
Hi, since my first reply to Stephen apparently didn't make it to this list, here's Take 2: PDF version is here: http://stsx.xs4all.nl/Net5501BoardConnections-1.pdf (for future archive diggers: may vanish any time) Looks very neat. Bill On Fri, 2008-02-15 at 16:43 +, Stephen Patrick wrote: Dear all I took the liberty to take Ralph's content and update it. Maybe I have later boards here, there are a lot of Pin1 idents both top and bottom so I've put those in. I put an annotated photo of the board in the document, and some content from the 4801 manual so the info is in one place. Maybe I should have/could put this on the Wiki but for now it's up on personal googlepages as a Word doc. http://stephenjpatrick.googlepages.com/Net5501BoardConnections.doc If anyone wants to put content from this on the Wiki, or use as a base for an official manual, then welcome, and I'd appreciate knowing what's been done. Comments/additions/corrections welcome. Regards Stephen Patrick www.cablefreesolutions.com [ps. I sent this as text-only format from outlook. If it ends up on the list as HTML it was our company exchange server at fault, and apologies in advance for that] -Original Message- From: Ralph Becker-Szendy [mailto:[EMAIL PROTECTED] Sent: 14 February 2008 22:13 To: soekris-tech@lists.soekris.com Subject: [Soekris] 5501 interface information updated on the Wiki Given all the whining in the last few days (and months and years!) about missing documentation from Soren, I updated the unofficial Soekris Wiki with a page about interfaces on the 5501 board: http://wiki.soekris.info/5501_Interfaces This was just a quick effort to get started; I documented everything I could remember off-hand, and from the discussion in the last few days. The next time I have my 5501 apart, I will update the documentation much further. Others are free and welcome to contribute too. At least, it documents the disk power and USB connectors. -- Ralph Becker-Szendy[EMAIL PROTECTED] (408)395-1435 735 Sunset Ridge Road; Los Gatos, CA 95033 ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- This email has been verified as Virus free. Virus Protection and more available at http://www.plus.net -- No virus found in this incoming message. Checked by AVG Free Edition. Version: 7.5.516 / Virus Database: 269.20.5/1279 - Release Date: 14/02/2008 18:35 ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris PXE boot
Hi Voipcrazy, On Mon, 2008-02-18 at 18:05 +0100, voip crazy wrote: Dear list, Five days ago, I started to try boot my soekris, vía PXE, without success, my soekris get an IP address and It stop to boot. I do not know why It is doing that, I search google, and find some links but all of them fails to boot. I want to install Debian on it an my soekris is 5501 model. http://roland.entierement.nu/pages/debian-on-soekris-howto.html Someone, could send me the TFTP files you use to boot a debian instalation? Or explain me where to get the Debian netboot.tar.gz, to boot the soekris 5501? http://ftp.debian.org/debian/dists/etch/main/installer-i386/current/images/netboot/ Select the right distrib dir of course.. You'll also need the matching pxelinux.0 sitting next to it. I've run it succesfully as follows: # mkdir /tftpboot # tar xzvpf netboot.tar.gz # mv debian-installer/i386/* /tftpboot I run tftpd-hpa as a standalone daemon: -- #/etc/default/tftpd-hpa #Defaults for tftpd-hpa RUN_DAEMON=yes #OPTIONS=-l -s /var/lib/tftpboot OPTIONS=-l -s /tftpboot -- You may have to add an entry to /etc/hosts.allow: -- # /etc/hosts.allow in.tftpd: LOCAL 192.168.1. : allow -- Or something similar... I never managed to get it work through inetd with Ubuntu/Debian GNU/Linux. Of course no-one will stop you from running it from /var/lib/tftpboot, but I don't like having persistent files under /var (while many others abhor having non-FSSTND stuff inside / - whatever..). My PXE config looks like this: -- # /tftpboot/pxelinux.cfg/default SERIAL 0 19200 0 DISPLAY boot.txt DEFAULT linux LABEL linux kernel linux append vga=normal initrd=initrd.gz console=ttyS0,19200n8 ramdisk_size=14332 root=/dev/rd/0 rw -- PROMPT 1 TIMEOUT 0 -- Common trouble: - missing hosts.allow entry (I'm not sure if a standalone tftpd actually requires one, but it's always safe) - there is any firewall running on the TFTP server blocking TFTP - wrong next-server in dhcpd.conf - files under /tftpboot aren't world-readable (0644 root.root) (on Debian, 'tftpd' runs as user nobody I think). That should get you going. Bill Thanks in advance. Best Regards. VoipCrazy. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris PXE boot
Errata/Supplement: On Tue, 2008-02-19 at 08:27 +0100, Bill Maas wrote: [...] http://ftp.debian.org/debian/dists/etch/main/installer-i386/current/images/netboot/ Select the right distrib dir of course.. You'll also need the matching pxelinux.0 sitting next to it. Not: it's in the tarball. [...] Common trouble: - missing hosts.allow entry (I'm not sure if a standalone tftpd actually requires one, but it's always safe) - there is any firewall running on the TFTP server blocking TFTP - wrong next-server in dhcpd.conf - files under /tftpboot aren't world-readable (0644 root.root) (on Debian, 'tftpd' runs as user nobody I think). - symblic links from inside /tftpboot to locations outside it (chroot..) ;) Bill That should get you going. Bill Thanks in advance. Best Regards. VoipCrazy. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501 under Debian can not see harddisk 5K160
On Fri, 2008-02-15 at 09:55 +0100, René Luckow wrote: You could just use a IDE2CF adapter on the PATA thingie? Note the expressions smaller and on-board in my text. To me, that PATA/SATA thing is just unnecessary complexity, weight and cost. I don't use it and I'm not planning to. I would very much appreciate a scaled-down Soekris model with dual CF. And slots in the side for the CF cards, if possible;). Just a small-scale brainstorm. Bill On Fri, Feb 15, 2008 at 9:46 AM, Bill Maas [EMAIL PROTECTED] wrote: Hi, this is what I was referring to in my previous e-mail: http://lists.soekris.com/pipermail/soekris-tech/2007-January/011447.html or: http://readlist.com/lists/lists.soekris.com/soekris-tech/0/1379.html There were lots of CF/HDD trouble posts around this period, so this should serve as a good starting point for searching more info about CF/HDD issues. Speaking of CF/IDE/ATA/SATA: are there any plans for a smaller board with _no_ HDD support but with a second on-board CF adapter? To me, that would be _really_ useful. Bill On Fri, 2008-02-15 at 08:51 +0100, Pontus Pihlgren wrote: I think its a misconception on Otfrieds part. You can use the CF and the Harddrive at the same time. However you cannot use the IDE connector, SATA connector and CF at the same time, it is two out of three if I recall correctly. There was a thread about it, but I can't locate it. Regards, Pontus. MikaelArhelger skrev: How does that make sense? I want to boot linux from CF card and store data on hdd: That makes sense, does it not? Mikael Otfried Geffert-2 wrote: I remember: you can either use one or the other. Either you use the harddisk and do not seethe CF card or vice versa. I think that is a hardware issue - the board is designed that way. Otfried - Original Message - From: Mikael Arhelger [EMAIL PROTECTED] To: soekris-tech@lists.soekris.com Sent: Thursday, February 14, 2008 4:07 PM Subject: [Soekris] 5501 under Debian can not see harddisk 5K160 Got my first 5501 and booting Debian from 2GB Kingston CF card. Can do fidsk -l to see only /dev/hdb but not the harddisk; a Hitachi 5K160 ATA 40GB. A sync command will make the yellow LED brighter for a short time (split second). Any idea? Bios can not see or ... Any pointer appreciated. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Anyone know what JP3 does on net5501?
On Thu, 2008-02-14 at 13:51 -0800, Trevor Talbot wrote: On Feb 14, 2008, at 8:26 AM, Bill Maas wrote: On Thu, 2008-02-14 at 16:07 +0100, Bernd Walter wrote: [...] Yes - having a manual for this board would be nice, but I don't expect that there is anything new in it. It will not tell about unpopulated headers - the 4501/4801 manual don't do it either. Amongst the questions sent to this list are which is the internal USB header, which one is the second COM port and how is the GPIO header wired. Those were documented as Soekris FEATURES in the documentation for previous boxes, not as there are unpopulated headers on the board and if you can find out what they are you can do lots of nice stuff with them. Er, they were populated headers, not unpopulated holes in the board. The thing that started this thread is apparently a set of holes, not a header. It's unreasonable to expect official documentation for it. The thing that started this and a few other threads is there not being a board doc. If you substitute unpopulated headers (as I copied from Bernd's text in my hurry) with various connectors the story remains exactly the same. If there is a board doc, it will at least be clear what is a board feature and what isn't. No-one has asked such questions about net4[58]x boards, as far as I can remember. Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] 5501 under Debian can not see harddisk 5K160
On Thu, 2008-02-14 at 16:58 -0800, MikaelArhelger wrote: How does that make sense? I want to boot linux from CF card and store data on hdd: That makes sense, does it not? If I remember it correctly, this is also a CF issue (M/S issues). You will have to search the archive for more info. Even if it can't be done, you will at least find a few suggestions for doing what you want to do - you're not the first to notice this;). Bill Mikael Otfried Geffert-2 wrote: I remember: you can either use one or the other. Either you use the harddisk and do not seethe CF card or vice versa. I think that is a hardware issue - the board is designed that way. Otfried - Original Message - From: Mikael Arhelger [EMAIL PROTECTED] To: soekris-tech@lists.soekris.com Sent: Thursday, February 14, 2008 4:07 PM Subject: [Soekris] 5501 under Debian can not see harddisk 5K160 Got my first 5501 and booting Debian from 2GB Kingston CF card. Can do fidsk -l to see only /dev/hdb but not the harddisk; a Hitachi 5K160 ATA 40GB. A sync command will make the yellow LED brighter for a short time (split second). Any idea? Bios can not see or ... Any pointer appreciated. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Storage device problem
Hi, I run OpenBSD from CF cards without any problem. CF is the cheap and flexible option. For the price of a single 2.5 HDD you can have 10 or more CF cards lying around, each with its own OS and/or configuration, or as backups. Writing speed is an issue with the cheaper cards, whether that is a problem depends on how much writing you are planning to do (if you're setting up e.g. a firewall/router probably almost none, apart from the install). I've had trouble with a fast Apacer card, now I'm using Dane-Elec XS cards for speed, but I bought those only recently, so I don't know yet how they will hold. Unless you're planning to write large amounts of data frequently, investing in a 2.5 HDD seems quite crazy. If I can give you an advice, it's to just buy the cheapest CF card you can find to start with, you won't be losing a lot if you decide to switch to HDD later. You would the other way around.. Good luck, Bill On Tue, 2008-02-12 at 16:44 +0100, voip crazy wrote: Dear list, I planned to install Debian on the soekris 5501 box but I'm not sure if install it on a CF card or in a SATA 2,5 Hdd drive. Which kind of install will be easier? which one would have less error or patching? How about the life of a CF card compare with the life of a SATA drive? Thanks in advance. VoipCrazy ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- SOD Against NLS ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Storage device problem
I forgot to mention: no moving parts, no mount trouble, no heat issues, no cables in weird places, no noise, no need for extra fans - CF!;) Bill On Tue, 2008-02-12 at 16:44 +0100, voip crazy wrote: Dear list, I planned to install Debian on the soekris 5501 box but I'm not sure if install it on a CF card or in a SATA 2,5 Hdd drive. Which kind of install will be easier? which one would have less error or patching? How about the life of a CF card compare with the life of a SATA drive? Thanks in advance. VoipCrazy ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- SOD Against NLS ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] OpenBSD Soekris - Soekris PXE boot
Hi Ervin, On Thu, 2008-02-07 at 23:37 +0100, ervin wrote: guidance for OpenBSD Bootstrap here - http://www.ultradesic.com/?section=74 Which does not reveal an awful lot about specific trouble and possible errors.. In fact, I recently tarted up my own doc, which is a bit more verbose, especially in the error section (which is why I wrote it in the first place): http://stsx.xs4all.nl/www.stsx.org/openbsd/obsd-bootsoekris.html (Added a few lines about firewalling this morning). Thanks anyway, and Brian as well, Bill By the way, I think my initial error was a faulty next-server, after that I went into a chain of overlapping fixes/new errors. It happens.. On 06/02/2008, Brian Candler [EMAIL PROTECTED] wrote: On Wed, Feb 06, 2008 at 12:16:45PM +0100, Bill Maas wrote: I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without problems, for booting my Soekris boxens. However, when I run [almost] the exact same config on a net4521, and try booting another Soekris box off that one, TFTP times out. On the boot server, 'inetd -d' shows a rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when the client tries to fetch the pxeboot image. The same happens by the way when I try to fetch files manually using 'tftp' from Linux or OpenBSD, but those implementations seem to have a reputation for stupidity and brokenness. The really weird thing is that the same config works on a PC as boot server, but not on a Soekris box. Anyone seen this before? No, but my suggestion would be to disable your OpenBSD pf temporarily and see if the problem goes away. TFTP is a pretty ugly protocol. An exchange looks like this: client UDP port XXX UDP port 69server UDP port XXX UDP port YYY UDP port XXX UDP port YYY ...etc That is, the server choses a dynamic port for sending its replies from. It's not easy to firewall properly. Maybe the difference between your clients is that one happens to have chosen XXX=69 as its source port, but the other has chosen XXX=dynamic high port (which is the normal thing to do). tcpdump could confirm or refute that hypothesis. Regards, Brian. ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- SOD Against NLS ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
[Soekris] OpenBSD Soekris - Soekris PXE boot
Hi,
I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
problems, for booting my Soekris boxens. However, when I run [almost]
the exact same config on a net4521, and try booting another Soekris box
off that one, TFTP times out. On the boot server, 'inetd -d' shows a
rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
the client tries to fetch the pxeboot image. The same happens by the way
when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
but those implementations seem to have a reputation for stupidity and
brokenness.
The really weird thing is that the same config works on a PC as boot
server, but not on a Soekris box. Anyone seen this before?
Files:
/etc/dhcpd.conf:
---
# $OpenBSD: dhcpd.conf,v 1.1 1998/08/19 04:25:45 form Exp $
#
# DHCP server options.
# See dhcpd.conf(5) and dhcpd(8) for more information.
#
# Network: 192.168.1.0/255.255.255.0
# Domain name: my.domain
# Name servers: 192.168.1.3 and 192.168.1.5
# Default router: 192.168.1.1
# Addresses:192.168.1.32 - 192.168.1.127
#
#shared-network LOCAL-NET {
# option domain-name my.domain;
# option domain-name-servers 192.168.1.3, 192.168.1.5;
#
# subnet 192.168.1.0 netmask 255.255.255.0 {
# option routers 192.168.1.1;
#
# range 192.168.1.32 192.168.1.127;
# }
#}
# Subnetwork declaration
subnet 192.168.1.0 netmask 255.255.255.0 {
option domain-name localdomain;
option subnet-mask 255.255.255.0;
option broadcast-address 192.168.1.255;
option domain-name-servers 194.109.6.66, 194.109.104.104;
option routers 192.168.1.1;
}
# Fixed addresses for local hosts
# (this is for PXE-booting the boor server itself)
group {
next-server 192.168.1.2;
host net4521 {
hardware ethernet 00:00:24:XX:XX:XX;
fixed-address 192.168.1.11;
filename pxeboot;
}
}
# (this is for my clients)
group {
next-server 192.168.1.11;
host net4801 {
hardware ethernet 00:00:24:YY:YY:YY;
fixed-address 192.168.1.8;
filename pxeboot;
}
host net5501 {
hardware ethernet 00:00:24:ZZ:ZZ:ZZ;
fixed-address 192.168.1.9;
filename pxeboot;
}
}
---
/etc/inetd.conf
---
[...]
tftpdgram udp wait
root/usr/libexec/tcpd /usr/libexec/tftpd -l -s /tftpboot
[...]
---
/etc/hosts.allow
---
# /etc/hosts.allow - see hosts_options(5)
sshd: ALL : allow
sendmail: LOCAL 192.168.1. .internal : allow
tftpd: LOCAL 192.168.1. .internal : allow
ftpd: LOCAL 192.168.1. .internal : allow
ALL: ALL : deny
---
/etc/hostname.sis0
---
inet 192.168.1.11 255.255.255.0 NONE
---
/etc/hostname.sis0
---
inet 10.0.1.11 255.0.0.0 NONE
---
(tried booting with and without hostname.sis1 configured - didn't make a
difference).
# pfctl -s rules
---
scrub in all fragment reassemble
block return all
block return in quick inet6 all
pass out all flags S/SA keep state
pass in quick on lo all flags S/SA keep state
block drop in on ! lo inet from 127.0.0.0/8 to any
block drop in on ! lo inet6 from ::1 to any
block drop in inet from 127.0.0.1 to any
block drop in on ! sis0 inet from 192.168.1.0/24 to any
block drop in inet from 192.168.1.111 to any
block drop in inet6 from ::1 to any
block drop in on lo0 inet6 from fe80::1 to any
block drop in on sis0 inet6 from fe80::200:24ff:fec1:ef50 to any
pass in quick on sis1 all flags S/SA keep state
pass in inet proto icmp from lan1 to any icmp-type echoreq code 0 keep
state
pass in log proto tcp from lan1 port = bootps to (sis0) port = bootpc
flags S/SA keep state
pass in log proto tcp from lan1 to (sis0) port = ftp flags S/SA keep
state
pass in log proto tcp from lan1 to (sis0) port 49151 flags S/SA keep
state
pass in log proto udp from lan1 port = bootps to (sis0) port = bootpc
keep state
pass in log proto udp from lan1 to (sis0) port = tftp keep state
pass in on sis0 proto tcp from any to (sis0) port = ssh flags S/SA keep
state
pass in log on sis0 proto tcp from any to (sis0) port = smtp flags S/SA
keep state
pass out log
Re: [Soekris] OpenBSD Soekris - Soekris PXE boot
Hi Yannick, On Wed, 2008-02-06 at 08:05 -0500, Yannick Gravel wrote: Hi Bill, Except for the first one, I have been doing just that with my net4501, net4801, net5501. Do you run CURRENT or some 3.8-4.2 on you net4521? The -l option for tftpd is new to 4.3, so if you are not running CURRENT that may be the reason for the rapid re-execution. I run 4.2 GENERIC, the '-l' was from later tests, without it the same thing occurs (Transfer timed out on the client side, by the way). Another big difference between PC and Soekris are of course disk access and program startup speed (net4521 == slo). Maybe some timeout, inetd waiting for tcpd, or PXE-tftp waiting for inetd. But I haven't found anything usable/relevant so far. Here's what it looks like on the server (net4521): [EMAIL PROTECTED]:~# inetd -d ADD: tftp proto=udp, wait.max=1.256 user:group=root:wheel builtin=0 server=/usr/libexec/tcpd [...] someone wants tftp 12472 execv /usr/libexec/tcpd reaping asked for 12472 reaped, status 100 restored tftp, fd 4 someone wants tftp 8305 execv /usr/libexec/tcpd reaping asked for 8305 reaped, status 100 restored tftp, fd 4 [...] More 'someone wants tftp' blocks follow in rapid succession. If the connect succeeds, as on the PC, they appear less often and with more time in between. On the client (net4801): boot f0 BootManage UNDI, PXE-2.0 (build 082) BootManage PXE-2.0 PROM 1.0, NATSEC 1.0, SDK 3.0/082 (OEM52) Copyright (C) 1989,2000 bootix Technology GmbH, D-41466 Neuss. PXE Software Copyright (C) 1997, 1998, 1999, 2000 Intel Corporation. Licensed to National Semiconductor CLIENT MAC ADDR: 00 00 24 C2 A3 F0 CLIENT IP: 192.168.1.4 MASK: 255.255.255.0 DHCP IP: 192.168.1.11 GATEWAY IP: 192.168.1.1 PXE-E32: TFTP open timeout TFTP.. I also wonder what that last line means, nothing seems to happen on the server side - I'd rather have my prompt back;). Bill Cheers Y. Bill Maas wrote: Hi, I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without problems, for booting my Soekris boxens. However, when I run [almost] the exact same config on a net4521, and try booting another Soekris box off that one, TFTP times out. On the boot server, 'inetd -d' shows a rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when the client tries to fetch the pxeboot image. The same happens by the way when I try to fetch files manually using 'tftp' from Linux or OpenBSD, but those implementations seem to have a reputation for stupidity and brokenness. The really weird thing is that the same config works on a PC as boot server, but not on a Soekris box. Anyone seen this before? [...] -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] OpenBSD Soekris - Soekris PXE boot
Hi Yannick,
it seems you're right: yet another test, with the '-l' removed again,
and:
(I)nstall, (U)pgrade or (S)hell?
I'm still not sure what went wrong in the first place, but at least it
works now.
Thanks.
Bill
On Wed, 2008-02-06 at 08:05 -0500, Yannick Gravel wrote:
Hi Bill,
Except for the first one, I have been doing just that with my net4501,
net4801, net5501.
Do you run CURRENT or some 3.8-4.2 on you net4521? The -l option
for tftpd is new to 4.3, so if you are not running CURRENT that may be
the reason for the rapid re-execution.
Cheers
Y.
Bill Maas wrote:
Hi,
I've been running dhcpd/tftpd on OpenBSD 3.8-4.2 on a PC without
problems, for booting my Soekris boxens. However, when I run [almost]
the exact same config on a net4521, and try booting another Soekris box
off that one, TFTP times out. On the boot server, 'inetd -d' shows a
rapid re-execution of tcpd (or of tftpd without using TCP wrappers) when
the client tries to fetch the pxeboot image. The same happens by the way
when I try to fetch files manually using 'tftp' from Linux or OpenBSD,
but those implementations seem to have a reputation for stupidity and
brokenness.
The really weird thing is that the same config works on a PC as boot
server, but not on a Soekris box. Anyone seen this before?
Files:
/etc/dhcpd.conf:
---
# $OpenBSD: dhcpd.conf,v 1.1 1998/08/19 04:25:45 form Exp $
#
# DHCP server options.
# See dhcpd.conf(5) and dhcpd(8) for more information.
#
# Network: 192.168.1.0/255.255.255.0
# Domain name: my.domain
# Name servers: 192.168.1.3 and 192.168.1.5
# Default router: 192.168.1.1
# Addresses:192.168.1.32 - 192.168.1.127
#
#shared-network LOCAL-NET {
# option domain-name my.domain;
# option domain-name-servers 192.168.1.3, 192.168.1.5;
#
# subnet 192.168.1.0 netmask 255.255.255.0 {
# option routers 192.168.1.1;
#
# range 192.168.1.32 192.168.1.127;
# }
#}
# Subnetwork declaration
subnet 192.168.1.0 netmask 255.255.255.0 {
option domain-name localdomain;
option subnet-mask 255.255.255.0;
option broadcast-address 192.168.1.255;
option domain-name-servers 194.109.6.66, 194.109.104.104;
option routers 192.168.1.1;
}
# Fixed addresses for local hosts
# (this is for PXE-booting the boor server itself)
group {
next-server 192.168.1.2;
host net4521 {
hardware ethernet 00:00:24:XX:XX:XX;
fixed-address 192.168.1.11;
filename pxeboot;
}
}
# (this is for my clients)
group {
next-server 192.168.1.11;
host net4801 {
hardware ethernet 00:00:24:YY:YY:YY;
fixed-address 192.168.1.8;
filename pxeboot;
}
host net5501 {
hardware ethernet 00:00:24:ZZ:ZZ:ZZ;
fixed-address 192.168.1.9;
filename pxeboot;
}
}
---
/etc/inetd.conf
---
[...]
tftpdgram udp wait
root/usr/libexec/tcpd /usr/libexec/tftpd -l -s /tftpboot
[...]
---
/etc/hosts.allow
---
# /etc/hosts.allow - see hosts_options(5)
sshd: ALL : allow
sendmail: LOCAL 192.168.1. .internal : allow
tftpd: LOCAL 192.168.1. .internal : allow
ftpd: LOCAL 192.168.1. .internal : allow
ALL: ALL : deny
---
/etc/hostname.sis0
---
inet 192.168.1.11 255.255.255.0 NONE
---
/etc/hostname.sis0
---
inet 10.0.1.11 255.0.0.0 NONE
---
(tried booting with and without hostname.sis1 configured - didn't make a
difference).
# pfctl -s rules
---
scrub in all fragment reassemble
block return all
block return in quick inet6 all
pass out all flags S/SA keep state
pass in quick on lo all flags S/SA keep state
block drop in on ! lo inet from 127.0.0.0/8 to any
block drop in on ! lo inet6 from ::1 to any
block drop in inet from 127.0.0.1 to any
block drop in on ! sis0 inet from 192.168.1.0/24 to any
block drop in inet from 192.168.1.111 to any
Re: [Soekris] What ever happened to the 5501 19 inch case from Soekris itself?
Hi Mark, Wim Vandeputte from KD85.com, who also happens to include a User Manual and directives for flashing the boot ROM with every Soekris board/box leaving the nest;), ships Soekris boards in every convceivable case format, I believe. See http://www.kd85.com/. Bill On Fri, 2008-01-25 at 12:22 -0800, Mark K. Mellis wrote: I may be asleep at the switch here - did the 19 inch case for the 5501 that is mentioned on the Soekris web site ever ship? If not, any alternatives to the applianceshop.eu one mentioned on the list the other day? Thanks, ...Mark Mark K. Mellis Http://www.mellis.com/~mkm ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net5501 and 24VDC ?
On Sun, 2008-01-20 at 14:08 +0100, Marc Balmer wrote: René Luckow wrote: As for the information on the wiki there, well, that is simply not true and I will edit it when I get to work tomorrow, but that's the thing about wiki's anyone can add any information, right or wrong. I wrote that, under the flag: It's a quick and dirty setup - feel free to modify it in any way., see http://lists.soekris.com/pipermail/soekris-tech/2007-September/012995.html I still think the term should applies, whether it's company policy or not (EU20 switches come with a User Manual). Bill -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] net5501 and 24VDC ?
On Sat, 2008-01-19 at 01:10 +0200, Jussi Peltola wrote: Still, I agree the manual should be released. But I doubt it is being intentionally delayed or anything... The introduction statement on http://www.soekris.com/index.htm suggsts you may be right: Welcome to Soekris Engineering's website. We're now introducing the new net4801 ;) Bill -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Smallest possible MFS /dev on OpenBSD
On Tue, 2008-01-15 at 14:56 +0200, Lars Noodén wrote: I've got OpenBSD 4.2 on a net4801 and am mounting /dev in RAM. I'd like to reduce the amount of memory used even further. Currently only %4 is used: Filesystem SizeUsed Avail Capacity Mounted on mfs:21800 853K 34.0K777K 4%/dev What should I be aware of in /etc/fstab to minimize the memory footprint of /dev? Currently, I have: swap /dev mfs rw,-P/dev.base,-i1,-s2300,-c264,-m5,nosuid 0 0 Couldn't resist .. here's ho far I got after about 45 mins. of testing: # mount_mfs -s 512 -b 32768 -f 32768 -P/var/dev -o nosuid swap /mnt/test/ # df /mnt/test Filesystem 512-blocks Used Avail Capacity Mounted on mfs:17962 192 1286467%/mnt/test/ # df -h /mnt/test Filesystem SizeUsed Avail Capacity Mounted on mfs:17962 96.0K 64.0K 32.0K67%/mnt/test/ Note: '-i' option is ignored here - it doesn't make any noticeable difference at all. Block and fragment size seem to be the issues here. With these set to the defaults, I never got past 13%. You would only need to change the default bpi if you run out of inodes. Maintaining fragments appears to be pretty disk-intensive, I suspect that if (blk_siz == frag_siz (or whatever)) then it's turned off altogether. I couldn't reproduce Jean-Yves' 23.0K (mount_mfs -i 256 -s 592 -P /var/run/dev swap /dev): # mount_mfs -i 256 -s 512 -P/var/dev -o nosuid swap /mnt/test # df -h /mnt/test mfs:29060 143K 15.0K121K11%/mnt/test I suspect that the '-s' value 592 was a typo, since it was rejected over here: mount_mfs: reduced number of fragments per cylinder group from 72 to 64 to enlarge last cylinder group mount_mfs: inode table does not fit in cylinder group I can't find any hard data - apart from claims and statements - about dynamic growth (nor shrinkage!) of the MFS partition, but should any process hijack /dev e.g. as it's temp dir, then 96K (or 23, if possible), should be a significant improvement over the 853K in Lars' original posting, even if it's only a max. value. Bill -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Net5501 Second USB Port
On Wed, 2008-01-16 at 17:56 +0100, Marc Balmer wrote: It's just hilarious (and arrogant) that Soekris does not provide a system manual. I agree. Also, a lynx from the Soekris web site to the Soekris-Info Wiki would be a nice gesture towards a never resting User Base;). Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] grub net4521
On Thu, 2008-01-17 at 10:53 -0500, burlen wrote: Expert command (m for help): s Number of sectors (1-63, default 62): 63 Warning: setting sector offset for DOS compatiblity I don't exactly know what that means Mostly non-info I think, since it's completely undocumented yet pops up in every example out there. I can't give you any advice about GRUB, I never use it, only temporarily against my will after a fresh and unflexible Ubuntu installation. However, if you switch to Lilo you stand a considerable chance at being able to boot without a single problem. There's just one detail you should know about: if you want to run lilo with the '-C' (chroot) option, make sure that versions on both sides match. Bill -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris 4801 openbsd kernel config file
Hello Lars, AFAIK, once you rem stuff out, it isn't GENERIC anymore. Bill On Tue, 2008-01-15 at 22:15 +0200, Lars Noodén wrote: Bill Maas wrote: ... But keep in mind that anything other than GENERIC is \officially\ not supported;). Does that mean GENERIC as-is or GENERIC with stuff rem'ed out ? -Lars -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Soekris 4801 openbsd kernel config file
Hi Lars, if there is an up-to-date config around' you'll probably find it at kernel-panic.it or at ultradesic.com. But keep in mind that anything other than GENERIC is \officially\ not supported;). Bill On Tue, 2008-01-15 at 19:26 +0200, Lars Noodén wrote: Is there an 'official' openbsd kernel config file to make a stripped-down kernel for OpenBSD 4.2 ? -Lars ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- What's a computer? - MES ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] mounting /dev in RAM in OpenBSD
Hello Lars,
not sure about the OS you're using, probably FreeBSD (?), but you will
have to re-create the device nodes in the MFS partition under /dev after
setting up the MFS. Mounting stuff in memory is more complex than just
editing fstab.. Best done from a script from /etc/rc{whatever}. For an
OpenBSD example see these:
http://stsx.xs4all.nl/www.stsx.org/software/openbsd/mfsmount.txt
http://stsx.xs4all.nl/www.stsx.org/software/openbsd/README.mfsmount.txt
Good luck with it,
Bill
On Sat, 2007-12-22 at 20:29 +0200, Lars Noodén wrote:
I'd like to move the directories that are frequently written to off of
the CF and into RAM so that / can be mounted read-only.
/tmp and /var can be managed simply by modifying /etc/fstab:
# /var.base has a template of directories to populat /var with
swap /var mfs -P/var.base,-s16000,async,nosuid,nodev,noatime,rw 0 0
swap /tmp mfs async,nosuid,nodev,noatime,rw,-s=15000 0 0
It's not the same with /dev, however. I've tried a few variations, e.g.
swap /dev mfs -P/dev.base,async,rw,-s=5000,-i=128 0 0
but none allow the system to finish booting.
What are the good ways to do work with /dev in RAM ?
Regards,
-Lars
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech
--
The only statistics you can trust are those you falsified yourself
Winston Churchill
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] PXE Installation of OpenBSD 4.2 on Soekris 4801
Hi Lars, try this: - if the systems appears to hang after a message like: booting hd0a:/bsd.rd 4464500+838332 [58+204240+181750]=0x56cfd0 entry point at 0x100120 - .. then you have either set the wrong COM port after the stty command in boot.conf (what Soekris calls COM1 is com0 in OpenBSD!), or the boot commands in boot.conf may be the wrong order. Note that boot.conf is a command file, not a configuration file. (from my personal howto @ http://stsx.xs4all.nl/www.stsx.org/openbsd/obsd-bootsoekris.txt. I also found a very useful txt here, the other day: http://www.packetwatch.net/documents/guides/openbsd/soekris.php). Bill On Thu, 2007-12-20 at 23:49 +0200, Lars Noodén wrote: I'm working with a new Soekris 4801 w/256MB RAM and wonder about getting OpenBSD installed so that the system boots. Once I can get to SSH, I'm home free. Netboot using PXE and tfptd goes well: I've gotten things set up so I can do the net boot + install and that whole installation procedure appears to run flawlessly. However, when I get to booting the installed system, things just appear to hang. Here is the output from the serial connection just before that happens: http://www-personal.umich.edu/~lars/net4801.txt Where can I find info about how OpenBSD needs to be configured differently from your average tower/rack mounted system so as to boot from CF with no HD? Regards -Lars ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- The only statistics you can trust are those you falsified yourself Winston Churchill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Choosing Compact flash (Was: Memory upgrade)
Hi Joel, On Tue, 2007-12-18 at 14:26 -0800, Joel Jaeggli wrote: for the transcend 8GB 266x cf card I got: 30.3MB/s block write on a 4GB file 37.5MB/s block read on a 4GB file (iozone) 6.9MB/s rewrite this is on a system with 2.6 linux kernel ext3fs and 512MB of ram. Manufacturers obviously don't plug their cards into Soekris boxes for testing and benchmarking, so average speed in the lab (the 45 Mb/s claim) is not what the average end user will ever see;). Still not bad, the fastest I've seen a slow card do after a few local tests is 2.35 Mb/s. However, in my experience, on an operational level it's the other hardware that matters on a Soekris box. Switching from a net4521 to a net4801, and later to a net5501, really made a difference (key generation, SSH login time etc). CF card speed has played only a minor role. Bill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech -- The only statistics you can trust are those you falsified yourself Winston Churchill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
Re: [Soekris] Choosing Compact flash (Was: Memory upgrade)
On Thu, 2007-12-20 at 22:46 +1300, Steve Holdoway wrote: [...] Hey Bill, Please tell me you mean MB, not Mb! Cheers, Steve I was being a bit over-zealous there: I usually make that other mistake;) Bill -- The only statistics you can trust are those you falsified yourself Winston Churchill ___ Soekris-tech mailing list Soekris-tech@lists.soekris.com http://lists.soekris.com/mailman/listinfo/soekris-tech
