Re: [Soekris] net6501: doesn't start if not first unplugged for 8 minutes

[Christian Weisgerber]

On 2017-03-19, Dries Verachtert  wrote:

> I have a net6501 soekris device and it has a strange issue: when the
> device is working correctly and I reboot, then it doesn't start
> anymore: [...]
> If I keep the device unplugged from a power source for +/- 8 minutes,
> then it does start again and everything works like it should.

Hardware failure.
My 6501-70 had exactly the same problem.  Since it was still under
warranty, I sent it to Soekris and they replaced the board.

-- 
Christian "naddy" Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] net6501 hanging on reboots after 012345 (take 2)

[Christian Weisgerber]

On 2016-07-24, Denis Fondras  wrote:

> I'd like to know if anyone had solution for this :
> http://marc.info/?l=soekris-tech=131703535427987=2

Unrelated.

> The issue is annoying me. This is a reboot issue where POST hangs at 012345 on
> NET6501-70 using BIOS 1.41c.
> 
> It happens only when the board has been powered-up for some time, I'd say > 
> ~10
> minutes. When the issue arise, the only way to make it usable again is to let 
> it
> unplugged for some time (no figure here, it seems to vary from a few seconds 
> to
> more than 20 minutes depending on how much time it was previously powered-up).

I had the same problem with a net6501-70.  The machine was still
under warranty, so I contacted Soekris who told me to send it in
for repair and they ended up replacing the board.

> Could this be a hardware issue ?

Yes.

-- 
Christian "naddy" Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] net5501 compact case for free plus postage

[Christian Weisgerber]

On 2016-07-06, Christian Weisgerber <na...@mips.inka.de> wrote:

> I have a compact case for the net5501 to give away.

It has been claimed.

-- 
Christian "naddy" Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

[Soekris] net5501 compact case for free plus postage

[Christian Weisgerber]

I have a compact case for the net5501 to give away.

  KD85.com case for Soekris 5501 with 4 ethernet ports, painted
  black, no company logo, includes the second serial port DB9
  flatcable.

http://soekris.kd85.com/images/tn/dsc08043.jpg.html

I'll be happy to send it anywhere (DHL parcel, uninsured, no tracking)
if you paypal me the postage:

Germany:  5 EUR
EU:  10 EUR
World:   17 EUR

Whoever claims it first.

-- 
Christian "naddy" Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] Dying net6501 servers

[Christian Weisgerber]

On 2016-02-21, "Hendrickson, Kenneth"  wrote:

> The warranty is clearly not long enough, and this is a *systemic*
> problem, with many people experiencing the same failures.  Soekris
> ought to replace them.

There's a bias problem here: You're only going to hear from the
people who lost machines, not from those who didn't.

-- 
Christian "naddy" Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] I see development of the net6801 has been dropped.

[Christian Weisgerber]

On 2015-10-06, Vadim Troshchinskiy  wrote:

> Besides, Lanner seems to have something out already.

These Rangeley-based systems by ADI Engineering seem to be actually
available and in quantities down to 1:

US:
http://store.netgate.com/Production-Systems-C209.aspx

EU:
https://shop.voleatech.de/en/product-category/network-hardware/

-- 
Christian "naddy" Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] FreeBSD-recent + nanobsd for 4801?

[Christian Weisgerber]

On 2015-07-01, Harlan Stenn har...@everett.org wrote:

 For a long while I've been gently trying to get FreeBSD/nanobsd built
 for a 4801 and installed on a Sandisk 4G CF card.

Have you considered a standard FreeBSD install?

I'm running FreeBSD 10, as an NTP server, on a net5501 using 2G of
CF.  I originally installed FreeBSD 9.x and keep updating it with
freebsd-update(8).  Even 2G of flash is enough for a standard
install.

The information at http://wiki.soekris.info/Installing_FreeBSD is
to be taken with a grain of salt.  E.g., for the net5501 its
recommendations for a custom kernel include Add 'options CPU_SOEKRIS'
and 'options CPU_GEODE'.  However, CPU_SOEKRIS only concerns the
net4xxx and CPU_GEODE isn't referenced anywhere in the 10-STABLE
kernel source.

Regarding the net4801, CPU_SOEKRIS and CPU_ELAN_* only appear to
concern the GPIO port, so if you don't hook up anything there, my
best guess is that a standard install with GENERIC should work just
fine.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] FreeBSD-recent + nanobsd for 4801?

[Christian Weisgerber]

On 2015-07-01, Christopher Sean Hilton ch...@vindaloo.com wrote:

 Regarding the net4801, CPU_SOEKRIS and CPU_ELAN_* only appear to
 concern the GPIO port, so if you don't hook up anything there, my
 best guess is that a standard install with GENERIC should work just
 fine.

 The net4801 is supposed to be perfect as an inexpensive gps driven
 clock for ntpd. But I thought that having CPU_ELAN and CPU_ELAN_PPS in
 the kernel added precision to the timekeeping?

Looking again, I see that it's the net4501 that is based on the AMD
ElanSC520.  The net4801 has an NSC SC1100 and is thus completely
unaffected by CPU_ELAN_*.

FreeBSD's special Elan support is concentrated in a single file,
sys/i386/i386/elan-mmcr.c:
https://svnweb.freebsd.org/base/stable/10/sys/i386/i386/elan-mmcr.c?view=log

CPU_ELAN_PPS is only of interest if you want to precisely timestamp
a PPS signal on a GPIO pin.  CPU_ELAN_XTAL is for people who want
to run the CPU from a different clock.  CPU_ELAN just pulls in the
support for the other CPU_ELAN_* defines.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] com speed changing during boot

[Christian Weisgerber]

Jeff Simmons:

  That's due to different defaults in the Soekris comBIOS (19200) and
  the OpenBSD boot loader (9600).  If you want to switch the latter to
  19200, add
  stty com0 19200
  to /etc/boot.conf.
 
 Tried that, doesn't work.

Yes, it does.

 Can't read the /etc/boot.conf file until the HD is mounted.

The boot loader can read the root file system.  It loads etc/boot.conf,
etc/random.seed, and, most significantly, the kernel that way.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] com speed changing during boot

[Christian Weisgerber]

On 2015-06-06, Jeff Simmons j...@j-simmons.net wrote:

 I have suddenly run into a problem with one of my Soekris net6501 boxes. When 
 I first boot it, the com speed is 19200, but just before the boot prompt it 
 switches to 9600. This is raising problems with upgrading the OS (OpenBSD). 
 Any idea what's wrong, and anything I can do to fix this?  

That's due to different defaults in the Soekris comBIOS (19200) and
the OpenBSD boot loader (9600).  If you want to switch the latter to
19200, add

stty com0 19200

to /etc/boot.conf.  Alternatively, you could set the comBIOS to
9600.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] OpenBSD 5.6 on a 4801?

[Christian Weisgerber]

On 2015-04-09, Peter Bako pe...@bakonet.org wrote:

 Has anybody successfully installed OpenBSD 5.6 on a net4801?  I have two of
 these units (both with the latest 133 firmware), each failing in the exact
 same way, at the exact same place, while doing a PXE install.  I can see my
 TFTP server being queried for pxeboot, bsd.rd and boot.conf, so I know the
 PXE setup is correct.  On my serial terminal I can see the box starting the
 BSD boot, loads the kernel then displays entry point at 0x200120 and
 stops.

Do you have a file etc/boot.conf

stty com0 19200
set tty com0

on your TFTP server?  Without this, the kernel will switch its
output to the nonexistent video console.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] 6501 with OpenBSD as firewall

[Christian Weisgerber]

On 2014-03-16, Marc Balmer m...@msys.ch wrote:

 What are people's experiences with the net 6501 and (recent)
 OpenBSDs, for firewalling and VPN purposes?

It works. *shrug*

There are two issues to be aware of:

* Some mSATA SSDs are not detected reliably at boot.  The KingSpec
  modules sold by Wim work, but the Transcend ones offered by Soekris
  are flakey.

* The OpenBSD installer does not recognize the 6501 as a multiprocessor
  system and will only install an SP kernel by default.  A manually
  installed MP kernel recognizes the 6501 as a dual-CPU system,
  though.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] net6501 + Transcend SSD + OpenBSD boot issues

[Christian Weisgerber]

David Paschal pasc...@surewest.net wrote:

 Has anybody had any success in finding an SLC mSATA SSD (either the
 Cactus 900S or otherwise) that (1) is currently and readily available
 for purchase, and (2) works reliably with the net6501 and OpenBSD?

The 4GB KingSpec one I bought from Wim Vandeputte works fine.
http://kd85.com/

(And yes, I also have a Transcend one that is flakey.)

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] net6501-50 normal working temperature

[Christian Weisgerber]

Lon Willett x_soek...@lonw.net wrote:

 My net6501s (standard case) were running rather hot too, but had no 
 problems.  Then when summer came, they even started throttling the CPU 
 because of high temperature.

How can you tell that the CPU is throttling?

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] FreeBSD 9.2 kernel config for 5501?

[Christian Weisgerber]

Martin Johnson martin.johnson.uk.li...@googlemail.com wrote:

 Or, can anyone confirm that the default GENERIC kernel in 9.2 runs
 properly on the 5501 hardware?

Yes, I can confirm this.  I originally installed 9.1 on my net5501
and later upgraded it to 9.2 with freebsd-update.  Plain GENERIC
all the way.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] net6501 : CPU architecture ?

[Christian Weisgerber]

Chris Cappuccio ch...@nmedia.net wrote:

 There is no advantage to 64bit on this box unless forward-looking binary
 compatibility is on your list.

If you are doing lots of crypto by way of OpenSSL's libcrypto, 64-bit
can be significantly faster for a number of popular algorithms (AES,
SHA256, RSA, DSA, ECDSA, ECDH).

-- net6501-50, OpenBSD 5.4/i386 --
OpenSSL 1.0.1c 10 May 2012
built on: date not available
options:bn(64,32) rc4(8x,mmx) des(ptr,risc1,16,long) aes(partial) idea(int) 
blowfish(idx) 
compiler: information not available
The 'numbers' are in 1000s of bytes per second processed.
type 16 bytes 64 bytes256 bytes   1024 bytes   8192 bytes
md2  0.00 0.00 0.00 0.00 0.00 
mdc2  1316.45k 1615.99k 1712.39k 1738.76k 1747.26k
md4   4936.01k17990.31k54015.57k   108121.13k   154028.65k
md5   3997.51k15227.37k47733.62k   102857.57k   155181.44k
hmac(md5) 5154.34k18515.84k0.89k   111454.41k   158165.48k
sha1  4546.58k15273.07k42995.41k78297.56k   103159.32k
rmd1603883.51k11894.88k27480.88k40856.58k47965.38k
rc4  52165.95k70481.37k76799.49k79048.72k79729.12k
des cbc  16145.39k17089.11k17401.37k17477.06k17499.85k
des ede3  5849.76k 5978.34k 6014.64k 6023.91k 6028.33k
idea cbc  8713.12k 9162.16k 9276.64k 9305.81k 9316.02k
seed cbc 0.00 0.00 0.00 0.00 0.00 
rc2 cbc   7512.73k 7903.23k 7994.26k 8017.48k 8025.98k
rc5-32/12 cbc0.00 0.00 0.00 0.00 0.00 
blowfish cbc 25897.38k29496.75k30504.48k30706.78k30911.87k
cast cbc 14125.50k15081.97k15407.46k15485.53k15510.37k
aes-128 cbc   8900.38k 9524.60k 9745.44k 9801.82k 9787.00k
aes-192 cbc   7336.21k 7803.41k 7963.81k 8004.21k 8017.82k
aes-256 cbc   6495.05k 6845.30k 6951.63k 6978.51k 6986.33k
camellia-128 cbc0.00 0.00 0.00 0.00 
0.00 
camellia-192 cbc0.00 0.00 0.00 0.00 
0.00 
camellia-256 cbc0.00 0.00 0.00 0.00 
0.00 
sha2564978.27k11987.99k21308.39k26482.61k28644.78k
sha5121707.65k 6828.93k10107.92k13988.66k15758.03k
whirlpool 2795.19k 6244.95k10795.29k13175.35k14149.57k
aes-128 ige   8487.77k 8848.31k 8958.21k 8984.66k 8994.87k
aes-192 ige   7056.14k 7304.10k 7373.99k 7402.06k 7402.74k
aes-256 ige   6319.89k 6482.58k 6540.25k 6553.60k 6559.04k
ghash26411.66k45536.45k55386.75k58784.40k59856.03k
  signverifysign/s verify/s
rsa  512 bits 0.001879s 0.000190s532.3   5252.2
rsa 1024 bits 0.011058s 0.000625s 90.4   1599.8
rsa 2048 bits 0.076061s 0.002336s 13.1428.0
rsa 4096 bits 0.566111s 0.009109s  1.8109.8
  signverifysign/s verify/s
dsa  512 bits 0.001942s 0.002161s515.0462.7
dsa 1024 bits 0.006225s 0.007201s160.6138.9
dsa 2048 bits 0.023164s 0.027479s 43.2 36.4
  signverifysign/s verify/s
 160 bit ecdsa (secp160r1)   0.0011s   0.0045s873.3222.0
 192 bit ecdsa (nistp192)   0.0015s   0.0063s683.9159.2
 224 bit ecdsa (nistp224)   0.0019s   0.0085s535.0117.7
 256 bit ecdsa (nistp256)   0.0024s   0.0111s425.2 90.1
 384 bit ecdsa (nistp384)   0.0053s   0.0274s190.2 36.5
 521 bit ecdsa (nistp521)   0.0113s   0.0646s 88.2 15.5
 163 bit ecdsa (nistk163)   0.0034s   0.0121s290.4 82.5
 233 bit ecdsa (nistk233)   0.0076s   0.0236s131.5 42.4
 283 bit ecdsa (nistk283)   0.0117s   0.0417s 85.7 24.0
 409 bit ecdsa (nistk409)   0.0308s   0.0955s 32.5 10.5
 571 bit ecdsa (nistk571)   0.0787s   0.2176s 12.7  4.6
 163 bit ecdsa (nistb163)   0.0035s   0.0132s289.6 76.0
 233 bit ecdsa (nistb233)   0.0075s   0.0259s132.5 38.7
 283 bit ecdsa (nistb283)   0.0117s   0.0465s 85.5 21.5
 409 bit ecdsa (nistb409)   0.0309s   0.1078s 32.3  9.3
 571 bit ecdsa (nistb571)   0.0788s   0.2478s 12.7  4.0
  op  op/s
 160 bit ecdh (secp160r1)   0.0039s257.3
 192 bit ecdh (nistp192)   0.0054s184.9
 224 bit ecdh (nistp224)   0.0073s137.9
 256 bit ecdh (nistp256)   0.0095s105.8
 384 bit ecdh (nistp384)   0.0236s 42.5
 521 bit ecdh (nistp521)   0.0545s 18.4
 163 bit ecdh (nistk163)   0.0059s168.5
 233 bit ecdh (nistk233)   0.0113s 88.7
 283 bit ecdh (nistk283)   0.0199s 50.2
 409 

Re: [Soekris] Net 6501 second serial port with Openbsd 5.3

[Christian Weisgerber]

Laurent Salle lsa...@taciturne.net wrote:

 I'm unsuccessful trying to use the second serial port on a Soekris Net 
 6501 with OpenBSD 5.3.
 
 $ cu -s 9600 -l cua01
 /dev/cua01: Device not configured
 link down
 $
 
 Anyone has been able to use this port ? With which command ?

It's cua03.

 See below dmesg output:

Exactly!

 com3 at puc0 port 0 apic 0 int 19: ti16750, 64 byte fifo
 com4 at puc1 port 0 apic 0 int 19: ti16750, 64 byte fifo
 com5 at puc2 port 0 apic 0 int 19: ti16750, 64 byte fifo
 com6 at puc3 port 0 apic 0 int 19: ti16750, 64 byte fifo
 com0 at isa0 port 0x3f8/8 irq 4: ti16750, 64 byte fifo
 com0: console

Note that it's com2 if you run amd64 and this has changed yet again
in OpenBSD -current.  In short, it's the port at puc0, and dmesg
will tell you as which com(4) device it shows up.  For com4 and
higher, you'll need to create the corresponding /dev entries with
MAKDEV tty04 etc. first.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] net6501 : CPU architecture ?

[Christian Weisgerber]

David Ruggiero thatseattle...@gmail.com wrote:

 So from this discussion, as far as I can tell,
 I have a choice of _four_ different OpenBSD kernels that could
 legitimately be booted on my net6501 and would probably run:
 
 32-bit i386 uniprocessor
 32-bit i386 SMP
 64-bit amd64 uniprocessor
 64-bit amd64 SMP

OpenBSD/i386 and /amd64 are different architectures, so the userland
is different, not just the kernel.  But yes, all of the above do
run on the net6501.

 A different company than Soekris might do some quick testing, or at
 least provide some pithy engineering insight and give its customers a
 quick rundown from their point of view on the advantages and
 disadvantages of 32 vs 64bit and non-smp vs smp - whether in general,
 or for specific common application needs.

I don't expect Soekris to comment on 32 bs 64-bit, because the Intel
CPUs used in the net6501 officially do not support 64-bit mode.
Yes, it happens to work fine, but that's not something you can
guarantee, much less advertise.  Now that public attention has been
drawn to the issue, I wonder if Soeren might have to fix it.  Oh
well.

Now that Intel has announced the new Silvermont Atoms with all the
goodies (AES-NI, RDRAND, and of course 64-bit mode), I'm wondering
how long it will take for these to trickle through to Soekris-format
boxes.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] net6501 : CPU architecture ?

[Christian Weisgerber]

Peter J. Holzer hjp+soek...@wsr.ac.at wrote:

 Is the BIOS even able to hide the lm bit? Or are operating systems
 supposed to ignore it and use information from the BIOS (DMI?) instead? 

For a decade, Intel and AMD have provided patches that work around
processor bugs and are implemented in microcode that is uploaded
by the BIOS.  Toggling CPU features in the same way isn't exactly
magic.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

[Soekris] Rubber feet dimensions?

[Christian Weisgerber]

Strange question time:  What are the nominal dimensions of the
little rubber feet attached to the bottom of the Soekris cases?
They tend to fall off.  I got a bunch of 12 x 3 mm ones as replacements,
but they aren't quite tall enough, at least on the net6501.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech

Re: [Soekris] Clock losing time on Net5501

[Christian Weisgerber]

Joakim Aronius joa...@aronius.se wrote:

[OpenBSD]
 I have not had time to look into the 'new' rc.d controlsystem for
 starting and stopping services but a cronjob with '/etc/rc.d/ntpd
 restart' should do the same thing as a reboot.. But I could not find
 anything on special handling like writing hw clock during shutdown,
 could be internal in ntpd.

ntpd(8) doesn't have anything to do with it.  The hardware clock
is set by the boot(9) call when halting/rebooting the system.

settimeofday(2) and clock_settime(2) will also set the hardware
clock.  However, only ntpd -s calls settimeofday, once at startup.
In normal operation, ntpd uses adjtime(2), which does not touch the
hardware clock.

-- 
Christian naddy Weisgerber  na...@mips.inka.de
___
Soekris-tech mailing list
Soekris-tech@lists.soekris.com
http://lists.soekris.com/mailman/listinfo/soekris-tech