Processeur Atom ?

classic Classic list List threaded Threaded
63 messages Options
1234
Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Tomas Bodzar-4
On Sat, Jun 12, 2010 at 11:53 AM, E.T <[hidden email]> wrote:
>> Saving 10 watts will save you (0.01kW * 24h * 365) = 87.6kWh per year.
>> Realistic savings might be around 20 watts, for a 35-40 watt P3 and
>> 15-20W Atom. Calculate for yourself if it is worth it.
>
> The future is processor ARM, Openbsd suppoorted ARM is good way. This
> month, canoncial and linaro((ARM, IBM, Freescale, TI, Ericsson, Samsung et
> Canonical)announces a job to make it compatible with the ARM world of open
> source and GNU / LINUX. It's good , no ?

Unix world was saying immediately after start of x86 what terrible
crap it is, but it was cheap and it's cheap so mostly it's still only
one option. But yes, ARM or MIPS are getting more and more presence
which is really good. Regarding that new alliance..... there is and
there was couple of them and most of them are just history. Maybe that
new will be more successful but who knows.

>
>> Since when have recycled machines cost 100 dollars a month?
>
> location server datacenter 100dollars/month

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

E.T-3
In reply to this post by Tomas Bodzar-4
> Are you sure that you know function of data center? Or maybe it's not
> standard in France, but here you have : access restrictions to
> datacenter with pictures, personal data, cameras are everywhere with
> long enough backup of data, encoded racks so only you or persons you
> allowed have access to your servers, fire protection, power backups
> and protections against overloading and a lot of other services.
> Security is not just about brilliantly written system/apps by OpenBSD
> engineers. When someone get access to your OpenBSD machine (which is
> much more easy at your home) and do boot> boot -s , then where is you
> security? ;-)

yes, datacenter in french, camera, backup, protection fire. Datacenter in
frech : inverter on fire, crash server datacenter attack, maintenance
operation catastrophy, server on fire site web black-out. It's reality in
french. There is no accident, datacenter america and canada?

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Tomas Bodzar-4
On Sat, Jun 12, 2010 at 12:22 PM, E.T <[hidden email]> wrote:

>> Are you sure that you know function of data center? Or maybe it's not
>> standard in France, but here you have : access restrictions to
>> datacenter with pictures, personal data, cameras are everywhere with
>> long enough backup of data, encoded racks so only you or persons you
>> allowed have access to your servers, fire protection, power backups
>> and protections against overloading and a lot of other services.
>> Security is not just about brilliantly written system/apps by OpenBSD
>> engineers. When someone get access to your OpenBSD machine (which is
>> much more easy at your home) and do boot> boot -s , then where is you
>> security? ;-)
>
> yes, datacenter in french, camera, backup, protection fire. Datacenter in
> frech : inverter on fire, crash server datacenter attack, maintenance
> operation catastrophy, server on fire site web black-out. It's reality in
> french. There is no accident, datacenter america and canada?
>

Can't find any physical. Only those like this one
http://tinyurl.com/hsbcbreach which is from employee of HSBC and those
datacenters are managed by different people and companies. Which is in
fact just confirmation that most of the problems come from inside of
companies. Because it's much more easy to apply social engineering or
simply pay some money to someone then try to break trough SW of
physical security.

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

E.T-3
> Can't find any physical. Only those like this one
> http://tinyurl.com/hsbcbreach which is from employee of HSBC and those
> datacenters are managed by different people and companies. Which is in
> fact just confirmation that most of the problems come from inside of
> companies. Because it's much more easy to apply social engineering or
> simply pay some money to someone then try to break trough SW of
> physical security.

And therefore, server web home it's dangerous ?. :)

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Tomas Bodzar-4
On Sat, Jun 12, 2010 at 12:41 PM, E.T <[hidden email]> wrote:
>> Can't find any physical. Only those like this one
>> http://tinyurl.com/hsbcbreach which is from employee of HSBC and those
>> datacenters are managed by different people and companies. Which is in
>> fact just confirmation that most of the problems come from inside of
>> companies. Because it's much more easy to apply social engineering or
>> simply pay some money to someone then try to break trough SW of
>> physical security.
>
> And therefore, server web home it's dangerous ?. :)

It depends on data which you provide and I don't think that Google
will show breaches to someones home. At least not now. Maybe in future
:D Anyway that part about hosting was very theoretical ;-) For many
people is server at home all they need and they are comfortable with
it. For some people it's not an option. So choose whatever you like.
Regarding performance of Atom see eg. this thread
http://marc.info/?l=openbsd-misc&m=126203835608528&w=2

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Miod Vallat
In reply to this post by E.T-3
> > Saving 10 watts will save you (0.01kW * 24h * 365) = 87.6kWh per year.
> > Realistic savings might be around 20 watts, for a 35-40 watt P3 and
> > 15-20W Atom. Calculate for yourself if it is worth it.
>
> The future is processor ARM

Bwahahahahahaha! Thanks for the laugh.

Miod

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

E.T-3
In reply to this post by Tomas Bodzar-4
> It looks same http://ark.intel.com/ProductCollection.aspx?familyID=29035
> , same cache size, same speed, no virtualization support and 510 has
> worse consumption, maybe because of integrated VGA. Why they don't
> specify eg. FSB for D510? Because it's worse? i86 is still same crap
> (and those unreadable and chaotic number of models and specifications
> prove it) because of backward compatibility to 86/186/286/386/486
> times like various privileged modes which completely bypass any
> security protection in SW.

Yes exactly, plateform Atom is bad. Low-cost ~60-70dollars. Soekris and
X7SPA-HF cost 250dollars.... . Budget student is impossible.  

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Tomas Bodzar-4
On Sat, Jun 12, 2010 at 2:02 PM, E.T <[hidden email]> wrote:

>> It looks same http://ark.intel.com/ProductCollection.aspx?familyID=29035
>> , same cache size, same speed, no virtualization support and 510 has
>> worse consumption, maybe because of integrated VGA. Why they don't
>> specify eg. FSB for D510? Because it's worse? i86 is still same crap
>> (and those unreadable and chaotic number of models and specifications
>> prove it) because of backward compatibility to 86/186/286/386/486
>> times like various privileged modes which completely bypass any
>> security protection in SW.
>
> Yes exactly, plateform Atom is bad. Low-cost ~60-70dollars. Soekris and
> X7SPA-HF cost 250dollars.... . Budget student is impossible.

Are we comparing CPUs or living standard in different countries and
professions? If we are talking about money then it's easy. Go for Atom
or any other Intel/AMD cheap line of CPUs

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

E.T-3
In reply to this post by Miod Vallat
> Et en quoi l'iPad est le futur de l'informatique ? Pourquoi donc n'y
> a-t-il pas d'ARM multiprocesseurs ou multicoeurs ? Pourquoi donc
> n'utilise-t-on jamais de processeurs ARM lorsque l'on a des besoins
> importants en entrC)es/sorties, par exemple pour des serveurs de fichiers
> ou de base de donnC)es ?


> Et en quoi l'iPad est le futur de l'informatique ?
Vous dC)formez mes propos lC  ou avez mal compris le sens de mon post. Arm
est le futur pour tout ce qui est mobile, mini-itx, pico-itx.

> Pourquoi donc n'y a-t-il pas d'ARM multiprocesseurs ou multicoeurs ?
http://www.arm.com/products/processors/classic/arm11/arm11-mpcore.php


> par exemple pour des serveurs de fichiers, ou de base de donnC)es ?
Attention, Mysql est trC)s mauvais pour la santC) de l'administrateur.

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Kevin Chadwick-2
In reply to this post by Jussi Peltola
On Sat, 12 Jun 2010 12:33:35 +0300
Jussi Peltola <[hidden email]> wrote:

> This is assuming 5 years service life,
> which is not very hard to get with carefully chosen hardware from the
> dumpster, but seems to be too much to ask when buying new...

Hear, Hear, Applies to most non specialist electronics. We just pulled
out an old microwave that my parents got in Canada over 20 years ago
because our 3 year old one just packed in. This is a recurring theme.

Old pcs without fans are great! and also improve the security of
schg on intel ;-)

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Kevin Chadwick-2
In reply to this post by Tomas Bodzar-4
 When someone get access to your OpenBSD machine (which is
> much more easy at your home) and do boot> boot -s , then where is you
> security? ;-)
>

A buildings a building and it depends on what measures are taken. At a
data center who knows who should be where. At home you can remove
secure from ttys asking for password on boot -s use solanoids which
lock the metal case to the pc from bios setting and set the bios to
prevent boot without pass or cdrom boot, lock doors, have alarms, know
there shouldn't be ANYONE going in there. It could easily though
unlikely be more secure aside from the cost of generators if your ups
goes flat.

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

E.T-3
> A buildings a building and it depends on what measures are taken. At a
> data center who knows who should be where. At home you can remove
> secure from ttys asking for password on boot -s use solanoids which
> lock the metal case to the pc from bios setting and set the bios to
> prevent boot without pass or cdrom boot, lock doors, have alarms, know
> there shouldn't be ANYONE going in there. It could easily though
> unlikely be more secure aside from the cost of generators if your ups
> goes flat.

chamber closed, rack closed

no boot :

Cd-rom
USB
PXE
ZIP

Enabled anti-malware bios, not read /boot :)

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Kevin Chadwick-2
On Sat, 12 Jun 2010 15:06:47 +0200
"E.T" <[hidden email]> wrote:

> > A buildings a building and it depends on what measures are taken. At a
> > data center who knows who should be where. At home you can remove
> > secure from ttys asking for password on boot -s use solanoids which
> > lock the metal case to the pc from bios setting and set the bios to
> > prevent boot without pass or cdrom boot, lock doors, have alarms, know
> > there shouldn't be ANYONE going in there. It could easily though
> > unlikely be more secure aside from the cost of generators if your ups
> > goes flat.
>
> chamber closed, rack closed
>
> no boot :
>
> Cd-rom
> USB
> PXE
> ZIP
>
> Enabled anti-malware bios, not read /boot :)
>

I'm not sure what you're saying, aside from cages are good?

Data centres are often complicated and have many with keys or lockpics,
kvms or people leaving fingerprints around etc and the machines have
been rebooted a lot, without precautions. We've always said fingerprint
readers were insecure but convenient, now most people are only just
realising this. The number of times I see security firms make stupid
mistakes is rediculous because security is often intangible unless
something terribly blatant happens.

If I thought physical security to my system at home was so important, I
can guarantee that I could make it more secure than most data centre's
and obviously ones that aren't in a military bunker and I would only
need to trust myself.

I also train most days which would add an extra layer, if I was in or
near.

Security is a process of choices, not a choice. At home you have more
control over choices, but likely less resources.

Of course that guy in Dubai has his own private spy satellite, his home
could really be something, if he cared and knew the right people.

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

E.T-3
> I'm not sure what you're saying, aside from cages are good?
>
> Data centres are often complicated and have many with keys or lockpics,
> kvms or people leaving fingerprints around etc and the machines have
> been rebooted a lot, without precautions. We've always said fingerprint
> readers were insecure but convenient, now most people are only just
> realising this. The number of times I see security firms make stupid
> mistakes is rediculous because security is often intangible unless
> something terribly blatant happens.

Can you say stupid mistakes in security?

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Kevin Chadwick-2
On Sat, 12 Jun 2010 16:13:02 +0200
"E.T" <[hidden email]> wrote:

> > I'm not sure what you're saying, aside from cages are good?
> >
> > Data centres are often complicated and have many with keys or lockpics,
> > kvms or people leaving fingerprints around etc and the machines have
> > been rebooted a lot, without precautions. We've always said fingerprint
> > readers were insecure but convenient, now most people are only just
> > realising this. The number of times I see security firms make stupid
> > mistakes is rediculous because security is often intangible unless
> > something terribly blatant happens.
>
> Can you say stupid mistakes in security?
>

That's what I meant??

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Landry Breuil-4
In reply to this post by E.T-3
On 6/12/10, E.T <[hidden email]> wrote:
>
> yes, datacenter in french, camera, backup, protection fire. Datacenter in
> frech : inverter on fire, crash server datacenter attack, maintenance
> operation catastrophy, server on fire site web black-out. It's reality in
> french. There is no accident, datacenter america and canada?

Can you try learning english ? This is an english-only mailing list.
You make us frogs really ashamed by you simili-engrish.

Oh, and stop talking about things you don't know, especially here...

Landry

Reply | Threaded
Open this post in threaded view
|

Processeur Atom ?

E.T-3
In reply to this post by Kevin Chadwick-2
Thank you for all the answers, I think we went around the issue.

bye

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

VICTOR TARABOLA CORTIANO
In reply to this post by E.T-3
Looks like you love Atom, I don't see why.

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Stuart Henderson
In reply to this post by Henning Brauer
On 2010-06-12, Henning Brauer <[hidden email]> wrote:
> * Nick <[hidden email]> [2010-06-11 12:55]:
>> If you want low power consumption and low cost, I'd suggest a small
>> PIII or Celeron based system, hard to beat for the price (usually,
>> free!).  IF the new, cool stuff has any real power savings, you are
>> unlikely to ever recoup the initial cost over recycled hardware.
>
> that might be (I am not convinced tho) with the electricity price in
> the US, but certainly isn't universal.
>

Especially the price of electricity in externally owned datacentres (*)
- and restrictions on current drawn; there are still places which allow
just 4A (@240V) per rack footprint (and 8A/footprint is fairly common).

(*) http://redinterbushouseexchangegatenorth.net/pricing.php        ;)

Reply | Threaded
Open this post in threaded view
|

Re: Processeur Atom ?

Nick Holland
In reply to this post by E.T-3
On 06/12/10 04:53, E.T wrote:
>> * Nick <[hidden email]> [2010-06-11 12:55]:
>>> If you want low power consumption and low cost, I'd suggest a small
>>> PIII or Celeron based system, hard to beat for the price (usually,
>>> free!).  IF the new, cool stuff has any real power savings, you are
>>> unlikely to ever recoup the initial cost over recycled hardware.
>
> it is a very bad idea, PIII low performance, low puissance, high hot, high
> electricity.

AGAIN, you are confusing "processors" for "systems".
I can show you PIII systems which idle at less than 30W of power.
Off-the-shelf, 100% conventional systems.

I can also show you PIII systems that draw more than 300W, and I
discarded one a while ago that probably could have maxed out at well
over 500w.

Hard numbers:
Compaq Deskpro EN PIII 800MHz
  128M RAM
  Onboard fxp
  PCI slot "router card" (rl chip attached to five port switch,
exposes four ports to the back of machine. Nice for tiny offices)
  unplugged the CDROM (saves a half watt or so)
  CF flash card in adapter (most expensive parts!)
  TWO unused PCI slots
dmesg below.


25w idle (power factor: 0.60, 42VA.  Yes, that power factor (PF)
sucks.  For those without an EE background, VA is what your
UPS/generator/power company must put out to power the device, Watts is
what the power company bills you for (at least in the US, I believe in
much of the rest of the world), Power Factor is the ratio between
them.  At least some areas are trying to force PFs towards unity, so
that power generated and power billed for are closer to the same)

In order to top that, you will need one very efficient power supply on
the Atom machine.  I have no doubt that the Atom chip and basic main
board draws less than that, but getting the second NIC and switch AND
power supply efficiencies to under 25W at the power cord will be
difficult.

>> that might be (I am not convinced tho) with the electricity price in
>> the US, but certainly isn't universal.

The calculations are.
Cost of money (i.e., interest rate), watts saved (if any), cost of a
kWh, initial costs, etc.  Plug in your numbers, find out what the ROI
is.  Add in what your AC costs are (watts in have to be removed, and
that's more watts to pump them out).  Evaluate results.

Going simpler, ignoring cost of money, IF your Atom machine draws 50%
of the power of my PIII, my quickie calc indicates you will save
105kWh.  If you also have to pay for AC, maybe double that number.

Granted, ROI (Return on Investment) isn't everything.  There's also
coolness, there's fun, etc.  If you are trying to implement 100
identical systems, recycling old hardware is probably more difficult
(=expensive) than buying new hardware.  Low power rack mount equipment
is hard to find now (Dell used to make a PowerEdge 350 system, which
was basically just a cheap desktop with a very long BIOS POST (ta-da!
now it is a server!).  Used a few of them in a past life, only had an
analog clamp-on ammeter at the time, but they appeared to draw under 60W.

If you want to talk about "power savings", get a wattmeter and quit
reading glossy sheets of one tiny part of the entire computer system.
 The numbers will surprise you.  (Fans ALONE on one Dell 1U system
draw over 50W at full speed.  Hopefully, they aren't at full speed
very often.  If my 40 year old 'fridge ever fails and I decide to buy
a new one, I'm getting a manual defrost unit)

For giggles, let's compare a soekris 4501:
   486 133MHz
   64M RAM
   two NICs (yes, supposed to have three, but one failed, and I pried
the overheating chip off the board.  Power savings!)
   No switch
   no expansion potential
   only serial port is in use for console
   1g CF storage

4W at idle.  (similar power factor as the Compaq deskpro).
Yes, as configured, the Soekris wins hands-down for power.  Even if
you add a five port mini-switch to balance things out, it will still
win.  However, it is an anemic little thing.  Home Use/Very Small
Office only...

My PIII system will pump a LOT of data.  I've used lesser systems
(with a much better than rl(4) NIC) to filter a 45Mbps line on a 700
user network.  The PCI bus will probably be a bottleneck before the
CPU will be.  Looks like the referenced Atom board is pretty good in
this regard: appears to have a PCI-e slot and a couple decent on-board
gigabit NICs.  Again, it is not the CPU, it's the surrounding hardware
that matters. IF you need and will use that much bandwidth, then yes,
THIS Atom board wins.  However, at no point did I think you were
worrying about a large office, high-speed environment, sounds like you
are looking at a home environment.

Nick.



OpenBSD 4.7 (GENERIC) #552: Sun Mar  7 21:43:02 MST 2010
    [hidden email]:/usr/src/sys/arch/i386/compile/GENERIC
cpu0: Intel Pentium III ("GenuineIntel" 686-class) 798 MHz
cpu0:
FPU,V86,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR,SSE
real mem  = 132542464 (126MB)
avail mem = 119754752 (114MB)
mainbus0 at root
bios0 at mainbus0: AT/286+ BIOS, date 06/06/02, BIOS32 rev. 0 @
0xe7300, SMBIOS rev. 2.3 @ 0xfd33c (47 entries)
bios0: vendor Compaq version "686P2 v3.13" date 06/06/2002
bios0: Compaq Deskpro
acpi0 at bios0: rev 0
acpi0: tables DSDT FACP SSDT SSDT SSDT APIC SSDT SSDT SSDT SSDT SSDT SSDT
acpi0: wakeup devices PCI0(S4) HUB_(S4) COM1(S4) COM2(S4) USB1(S3)
USB2(S3) PBTN(S4)
acpitimer0 at acpi0: 3579545 Hz, 24 bits
acpimadt0 at acpi0 addr 0xfee00000: PC-AT compat
cpu0 at mainbus0: apid 0 (boot processor)
cpu0: apic clock running at 132MHz
ioapic0 at mainbus0: apid 8 pa 0xfec00000, version 20, 24 pins
ioapic0: misconfigured as apic 0, remapped to apid 8
acpiprt0 at acpi0: bus 0 (PCI0)
acpiprt1 at acpi0: bus 2 (HUB_)
acpicpu0 at acpi0
acpibtn0 at acpi0: PBTN
bios0: ROM list: 0xc0000/0x8000 0xe0000/0x10000
pci0 at mainbus0 bus 0: configuration mode 1 (bios)
pchb0 at pci0 dev 0 function 0 "Intel 82815 Host" rev 0x02
vga1 at pci0 dev 2 function 0 "Intel 82815 Video" rev 0x02
wsdisplay0 at vga1 mux 1: console (80x25, vt100 emulation)
wsdisplay0: screen 1-5 added (80x25, vt100 emulation)
intagp0 at vga1
agp0 at intagp0: aperture at 0x44000000, size 0x4000000
ppb0 at pci0 dev 30 function 0 "Intel 82801BA Hub-to-PCI" rev 0x01
pci1 at ppb0 bus 2
fxp0 at pci1 dev 8 function 0 "Intel 82562" rev 0x01, i82562: apic 8
int 20 (irq 5), address 00:02:a5:56:0d:f8
inphy0 at fxp0 phy 1: i82562EM 10/100 PHY, rev. 0
rl0 at pci1 dev 9 function 0 "Realtek 8139" rev 0x10: apic 8 int 18
(irq 10), address 00:e0:4c:15:2a:07
rlphy0 at rl0 phy 0: RTL internal PHY
ichpcib0 at pci0 dev 31 function 0 "Intel 82801BA LPC" rev 0x01
pciide0 at pci0 dev 31 function 1 "Intel 82801BA IDE" rev 0x01: DMA,
channel 0 wired to compatibility, channel 1 wired to compatibility
wd0 at pciide0 channel 0 drive 0: <CF 1GB>
wd0: 1-sector PIO, LBA, 967MB, 1981728 sectors
wd0(pciide0:0:0): using PIO mode 4, DMA mode 2
pciide0: channel 1 disabled (no drives)
uhci0 at pci0 dev 31 function 4 "Intel 82801BA USB" rev 0x01: apic 8
int 23 (irq 10)
auich0 at pci0 dev 31 function 5 "Intel 82801BA AC97" rev 0x01: apic 8
int 17 (irq 5), ICH2 AC97
ac97: codec id 0x41445360 (Analog Devices AD1885)
ac97: codec features headphone, Analog Devices Phat Stereo
audio0 at auich0
isa0 at ichpcib0
isadma0 at isa0
com0 at isa0 port 0x3f8/8 irq 4: ns16550a, 16 byte fifo
com1 at isa0 port 0x2f8/8 irq 3: ns16550a, 16 byte fifo
pckbc0 at isa0 port 0x60/5
pckbd0 at pckbc0 (kbd slot)
pckbc0: using irq 1 for kbd slot
wskbd0 at pckbd0: console keyboard, using wsdisplay0
pmsi0 at pckbc0 (aux slot)
pckbc0: using irq 12 for aux slot
wsmouse0 at pmsi0 mux 0
pcppi0 at isa0 port 0x61
midi0 at pcppi0: <PC speaker>
spkr0 at pcppi0
lpt0 at isa0 port 0x378/4 irq 7
npx0 at isa0 port 0xf0/16: reported by CPUID; using exception 16
fdc0 at isa0 port 0x3f0/6 irq 6 drq 2
usb0 at uhci0: USB revision 1.0
uhub0 at usb0 "Intel UHCI root hub" rev 1.00/1.00 addr 1
mtrr: Pentium Pro MTRR support
vscsi0 at root
scsibus0 at vscsi0: 256 targets
softraid0 at root
wd0(pciide0:0:0): timeout
        type: ata
        c_bcount: 512
        c_skip: 0
pciide0:0:0: bus-master DMA error: missing interrupt, status=0x21
wd0c: device timeout reading fsbn 0 (wd0 bn 0; cn 0 tn 0 sn 0), retrying
wd0(pciide0:0:0): timeout
        type: ata
        c_bcount: 512
        c_skip: 0
pciide0:0:0: bus-master DMA error: missing interrupt, status=0x21
wd0: transfer error, downgrading to PIO mode 4
wd0(pciide0:0:0): using PIO mode 4
wd0c: device timeout reading fsbn 0 (wd0 bn 0; cn 0 tn 0 sn 0), retrying
wd0: soft error (corrected)
root on wd0a swap on wd0b dump on wd0b

1234