zfs discuss - Feb 2009 - Motherboard for home zfs/solaris file server

Hello,
I am building a home file server and am looking for an ATX mother board 
that will be supported well with OpenSolaris (onboard SATA controller, 
network, graphics if any, audio, etc). I decided to go for Intel based 
boards (socket LGA 775) since it seems like power management is better 
supported with Intel processors and power efficiency is an important 
factor. After reading several posts about ZFS it looks like I want ECC 
memory as well.

Does anyone have any recommendations?

Here are a few that I found. Any comments about those?

Supermicro C2SBX+
http://www.supermicro.com/products/motherboard/Core2Duo/X48/C2SBX+.cfm

Gigabyte GA-X48-DS4
gigabyte: 
http://www.gigabyte.com.tw/Products/Motherboard/Products_Overview.aspx?ProductID=2810

Intel S3200SHV
http://www.intel.com/Products/Server/Motherboards/Entry-S3200SH/Entry-S3200SH-overview.htm

Thanks for any help,
-Ilya

On 02/23/09 20:24, Ilya Tatar wrote:
> Hello,
> I am building a home file server and am looking for an ATX mother 
> board that will be supported well with OpenSolaris (onboard SATA 
> controller, network, graphics if any, audio, etc). I decided to go for 
> Intel based boards (socket LGA 775) since it seems like power 
> management is better supported with Intel processors and power 
> efficiency is an important factor. After reading several posts about 
> ZFS it looks like I want ECC memory as well.
>
> Does anyone have any recommendations?
Any motherboard for the Core2  or Core i7 Intel processors with the ICH 
southbridge (desktop boards) or
ESB2 soutbridge (server boards) will be well supported.  I recommend an 
actual Intel
board since they also always use the Intel network chip (well supported 
and tuned).   Many of the third
party boards from MSI, Gigabyte, Asus, DFI, ECS, and others also work, 
but for some (penny pinching)
reason, they tend to use network chips like Marvell that are not yet 
supported, or Realtek,
for which some of the models are supported.

So using an actual board from Intel Corp will be best supported right 
out of the box.
For that matter, because of the work we do with Intel, almost any of 
their boards will
be supported using the ICH 6, 7, 8, 9, or ICH10 SATA ports in either 
legacy or AHCI
mode.   Again, almost any version of the Intel network (NIC) chips are 
supported across
all their boards.  If you are able to find one that is not, I''d love to
hear about it and
add it to our work queue.

In the most recent builds of Solaris Nevada (SXCE), the integrated Intel 
graphics
found on many of the boards is well supported.  On other boards, use a 
low end
VGA card.
Again, if you find an Intel board where the graphics is not supported or 
not working,
please let us know the specifics and we''ll fix it.

Cheers,

Neal
>
> Here are a few that I found. Any comments about those?
>
> Supermicro C2SBX+
> http://www.supermicro.com/products/motherboard/Core2Duo/X48/C2SBX+.cfm
>
> Gigabyte GA-X48-DS4
> gigabyte: 
>
http://www.gigabyte.com.tw/Products/Motherboard/Products_Overview.aspx?ProductID=2810
>
>
> Intel S3200SHV
>
http://www.intel.com/Products/Server/Motherboards/Entry-S3200SH/Entry-S3200SH-overview.htm
>
>
> Thanks for any help,
> -Ilya
>
>
>
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Neal Pollack wrote:
> On 02/23/09 20:24, Ilya Tatar wrote:
...
>> efficiency is an important factor. After reading several posts about 
>> ZFS it looks like I want ECC memory as well.
...
> 
> Any motherboard for the Core2  or Core i7 Intel processors with the ICH 
Not. Intel decided we don''t need ECC memory on the Core i7 (one of the 
few truly idiotic things I can remember them doing lately). The OP 
specified ECC RAM, so Core i7 is a no go. Thanks for nothing, Intel.

-- 
Carson

> Not. Intel decided we don''t need ECC memory on the Core i7 
I thought that was a Core i7 vs Xeon E55xx for socket
LGA-1366 so that''s why this X58 MB claims ECC support:
http://supermicro.com/products/motherboard/Xeon3000/X58/X8SAX.cfm

>>>>> "rl" == Rob Logan <Rob at Logan.com>
writes:
    rl> that''s why this X58 MB claims ECC support:

the claim is worth something.  People always say ``AMD supports ECC
because the memory controller is in the CPU so they all support it, it
cannot be taken away from you by lying idiot motherboard manufacturers
or greedy marketers trying to segment users into different demand
groups'''' but you still need some motherboard BIOS to flip the
ECC
switch to ``wings stay on'''' mode before you start down the
runway.

Here is a rather outdated and Linux-specific workaround for cheapo AMD
desktop boards that don''t have an ECC option in their BIOS:

 
http://newsgroups.derkeiler.com/Archive/Alt/alt.comp.periphs.mainboard.asus/2005-10/msg00365.html
  http://hyvatti.iki.fi/~jaakko/sw/

The discussion about ECC-only vs scrub-and-fix, about how to read from
PCI if ECC errors are happening (though not necessarily which stick),
and his 10-ohm testing method, is also interesting.  I still don''t
understand what chip-kill means.

I remember something about a memory scrubbing kernel thread in
Solaris.  This sounds like the AMD chips have a hardware scrubber?
Also how are ECC errors reported in Solaris?  I guess this is getting
OT though.

Anyway ECC is not just a feature bullet to gather up and feel good.
You have to finish the job and actually interact with it.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 304 bytes
Desc: not available
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090224/f2ed572f/attachment.bin>

Rob Logan wrote:
> 
>> Not. Intel decided we don''t need ECC memory on the Core i7 
> 
> I thought that was a Core i7 vs Xeon E55xx for socket
> LGA-1366 so that''s why this X58 MB claims ECC support:
> http://supermicro.com/products/motherboard/Xeon3000/X58/X8SAX.cfm
They lie*. Read the Intel Core i7 specs - no ECC on any of them.

* They claim "future Nehalem processor families". These mysterious 
future CPUs may indeed support ECC. The Core i7-(920|940|965) do not.

-- 
Carson

On Tue, Feb 24, 2009 at 4:04 PM, Carson Gaspar <carson at taltos.org>
wrote:
>
> They lie*. Read the Intel Core i7 specs - no ECC on any of them.
>
> * They claim "future Nehalem processor families". These
mysterious future
> CPUs may indeed support ECC. The Core i7-(920|940|965) do not.
>

Given the current state of AMD, I think we all know that''s not likely. 
Why
cut into the revenue of your server line chips when you don''t have to?
 Right?

--Tim
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090224/edbfc4e6/attachment.html>

On Tue, Feb 24, 2009 at 2:29 PM, Tim <tim at tcsac.net>
wrote:
> Given the current state of AMD, I think we all know that''s not
likely.  Why
> cut into the revenue of your server line chips when you don''t have
to?
>  Right?
AMD has nothing to do with whether ECC exists on the Nehalem.

Most likely ECC is in the memory controller of the Nehalem die, it''s
just disabled on the i7. It wouldn''t make any sense to tape out a
whole new die for the server version of the chip. The Xeon could use
another stepping, but I''d expect Intel to use the same on both
consumer and server versions of the chip.

-B

-- 
Brandon High : bhigh at freaks.com

On Wed, Feb 25, 2009 at 2:39 PM, Brandon High <bhigh at freaks.com> wrote:
> On Tue, Feb 24, 2009 at 2:29 PM, Tim <tim at tcsac.net> wrote:
> > Given the current state of AMD, I think we all know that''s
not likely.
>  Why
> > cut into the revenue of your server line chips when you don''t
have to?
> >  Right?
>
> AMD has nothing to do with whether ECC exists on the Nehalem.
>
Of course it does.  Competition directly affects the features provided on
everyone in a market segment''s products.

>
> Most likely ECC is in the memory controller of the Nehalem die,
it''s
> just disabled on the i7. It wouldn''t make any sense to tape out a
> whole new die for the server version of the chip. The Xeon could use
> another stepping, but I''d expect Intel to use the same on both
> consumer and server versions of the chip.
>
The fact Intel put a memory controller on die is PROOF that AMD has a direct
effect on their product roadmap.  Do you think Intel would have willingly
killed off their lucrative northbridge chipset business without AMD forcing
their hand?  Please.

--Tim
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090225/e3449e42/attachment.html>

On Wed, Feb 25, 2009 at 12:52 PM, Tim <tim at tcsac.net>
wrote:
> Of course it does.  Competition directly affects the features provided on
> everyone in a market segment''s products.
The server and workstation market demands ECC. Any die that would be
used in the server or workstation market would need to have ECC.
> The fact Intel put a memory controller on die is PROOF that AMD has a
direct
> effect on their product roadmap.  Do you think Intel would have willingly
> killed off their lucrative northbridge chipset business without AMD forcing
> their hand?  Please.
Intel moved to on-die memory controller because the front side bus
architecture was becoming a bottleneck as the number of cores
increased.

The fact that AMD''s chips already have an on-die controller certainly
influenced Intel''s direction - I''m not disputing that. The
fact of the
matter is that an on-die MC is an efficient way to to have high
bandwidth and low latency access to memory. The IBM POWER 6 has on-die
memory controllers as well, which is less likely to be due to any
market pressure caused by AMD since the two firms'' products
don''t
directly compete. It''s just a reasonable engineering decision.

-B

-- 
Brandon High : bhigh at freaks.com

IIRC, the AMD board I have at my office has hardware ECC scrub.  I
have no idea if Solaris knows about this or makes any use of it (or
needs to?)

On Tue, Feb 24, 2009 at 2:50 PM, Miles Nordin <carton at ivy.net>
wrote:
>>>>>> "rl" == Rob Logan <Rob at Logan.com>
writes:
>
> ? ?rl> that''s why this X58 MB claims ECC support:
>
> the claim is worth something. ?People always say ``AMD supports ECC
> because the memory controller is in the CPU so they all support it, it
> cannot be taken away from you by lying idiot motherboard manufacturers
> or greedy marketers trying to segment users into different demand
> groups'''' but you still need some motherboard BIOS to flip
the ECC
> switch to ``wings stay on'''' mode before you start down
the runway.
>
> Here is a rather outdated and Linux-specific workaround for cheapo AMD
> desktop boards that don''t have an ECC option in their BIOS:
>
>
?http://newsgroups.derkeiler.com/Archive/Alt/alt.comp.periphs.mainboard.asus/2005-10/msg00365.html
> ?http://hyvatti.iki.fi/~jaakko/sw/
>
> The discussion about ECC-only vs scrub-and-fix, about how to read from
> PCI if ECC errors are happening (though not necessarily which stick),
> and his 10-ohm testing method, is also interesting. ?I still don''t
> understand what chip-kill means.
>
> I remember something about a memory scrubbing kernel thread in
> Solaris. ?This sounds like the AMD chips have a hardware scrubber?
> Also how are ECC errors reported in Solaris? ?I guess this is getting
> OT though.
>
> Anyway ECC is not just a feature bullet to gather up and feel good.
> You have to finish the job and actually interact with it.
>
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
>
>

When I type `zpool import` to see what pools are out there, it gets to

/1:     open("/dev/dsk/c5t2d0s0", O_RDONLY)             = 6
/1:     stat64("/usr/local/apache2/lib/libdevid.so.1", 0x08042758)
Err#2 ENOENT
/1:     stat64("/usr/lib/libdevid.so.1", 0x08042758)    = 0
/1:         d=0x02D90002 i=241208 m=0100755 l=1  u=0     g=2     sz=61756
/1:             at = Apr 29 23:41:17 EDT 2009  [ 1241062877 ]
/1:             mt = Apr 27 01:45:19 EDT 2009  [ 1240811119 ]
/1:             ct = Apr 27 01:45:19 EDT 2009  [ 1240811119 ]
/1:         bsz=61952 blks=122   fs=zfs
/1:     resolvepath("/usr/lib/libdevid.so.1",
"/lib/libdevid.so.1", 1023) = 18
/1:     open("/usr/lib/libdevid.so.1", O_RDONLY)        = 7
/1:     mmapobj(7, 0x00020000, 0xFEC70640, 0x080427C4, 0x00000000) = 0
/1:     close(7)                                        = 0
/1:     memcntl(0xFEC50000, 4048, MC_ADVISE, MADV_WILLNEED, 0, 0) = 0
/1:     fxstat(2, 6, 0x080430C0)                        = 0
/1:         d=0x04A00000 i=50333315 m=0060400 l=1  u=0     g=0    
rdev=0x01800340
/1:             at = Nov 19 21:19:26 EST 2008  [ 1227147566 ]
/1:             mt = Nov 19 21:19:26 EST 2008  [ 1227147566 ]
/1:             ct = Apr 29 23:23:11 EDT 2009  [ 1241061791 ]
/1:         bsz=8192  blks=1     fs=devfs
/1:     modctl(MODSIZEOF_DEVID, 0x01800340, 0x080430BC, 0xFEC51239, 0xFE8E92C0)
= 0
/1:     modctl(MODGETDEVID, 0x01800340, 0x00000038, 0x080D5A48, 0xFE8E92C0) = 0
/1:     fxstat(2, 6, 0x080430C0)                        = 0
/1:         d=0x04A00000 i=50333315 m=0060400 l=1  u=0     g=0    
rdev=0x01800340
/1:             at = Nov 19 21:19:26 EST 2008  [ 1227147566 ]
/1:             mt = Nov 19 21:19:26 EST 2008  [ 1227147566 ]
/1:             ct = Apr 29 23:23:11 EDT 2009  [ 1241061791 ]
/1:         bsz=8192  blks=1     fs=devfs
/1:     modctl(MODSIZEOF_MINORNAME, 0x01800340, 0x00006000, 0x080430BC,
0xFE8E92C0) = 0
/1:     modctl(MODGETMINORNAME, 0x01800340, 0x00006000, 0x00000002, 0x0808FFC8)
= 0
/1:     close(6)                                        = 0
/1:     ioctl(3, ZFS_IOC_POOL_STATS, 0x08042220)        = 0

and then the machine dies consistently with:

panic[cpu1]/thread=ffffff01d045a3a0:
BAD TRAP: type=e (#pf Page fault) rp=ffffff000857f4f0 addr=260 occurred in
module "unix" due to a NULL pointer dereference

zpool:
#pf Page fault
Bad kernel fault at addr=0x260
pid=576, pc=0xfffffffffb854e8b, sp=0xffffff000857f5e8, eflags=0x10246
cr0: 8005003b<pg,wp,ne,et,ts,mp,pe> cr4:
6f8<xmme,fxsr,pge,mce,pae,pse,de>
cr2: 260
cr3: 12b690000
cr8: c

         rdi:              260 rsi:                4 rdx: ffffff01d045a3a0
         rcx:                0  r8:               40  r9:            21ead
         rax:                0 rbx:                0 rbp: ffffff000857f640
         r10:          bf88840 r11: ffffff01d041e000 r12:                0
         r13:              260 r14:                4 r15: ffffff01ce12ca28
         fsb:                0 gsb: ffffff01ce985ac0  ds:               4b
          es:               4b  fs:                0  gs:              1c3
         trp:                e err:                2 rip: fffffffffb854e8b
          cs:               30 rfl:            10246 rsp: ffffff000857f5e8
          ss:               38

ffffff000857f3d0 unix:die+dd ()
ffffff000857f4e0 unix:trap+1752 ()
ffffff000857f4f0 unix:cmntrap+e9 ()
ffffff000857f640 unix:mutex_enter+b ()
ffffff000857f660 zfs:zio_buf_alloc+2c ()
ffffff000857f6a0 zfs:arc_get_data_buf+173 ()
ffffff000857f6f0 zfs:arc_buf_alloc+a2 ()
ffffff000857f770 zfs:dbuf_read_impl+1b0 ()
ffffff000857f7d0 zfs:dbuf_read+fe ()
ffffff000857f850 zfs:dnode_hold_impl+d9 ()
ffffff000857f880 zfs:dnode_hold+2b ()
ffffff000857f8f0 zfs:dmu_buf_hold+43 ()
ffffff000857f990 zfs:zap_lockdir+67 ()
ffffff000857fa20 zfs:zap_lookup_norm+55 ()
ffffff000857fa80 zfs:zap_lookup+2d ()
ffffff000857faf0 zfs:dsl_pool_open+91 ()
ffffff000857fbb0 zfs:spa_load+696 ()
ffffff000857fc00 zfs:spa_tryimport+95 ()
ffffff000857fc40 zfs:zfs_ioc_pool_tryimport+3e ()
ffffff000857fcc0 zfs:zfsdev_ioctl+10b ()
ffffff000857fd00 genunix:cdev_ioctl+45 ()
ffffff000857fd40 specfs:spec_ioctl+83 ()
ffffff000857fdc0 genunix:fop_ioctl+7b ()
ffffff000857fec0 genunix:ioctl+18e ()
ffffff000857ff10 unix:brand_sys_sysenter+1e6 ()

the offending disk, c5t2d0s0, is part of a mirror that if removed I can
see the results (from the other mirror half) and the machine does not crash.
all 8 labels look diff perfect

     version=13
     name=''r''
     state=0
     txg=2110897
     pool_guid=10861732602511278403
     hostid=13384243
     hostname=''nas''
     top_guid=6092190056527819247
     guid=16682108003687674581
     vdev_tree
         type=''mirror''
         id=0
         guid=6092190056527819247
         whole_disk=0
         metaslab_array=23
         metaslab_shift=31
         ashift=9
         asize=320032473088
         is_log=0
         children[0]
                 type=''disk''
                 id=0
                 guid=16682108003687674581
                 path=''/dev/dsk/c5t2d0s0''
                 devid=''id1,sd at
f31cd3f064658835c000a06290005/a''
                 phys_path=''/pci at 0,0/pci15d9,d280 at 1f,2/disk at
2,0:a''
                 whole_disk=0
                 DTL=72
         children[1]
                 type=''disk''
                 id=1
                 guid=3306076269030000850
                 path=''/dev/dsk/c5t1d0s0''
                 devid=''id1,sd at
SATA_____WDC_WD3200JD-00K_____WD-WCAMR2427509/a''
                 phys_path=''/pci at 0,0/pci15d9,d280 at 1f,2/disk at
1,0:a''
                 whole_disk=0
                 DTL=54

my question:

How can I import half a mirror?

37 % zpool import -f 10861732602511278403 newpool
cannot import ''r'' as ''newpool'': invalid vdev
configuration

Ok, so the choice for a MB boils down to:

- Intel desktop MB, no ECC support
- Intel server MB, ECC support, expensive (requires a Xeon for speedstep
support). It is a shame to waste top kit doing nothing 24/7.
- AMD K8: ECC support(right?), no Cool''n''quiet support (but
maybe still cool enough with the right CPU?)
- AMD K10: should have the best all of both worlds: ECC support,
Cool''n''quiet, cheap-ish and lowish-power CPU like Athlon II
250

Is my understanding correct? Like many I want reliable, cheap, low power, ECC
supporting MB. Integrated video and low power chipset would be best. The sata
ports will have to come from an additional controller it seems, but
that''s life.

Intel gear is best supported, but they shoot themselves (or is that that us?) in
the foot with their ECC-on-server MB policy.

AMD K10 seems the most tempting as it has it all. I wonder about solaris support
though. For example, is an AM3 MB OK with solaris?

I''d like this hopefully to work right away with opensolaris 2009.06,
without fiddling with drivers, I dont have much time or skills.

What AM3 MB do you guys know that is trouble free with solaris? 

If none, maybe top quality ram (suggestions?) would allow me to forego ECC and
use a well supported low power intel board (suggestions?) instead? and a E5200?

Thanks for your insight.
-- 
This message posted from opensolaris.org

On Mon, 20 Jul 2009, chris wrote:
>
> If none, maybe top quality ram (suggestions?) would allow me to 
> forego ECC and use a well supported low power intel board 
> (suggestions?) instead? and a E5200?
Even top quality RAM will not protect you from an alpha particle.

I would be surprised if the AMD K10 CPU caused any problem for 
Solaris.  The chipset used on the motherboard is probably what you 
should pay attention to.

Bob
--
Bob Friesenhahn
bfriesen at simple.dallas.tx.us, http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/

Thanks for your reply.
What if I wrap the ram in a sheet of lead?    ;-)
(hopefully the lead itself won''t be radioactive)

I found these 4 AM3 motherboard with "optional" ECC memory support. I
don''t know whether this means ECC works, or ECC memory can be used but
ECC will not. Do you?

Asus  M4N78 SE, Nvidia nForce 720D Chipset, 4xsata
Asus  M4N78-VM, Nvidia GeForce 8200 Chipset, 6xsata, onboard video
Asus  M4N82 Deluxe,  NVIDIA nForce 980a Chipset, 6xsata
Gigabyte  GA-MA770T-UD3P, AMD 770 Chipset, 6xsata

The 2nd one looks the most promising, and GeForce 8200 seems somewhat supported
by solaris except for sound(don''t care) and network (can add another
card. A workaround is described in
http://pegolon.wordpress.com/2009/07/13/setting-up-my-solaris-fileserver-part-1/
but i''d be clueless if anything goes wrong).

What do you think?
-- 
This message posted from opensolaris.org

>    hopefully the lead itself won''t be             >radioactive)
Or the chips themselves don''t have some alpha particle generation. It  
has happened and from premium vendors

There is no replacement for good system design :)

khbkhb at gmail.com
Sent from my iPod

chris wrote:
> Thanks for your reply.
> What if I wrap the ram in a sheet of lead?    ;-)
> (hopefully the lead itself won''t be radioactive)
>
>   
I''ve been looking at the same thing recently.
> I found these 4 AM3 motherboard with "optional" ECC memory
support. I don''t know whether this means ECC works, or ECC memory can
be used but ECC will not. Do you?
>
>   
That''s a good question. The ASUS specs definitely say unbuffered ECC 
memory is compatible, but until you mentioned it I never thought about 
whether the ECC functionality would actually be used.
> Asus  M4N78 SE, Nvidia nForce 720D Chipset, 4xsata
> Asus  M4N78-VM, Nvidia GeForce 8200 Chipset, 6xsata, onboard video
> Asus  M4N82 Deluxe,  NVIDIA nForce 980a Chipset, 6xsata
> Gigabyte  GA-MA770T-UD3P, AMD 770 Chipset, 6xsata
>   
I hadn''t located the Gigabyte board yet I''ll have to look at
that.

The ASUS boards with the AMD chipsets (the models that start with M4A - 
like the M4A79T) are all true AM3 boards - they take DDR3 memory. All 
the nVidia chipset boards (even the 980a one) are AM2+/AM3 boards, and 
(as far as I know) only take DDR2 memory, but that may not matter to you 
since this will only be a server for you. The chipset isn''t supposed to
dictate the memory type that up to the CPU, but the MB does need to 
support it in other ways.

DDR3 doesn''t appear (in any reviews I''ve seen) to give much
benefits
with the current processors anyway. What I find more discouraging (since 
I''m trying to build a desktop/workstation) is that when you go to look 
for RAM the only ECC memory available (doesn''t matter if it''s
DDR2 or 3)
is rated much slower than what is available for non-ECC. For example you 
can find DD2 at 1066mhz, or even 1200mhz, but the fasted ECC DDR2 you 
can get is 800mhz. - It''s cheap though, unless you want 4GB DIMMs then 
it''s outrageous!
> The 2nd one looks the most promising, and GeForce 8200 seems somewhat
supported by solaris except for sound(don''t care) and network (can add
another card.
I don''t see the the 1st or the  2nd one at usa.asus.com. The 3rd is the
one I''ve been considering hard lately. In my searching the other brands
don''t seem to support ECC memeory at all.

Another thing to remember is the expansion slots. You mentioned putting 
in a SATA controller for more drives, You''ll want to make sure the
board
has a slot that can handle the card you want. If you''re not using 
graphics then any board with a single PCI-E x16 slot should handle 
anything. But if you do put in a graphics board you''ll want to look at 
what other slots are available. Not many consumer boards have PCI-X 
slots, and only some have PCI-E x4 slots. PCI-E x1 slots are getting 
scarce too. Most of the PCI-E SATA controlers I''ve seen want a slot at 
least x4, and many are x8.

  -Kye

>Another thing to remember is the expansion slots. You mentioned putting
>in a SATA controller for more drives, You''ll want to make sure the
board
>has a slot that can handle the card you want. If you''re not using
>graphics then any board with a single PCI-E x16 slot should handle
>anything. But if you do put in a graphics board you''ll want to look
at
>what other slots are available. Not many consumer boards have PCI-X
>slots, and only some have PCI-E x4 slots. PCI-E x1 slots are getting
>scarce too. Most of the PCI-E SATA controlers I''ve seen want a slot
at
>least x4, and many are x8.
Better check that, almost *all* consumer boards that have 1 16 lane PCIe
slot can only have use a graphics card in that slot. I can confirm this
to be true and most Intel boards are that way and some Asus boards I have
used behave this way as well.

As far as ECC for a home system, I run two ESXi servers, an asterisk PBX,
a red hat iSCSI server etc etc all on commodity mobo''s without ECC and
have perfect uptime. I wouldn''t do this at work, but for the 1/2 dozen
people
that use it _at home_, it works perfectly.

jlc

Joseph L. Casale wrote:
>> Another thing to remember is the expansion slots. You mentioned putting
>> in a SATA controller for more drives, You''ll want to make sure
the board
>> has a slot that can handle the card you want. If you''re not
using
>> graphics then any board with a single PCI-E x16 slot should handle
>> anything. But if you do put in a graphics board you''ll want to
look at
>> what other slots are available. Not many consumer boards have PCI-X
>> slots, and only some have PCI-E x4 slots. PCI-E x1 slots are getting
>> scarce too. Most of the PCI-E SATA controlers I''ve seen want a
slot at
>> least x4, and many are x8.
>>     
>
> Better check that, almost *all* consumer boards that have 1 16 lane PCIe
> slot can only have use a graphics card in that slot. I can confirm this
> to be true and most Intel boards are that way and some Asus boards I have
> used behave this way as well.
>   
Really? That''s good to know.  Is that only for the first x16 slot? or 
(for MB that have them) all the x16 slots?
I wonder how it knows if it''s a video card... hmm.

That makes the prospects of adding more SATA ports even harder. All the 
8 port cards I''ve seen are either PCI-X or PCI-E, and many are 
physically x8 or more, even if they''ll run with fewer lanes.
> As far as ECC for a home system, I run two ESXi servers, an asterisk PBX,
> a red hat iSCSI server etc etc all on commodity mobo''s without ECC
and
> have perfect uptime. I wouldn''t do this at work, but for the 1/2
dozen people
> that use it _at home_, it works perfectly.
>   
I know it might be over kill. But I have other reasons to liek the MB, 
and if it can do ECC, and the memory doesn''t cost more (and with 2GB 
DIMMs it doesn''t) I figured why not. Though it is slower. Everything is
a trade off.

  -Kyle
> jlc
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss
>

Regarding the SATA card and the mainboard slots, make sure that
whatever you get is compatible with the OS. In my case I chose
OpenSolaris which lacks support for Promise SATA cards. As a result,
my choices were very limited since I had chosen a Chenbro ES34069 case
and Intel Little Falls 2 mainboard. Basically I had to go with the
SYBA Sil3124 card and a flexible PCI adapter. More details here:
http://cafenate.wordpress.com/2009/07/13/building-a-nas-box/

No ECC memory, but I don''t mind because the case has a great form
factor and hot swappable drive bays. If I could find a low power board
that supported ECC and OpenSolaris, I''d consider switching.

Good luck.

n

On Tue, Jul 21, 2009 at 4:20 PM, chris <no-reply at opensolaris.org>
wrote:
> Thanks for your reply.
> What if I wrap the ram in a sheet of lead?    ;-)
> (hopefully the lead itself won''t be radioactive)
>
> I found these 4 AM3 motherboard with "optional" ECC memory
support. I don''t
> know whether this means ECC works, or ECC memory can be used but ECC will
> not. Do you?
>
Often this means, ECC memory will work but the ECC aspect will not work. So
the memory is usable, but not as you expect.

Nicholas
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090722/6f51a82d/attachment.html>

The i7 and Xeon 3300 m/b that say they have ECC support have exactly this
problem as well.

On Wed, Jul 22, 2009 at 4:53 PM, Nicholas Lee <emptysands at gmail.com>
wrote:
>
>
> On Tue, Jul 21, 2009 at 4:20 PM, chris <no-reply at opensolaris.org>
wrote:
>
>> Thanks for your reply.
>> What if I wrap the ram in a sheet of lead?    ;-)
>> (hopefully the lead itself won''t be radioactive)
>>
>> I found these 4 AM3 motherboard with "optional" ECC memory
support. I
>> don''t know whether this means ECC works, or ECC memory can be
used but ECC
>> will not. Do you?
>>
>
> Often this means, ECC memory will work but the ECC aspect will not work. So
> the memory is usable, but not as you expect.
>
> Nicholas
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090722/f0fd0c04/attachment.html>

i7 doesn''t support ECC even motherboard supports it, you need XEON
W3500 which costs the same as i7 to support ECC.
-- 
This message posted from opensolaris.org

Good news; the manual for the M4N78-VM mentions ECC and gives the following BIOS
options: disabled/basic/good/super/maxi/user.

Unsure what these mean but that''s a start.
-- 
This message posted from opensolaris.org

Found this:

ECC Mode [Disabled]
Disables or sets the DRAM ECC mode that allows the hardware to report and
correct memory errors. Set this item to [Basic] [Good] or [Max] to allow ECC
mode
auto-adjustment. Set this item to [Super] to adjust the DRAM BG Scrub sub-item
manually. You may also adjust all sub-items by setting this item to [User].
Configuration options: [Disabled] [Basic] [Good] [Super] [Max] [User]

I would have thought the checksum was either good or not. Apparently
it''s not so simple. Now about that unique PCIe-16 slot?
-- 
This message posted from opensolaris.org

Hi,

I''m using asus m3a78 boards (with the sb700) for opensolaris and m2a*
boards (with the sb600) for linux some of them with 4*1GB and others with 4*2Gb
ECC memory. Ecc faults will be detected and reported. I tested it with a small
tungsten light. By moving the light source slowly towards the memory banks
you''ll heat them up in a controlled way and at a certain point bit
flips will occur.
I recommend you to go for a m4a board since they support up to 16 GB. 
I don''t know if you can run opensolaris without a videocard after
installation I think you can disable the "halt on no video card" in
the bios. But Simon Breden had some trouble with it, see his homeserver blog.
But you can go for one of the three m4a boards with a 780g onboard. Those will
give you 2 pci-e x16 connectors. I don''t think the onboard nic is
supported. I always put an intel (e1000) in, just to prevent any trouble. I
don''t have any trouble with the sb700 in ahci mode. Hotplugging works
like a charm. Transfering a couple of GB''s over esata takes
considerable less time than via usb.
I have a pata to dual cf adapter and two industrial 16gb cf cards as mirrored
root pool. It takes for ever to install nevada, at least 14 hours. I suspect the
cf cards lack caches. But I don''t update that regularly, still on
snv104.  And have 2 mirrors and a hot spare. The sixth port is an esata port I
use to transfer large amounts of data. This system consumes about 73 watts idle
and 82 under load i/o load. (5 disks , a separate nic  ,8 gb ram and a be2400
all using just 73 watts!!!)
Please note that frequency scaling is only supported on the K10 architecture.
But don''t expect to much power saving from it. A lower voltage yields
far greater savings than a lower frequency.
In september I''ll do a post about the afore mentioned M4A boards and an
lsi sas controller in one of the pcie x16 slots.
-- 
This message posted from opensolaris.org

On Jul 23, 2009, at 5:42 AM, F. Wessels wrote:
> Hi,
>
> I''m using asus m3a78 boards (with the sb700) for opensolaris and  
> m2a* boards (with the sb600) for linux some of them with 4*1GB and  
> others with 4*2Gb ECC memory. Ecc faults will be detected and  
> reported. I tested it with a small tungsten light. By moving the  
> light source slowly towards the memory banks you''ll heat them up
in
> a controlled way and at a certain point bit flips will occur.
I am impressed!  I don''t know very many people interested in inducing
errors in their garage.  This is an excellent way to demonstrate random
DRAM errors. Well done!
> I recommend you to go for a m4a board since they support up to 16 GB.
> I don''t know if you can run opensolaris without a videocard after
> installation I think you can disable the "halt on no video card"
in
> the bios. But Simon Breden had some trouble with it, see his  
> homeserver blog. But you can go for one of the three m4a boards with  
> a 780g onboard. Those will give you 2 pci-e x16 connectors. I
don''t
> think the onboard nic is supported. I always put an intel (e1000)  
> in, just to prevent any trouble. I don''t have any trouble with the
> sb700 in ahci mode. Hotplugging works like a charm. Transfering a  
> couple of GB''s over esata takes considerable less time than via
usb.
> I have a pata to dual cf adapter and two industrial 16gb cf cards as  
> mirrored root pool. It takes for ever to install nevada, at least 14  
> hours. I suspect the cf cards lack caches. But I don''t update that
> regularly, still on snv104.  And have 2 mirrors and a hot spare. The  
> sixth port is an esata port I use to transfer large amounts of data.  
> This system consumes about 73 watts idle and 82 under load i/o load.  
> (5 disks , a separate nic  ,8 gb ram and a be2400 all using just 73  
> watts!!!)
How much power does the tungsten light burn? :-)
  -- richard

On 07/23/09 09:19 AM, Richard Elling wrote:
> On Jul 23, 2009, at 5:42 AM, F. Wessels wrote:
>
>> Hi,
>>
>> I''m using asus m3a78 boards (with the sb700) for opensolaris
and m2a*
>> boards (with the sb600) for linux some of them with 4*1GB and others 
>> with 4*2Gb ECC memory. Ecc faults will be detected and reported. I 
>> tested it with a small tungsten light. By moving the light source 
>> slowly towards the memory banks you''ll heat them up in a
controlled
>> way and at a certain point bit flips will occur.
>
> I am impressed!  I don''t know very many people interested in
inducing
> errors in their garage.  This is an excellent way to demonstrate random
> DRAM errors. Well done!
>
>> I recommend you to go for a m4a board since they support up to 16 GB.
>> I don''t know if you can run opensolaris without a videocard
after
>> installation I think you can disable the "halt on no video
card" in
>> the bios. But Simon Breden had some trouble with it, see his 
>> homeserver blog. But you can go for one of the three m4a boards with 
>> a 780g onboard. Those will give you 2 pci-e x16 connectors. I
don''t
>> think the onboard nic is supported. 

What is the specific model of the onboard nic chip?
We may be working on it right now.

Neal

>> I always put an intel (e1000) in, just to prevent any trouble. I 
>> don''t have any trouble with the sb700 in ahci mode.
Hotplugging works
>> like a charm. Transfering a couple of GB''s over esata takes 
>> considerable less time than via usb.
>> I have a pata to dual cf adapter and two industrial 16gb cf cards as 
>> mirrored root pool. It takes for ever to install nevada, at least 14 
>> hours. I suspect the cf cards lack caches. But I don''t update
that
>> regularly, still on snv104.  And have 2 mirrors and a hot spare. The 
>> sixth port is an esata port I use to transfer large amounts of data. 
>> This system consumes about 73 watts idle and 82 under load i/o load. 
>> (5 disks , a separate nic  ,8 gb ram and a be2400 all using just 73 
>> watts!!!)
>
> How much power does the tungsten light burn? :-)
>  -- richard
>
> _______________________________________________
> zfs-discuss mailing list
> zfs-discuss at opensolaris.org
> http://mail.opensolaris.org/mailman/listinfo/zfs-discuss

Thanks for this, good news! 

Yes, I would try to use onboard video.
> Please note that frequency scaling is only supported
> on the K10 architecture. But don''t expect to much
> power saving from it. A lower voltage yields far
> greater savings than a lower frequency.
Doesn''t Cool''n''quiet step the voltage as well? 
An Athlon X2 5050e and an Athlon II X2 250 are the same price.
The former has a TDP of 45W, while the latter is 65W.
But the 250 uses 45nm technology and the K10 architecture, so I would hope that
its power consumption at idle would be lower. Would you agree?

Also, out of interest, do you know what the ECC BIOS modes mean?
-- 
This message posted from opensolaris.org

The Asus M4N78-VM uses a Nvidia GeForce 8200 Chipset (This board only has 1
PCIe-16 slot though, I should look at those that have 2 slots).
-- 
This message posted from opensolaris.org

Oh, and another unrelated question: 

Would I better off using OpenSolaris or Solaris Community Edition? 

I suspect SCE has more drivers (though mayby in a more beta state?), but its
huge download size (several days in backward New Zealand, thanks Telecom NZ!)
means I would only try if there is good justification.
What would you guys recommend (I know, this is an OpenSolaris forum, but at
least can you tell me how these 2 differ)?
-- 
This message posted from opensolaris.org

>>>>> "c" == chris  <no-reply at opensolaris.org>
writes:
     c> do you know what the ECC BIOS modes mean?

It''s about the hardware scrubbing feature I mentioned.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 304 bytes
Desc: not available
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090723/7ff6bb8e/attachment.bin>

I''m going the other route here, and using a Intel small server
motherboard.

I''m currently trying the Supermicro X7SBE, which supports a non-Xeon
CPU, and _should_ actually use the (unbuffered) ECC RAM I have in it.
It can also support a network KVM IPMI board, which is nice (though not
cheap - i.e. $100 or so). 


The Supermicro X7SBL-LN[12] boards also look good, though they won''t
support the network KVM option.


--
Erik Trimble
Java System Support
Mailstop:  usca22-123
Phone:  x17195
Santa Clara, CA
Timezone: US/Pacific (GMT-0800)

chris wrote:
> Ok, so the choice for a MB boils down to:
>
> - Intel desktop MB, no ECC support
>   
This is mostly true.  The exceptions are some implementations of the 
Socket T LGA 775 (i.e. late Pentium 4 series, and Core 2) D975X and X38 
chipsets, and possibly some X48 boards as well.  Intel''s other desktop 
chipsets do not support ECC.  Some motherboard examples include:

Intel DX38BT - ECC support is mentioned in the documentation and is a 
BIOS option
Gigabyte GA-X38-DS4, GA-EX38-DS4 - ECC support is mentioned in the 
documentation and is listed in the website FAQ
The Sun Ultra 24 also uses the X38 chipset.

It''s not clear how well ECC support has actually been implemented on
the
Intel and Gigabyte boards, i.e. whether it is simply unbuffered ECC 
memory compatible, or actually able to initialize and use the ECC 
capability.  I mentioned the X48 chipset above because discussions 
surrounding it say it is just a higher binned X38 chip.

On Linux, the EDAC project maintains software to manage the 
motherboard''s ECC capability.  A list of memory controllers currently 
supported by Linux EDAC is here:
http://buttersideup.com/edacwiki/Main_Page

A prior discussion thread in ''fm'' titled ''X38/975x
ECC memory support''
is here:
http://opensolaris.org/jive/thread.jspa?threadID=52440&tstart=60

Thread links:
http://www.madore.org/~david/linux/#ECC_for_82x
http://developmentonsolaris.wordpress.com/2008/03/12/intel-82975x-mch-and-logging-of-ecc-events-on-solaris/

Note that the ''ecccheck.pl'' script depends on the
''pcitweak'' utility
which is no longer present in OpenSolaris 2009.06 and Ubuntu 8.10 
because of Xorg changes.  One Linux user needing the utility copied it 
from another distro.  The version of pcitweak included with previous 
versions of OpenSolaris might work on 2009.06.
http://opensolaris.org/jive/thread.jspa?threadID=105975&tstart=90
http://ubuntuforums.org/showthread.php?t=1054516

Finally, on unbuffered ECC memory prices and speeds...they are a bit 
behind in price and speed vs. regular unbuffered RAM, but both are still 
reasonable.  Keep When comparing prices, remember that ECC RAM uses 9 
chips where non-ECC uses 8, so expect at least a 12.5% price increase.  
Consider:

DDR2: $64 for Crucial 4GB kit (2GBx2), 240-pin DIMM, Unbuffered DDR2 
PC2-6400 memory module
http://www.crucial.com/store/partspecs.aspx?IMODULE=CT2KIT25672AA800

DDR3: $108 for Crucial 6GB (3 x 2GB) 240-Pin DDR3 SDRAM ECC Unbuffered 
DDR3 1333 (PC3 10600) Triple Channel Kit Server Memory Model 
CT3KIT25672BA1339 - Retail
http://www.newegg.com/Product/Product.aspx?Item=N82E16820148259

-hk
> - Intel server MB, ECC support, expensive (requires a Xeon for speedstep
support). It is a shame to waste top kit doing nothing 24/7.
> - AMD K8: ECC support(right?), no Cool''n''quiet support
(but maybe still cool enough with the right CPU?)
> - AMD K10: should have the best all of both worlds: ECC support,
Cool''n''quiet, cheap-ish and lowish-power CPU like Athlon II
250
>
> Is my understanding correct? Like many I want reliable, cheap, low power,
ECC supporting MB. Integrated video and low power chipset would be best. The
sata ports will have to come from an additional controller it seems, but
that''s life.
>
> Intel gear is best supported, but they shoot themselves (or is that that
us?) in the foot with their ECC-on-server MB policy.
>
> AMD K10 seems the most tempting as it has it all. I wonder about solaris
support though. For example, is an AM3 MB OK with solaris?
>
> I''d like this hopefully to work right away with opensolaris
2009.06, without fiddling with drivers, I dont have much time or skills.
>
> What AM3 MB do you guys know that is trouble free with solaris? 
>
> If none, maybe top quality ram (suggestions?) would allow me to forego ECC
and use a well supported low power intel board (suggestions?) instead? and a
E5200?
>
> Thanks for your insight.
>

Cheers Miles, and thanks also for the tip to look in the BIOS options to see if
ECC is actually used.
Which mode woud you use? Max seems the most appealing, why would anyone use
something called basic? But there must be a catch if they provided several  ECC
support modes.

I am glad this thread seems to be going somewhere with lots of valuable
contributions  =:^)
-- 
This message posted from opensolaris.org

More choice is good!

It seems Intel''s server boards sometimes accept desktop CPUS, but
don''t support speedstep. Is all OK with those?
-- 
This message posted from opensolaris.org

>Note that the ''ecccheck.pl'' script depends on the
''pcitweak'' utility
>which is no longer present in OpenSolaris 2009.06 and Ubuntu 8.10 
>because of Xorg changes. 
This is exactly the kind of hidden trap I fear. One does everything right and
then discovers that xx is missing or has been changed or depends on yy or
doesn''t work with zz. And that discovery comes after hours/days/weeks
of trying to find out why something misbehaves. Thanks for the heads up!
2008.11 would be a safer bet then? Or Solaris CE?
-- 
This message posted from opensolaris.org

>>>>> "c" == chris  <no-reply at opensolaris.org>
writes:
>>>>> "hk" == Haudy Kazemi <kaze0010 at umn.edu>
writes:
     c> why would anyone use something called basic? But there must be
     c> a catch if they provided several ECC support modes.

They are just taiwanese.  They have no clue wtf they are doing and do
not care about quality since their customers don''t have any memory
past six months, and anyway if they get a bad reputation they''ll just
sell the same crap under a different brand name.  They are just trying
to ship kit for gamers as quickly as possible.  They view the design
of their box and website and little bubbly slogan blobs as their key
distinguishing asset over their competitors, not what''s inside.  Do
not try to reason with these people who have zero respect for you, and
certainly do not trust them to do something ``reasonable'''' and
then
try to work out what they''ve not documented based on blind faith in an
orderly world with competent stewardship.

Read jaakko''s script that I posted and set the timer to scrub the
amount of memory you have about once a day.  The script will give you
status, showing the current address being scrubbed, so you can watch
the rate at which the counter increases, and also watch the counter
wrap around to determine the number of zeroes it has hacked off from
the size of your memory.  With a couple observations spaced 10min
apart, plus a series of observations each spaced 4 hours apart, you
can convert the microseconds you feed to the script into
hours-per-complete-pass and accomplish this.  If you really care that
much.  I didn''t---just made sure it wasn''t crazy.

It isn''t really that important anyway---you should not think about it
so much.  I jsut blundered through it.  I''m spending more time writign
about it than doing it.  it is just a bunch of toy knobs for you to
play with.  The important thing is, will it actually correct errors?
will the OS count the errors?  will it localize them to a DIMM?  since
the L2 caches are 10x the size they used to be and etched
smaller/more-sensitive will ECC also work on the on-chip cache and get
counted and reported distinct from DIMM''s, or not?  All of these are
more important than whether it does any scrubbing at all, much less
the specific timing of the scrubbing.

I bought four different boards on purpose to get a cross-section of
crappyness, and the arena is incredibly stabby.  There is all kinds of
stuff in the BSoS like DIMM powerdown and C1E support that probably
doesn''t work at all.  One of these nvidia-northbridge boards, the
audio controller showed up on a different interrupt every time I
booted it.  Some other board, you needed an actual physical floppy
disk to update the BIOS---pxeboot/memdisk just froze, even though it
works for everything else I''ve tried.  Who even HAS a floppy disk?
Don''t get distracted playing with their stupid fisher price knobs like
an overclocker.  Just flip the ECC thing on and forget about it.  What
jaakko demonstrates is that support for AMD ECC probably belongs in
the OS or bootloader anyway, since it''s entirely in the CPU and not
integrator-specific and the people who write OS''s are less idiotic
than these ship-and-forget BIOS people and besides the error reporting
needs to be in the OS anyway, which is a more complicated piece, so
relying on someone else to ``turn it on'''' for you when turning
it on
boils down to a couple setpci lines in a shell script is completely
retarded.

    hk> DDR3: $108 for Crucial 6GB (3 x 2GB) 240-Pin DDR3 SDRAM ECC

yeah, but for 4GB parts it''s not a 12% difference any more. 
It''s raep.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 304 bytes
Desc: not available
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090724/031e6356/attachment.bin>

Hi,

thank you so much for this post. This is exactly what I was looking for.
I''ve been eyeing the M3A76-CM board, but will now look at 78 and M4A as
well.

Actually, not that many Asus M3A, let alone M4A boards show up yet on the
OpenSolaris HCL, so I''d like to encourage everyone to share their
hardware
experience by clicking on the "submit hardware" link on:

   http://www.sun.com/bigadmin/hcl/data/os/

I''ve done it a couple of times and it''s really just a matter
of 5-10 minutes
where you can help others know if a certain component works or not or if a
special driver or /etc/driver_aliases setting is required.

I''m also interested in getting the power down. Right now, I have the
Athlon X2 5050e (45W TDP) on my list, but I''d also like to know more
about
the possibilities of the Athlon II X2 250 and whether it has better potential
for power savings.

Neal, the M3A78 seems to have a RealTek RTL8111/8168B NIC chip. I pulled
this off a Gentoo Wiki, because strangely this information doesn''t show
up
on the Asus website.

Also, thanks for the CF to pata hint for the root pool mirror. Will try to
find fast CFs to boot from. The performance problems you see when writing
may be related to master/slave issues, but I''m not a good PC tweaker to
back
that up.

Cheers,
    Constantin


F. Wessels wrote:
> Hi,
> 
> I''m using asus m3a78 boards (with the sb700) for opensolaris and
m2a* boards
> (with the sb600) for linux some of them with 4*1GB and others with 4*2Gb
ECC
> memory. Ecc faults will be detected and reported. I tested it with a small
> tungsten light. By moving the light source slowly towards the memory banks
> you''ll heat them up in a controlled way and at a certain point bit
flips will
> occur. I recommend you to go for a m4a board since they support up to 16
GB.
>  I don''t know if you can run opensolaris without a videocard after
> installation I think you can disable the "halt on no video card"
in the bios.
> But Simon Breden had some trouble with it, see his homeserver blog. But you
> can go for one of the three m4a boards with a 780g onboard. Those will give
> you 2 pci-e x16 connectors. I don''t think the onboard nic is
supported. I
> always put an intel (e1000) in, just to prevent any trouble. I
don''t have any
> trouble with the sb700 in ahci mode. Hotplugging works like a charm.
> Transfering a couple of GB''s over esata takes considerable less
time than via
> usb. I have a pata to dual cf adapter and two industrial 16gb cf cards as
> mirrored root pool. It takes for ever to install nevada, at least 14 hours.
I
> suspect the cf cards lack caches. But I don''t update that
regularly, still on
> snv104.  And have 2 mirrors and a hot spare. The sixth port is an esata
port
> I use to transfer large amounts of data. This system consumes about 73
watts
> idle and 82 under load i/o load. (5 disks , a separate nic  ,8 gb ram and a
> be2400 all using just 73 watts!!!) Please note that frequency scaling is
only
> supported on the K10 architecture. But don''t expect to much power
saving from
> it. A lower voltage yields far greater savings than a lower frequency. In
> september I''ll do a post about the afore mentioned M4A boards and
an lsi sas
> controller in one of the pcie x16 slots.
-- 
Constantin Gonzalez                              Sun Microsystems GmbH, Germany
Principal Field Technologist                    http://blogs.sun.com/constantin
Tel.: +49 89/4 60 08-25 91       http://google.com/search?q=constantin+gonzalez

Sitz d. Ges.: Sun Microsystems GmbH, Sonnenallee 1, 85551 Kirchheim-Heimstetten
Amtsgericht Muenchen: HRB 161028
Geschaeftsfuehrer: Thomas Schroeder, Wolfgang Engels, Wolf Frenkel
Vorsitzender des Aufsichtsrates: Martin Haering

Ok, i am ready to try.

2 last questions before I go for it:
- which version of (open)solaris for Ecc support (which seems to have been
dropped from 200906) and general as-few-headaches-as-possible installation?

- do you think this issue with the AMD Athlon II X2 250 
http://www.anandtech.com/cpuchipsets/showdoc.aspx?i=3572&p=2&cp=4
would affect cool''n''quiet support in solaris?

thx for your insight.
-- 
This message posted from opensolaris.org

Hello,
your "(open)solaris for Ecc support (which seems to have been dropped from
200906)" is misunderstanding. OS 2009.06 also supports ECC as 2005 did.
Just install it and use my updated ecccheck.pl script to get informed about
errors. Also you might verify that Solaris'' memory scrubber is really
running if you are that curious:
http://developmentonsolaris.wordpress.com/2009/03/06/how-to-make-sure-memory-scrubber-is-running/
Karel
-- 
This message posted from opensolaris.org

On Thu, Sep 3, 2009 at 4:57 AM, Karel Gardas <karel.gardas at
centrum.cz>wrote:
> Hello,
> your "(open)solaris for Ecc support (which seems to have been dropped
from
> 200906)" is misunderstanding. OS 2009.06 also supports ECC as 2005
did. Just
> install it and use my updated ecccheck.pl script to get informed about
> errors. Also you might verify that Solaris'' memory scrubber is
really
> running if you are that curious:
>
http://developmentonsolaris.wordpress.com/2009/03/06/how-to-make-sure-memory-scrubber-is-running/
> Karel
> --
>

Is there something that needs to be done on the solaris side for memscrub
scans to occur?  I''m running snv_118, with a supermicro board running
ECC
memory and amd opteron CPU''s.  It would appear it''s doing a
lot of nothing.

Aug  8 03:56:23 fserv unix: [ID 950921 kern.info] cpu0: x86 (chipid 0x0
AuthenticAMD 40F13 family 15 model 65 step 3 clock 2010 MHz)
Aug  8 03:56:23 fserv unix: [ID 950921 kern.info] cpu0: Dual-Core AMD
Opteron(tm) Processor 2212

root at fserv:~# isainfo -v
64-bit amd64 applications
        tscp ahf cx16 sse3 sse2 sse fxsr amd_3dnowx amd_3dnow amd_mmx mmx
cmov
        amd_sysc cx8 tsc fpu
32-bit i386 applications
        tscp ahf cx16 sse3 sse2 sse fxsr amd_3dnowx amd_3dnow amd_mmx mmx
cmov
        amd_sysc cx8 tsc fpu




root at fserv:~# echo "memscrub_scans_done/U" | mdb -k
memscrub_scans_done:
memscrub_scans_done:            0
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090904/d6b8f3b7/attachment.html>

On Fri, Sep 4, 2009 at 7:14 PM, Tim Cook<tim at cook.ms>
wrote:
> Is there something that needs to be done on the solaris side for memscrub
> scans to occur?? I''m running snv_118, with a supermicro board
running ECC
> memory and amd opteron CPU''s.? It would appear it''s doing
a lot of nothing.
My AMD board from ASUS has a BIOS option to scrub memory, outside of
the OS. Check that?

-B

-- 
Brandon High : bhigh at freaks.com

What''s your uptime? Usually it scrubs memory during the idle time and
usually waits quite a long nearly till the deadline -- which is IIRC 12 hours.
So do you have more than 12 hours of uptime?
-- 
This message posted from opensolaris.org

On Mon, Sep 7, 2009 at 2:01 AM, Karel Gardas <karel.gardas at
centrum.cz>wrote:
> What''s your uptime? Usually it scrubs memory during the idle time
and
> usually waits quite a long nearly till the deadline -- which is IIRC 12
> hours. So do you have more than 12 hours of uptime?
> --
>

 10:43am  up 30 days  6:47,  1 user,  load average: 0.00, 0.00, 0.00
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://mail.opensolaris.org/pipermail/zfs-discuss/attachments/20090907/0b60b474/attachment.html>