Recommended pci-e 1x SATA cards.

Recommended pci-e 1x SATA cards.

[Steven Haigh]

I'm looking at adding a couple of SATA ports to my home server and 
figure its probably better to get a 1x pci-e card instead of using the 
PCI bus.

Has anyone got any recommendations on cheap pci-e SATA cards that work 
well with linux?

Any comments on these?
http://cgi.ebay.com.au/PCI-E-PCIE-RAID-Express-Card-SATA-II-Controller-A24-/220762966195?pt=AU_Computer_Components_Controller_Cards&hash=item33667f88b3

http://cgi.ebay.com.au/PCI-E-PCIE-RAID-Express-Card-2-SATA-II-2-eSATA-A25-/300543913241?pt=AU_Components&hash=item45f9d02d19

This is a jMicron adapter - but I haven't heard much good or bad about them.
http://cgi.ebay.com.au/PCI-E-PCIE-Express-Card-2-SATA-RAID-II-Controller-/260745100434?pt=AU_Computer_Components_Controller_Cards&hash=item3cb59e7c92

Any others that people recommend?

-- 
Steven Haigh

Email: netwiz@crc.id.au
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
Fax: (03) 8338 0299

There are no great men, only great challenges that ordinary men are 
forced by circumstances to meet. -- Admiral William Halsey

Re: Recommended pci-e 1x SATA cards.

[Roman Mamedov]

[-- Attachment #1: Type: text/plain, Size: 1993 bytes --]

On Thu, 14 Apr 2011 20:31:26 +1000
Steven Haigh <netwiz@crc.id.au> wrote:

> I'm looking at adding a couple of SATA ports to my home server and 
> figure its probably better to get a 1x pci-e card instead of using the 
> PCI bus.
> 
> Has anyone got any recommendations on cheap pci-e SATA cards that work 
> well with linux?
> 
> Any comments on these?
> http://cgi.ebay.com.au/PCI-E-PCIE-RAID-Express-Card-SATA-II-Controller-A24-/220762966195?pt=AU_Computer_Components_Controller_Cards&hash=item33667f88b3
> 
> http://cgi.ebay.com.au/PCI-E-PCIE-RAID-Express-Card-2-SATA-II-2-eSATA-A25-/300543913241?pt=AU_Components&hash=item45f9d02d19
> 
> This is a jMicron adapter - but I haven't heard much good or bad about them.
> http://cgi.ebay.com.au/PCI-E-PCIE-Express-Card-2-SATA-RAID-II-Controller-/260745100434?pt=AU_Computer_Components_Controller_Cards&hash=item3cb59e7c92
> 
> Any others that people recommend?
> 

JMicron is okay, it's an AHCI compatible controller, so does not even require
its own driver module, it just uses 'ahci'. Keep in mind that it is not a
PCI-E 2.0 device, and the transfer speed will be limited to 250 MB/sec over
both ports (and because of various overheads, something closer to 200 MB/sec).
A couple of modern disks can easily reach and become limited by that.

I suggest that you avoid Silicon Image 3132, they have data corruption issue
(on some board designs?), triggered or amplified by transferring via both
ports at the same time, at full speed.

If you can spend a bit more, try finding a controller based on Marvell 9123,
which is a 2-port SATA3 chip for PCI-E 2.0. For example "ASRock SATA3 Card"
which is sold separately (from motherboards) for about $33.
https://encrypted.google.com/search?hl=en&source=hp&q=ASRock%20SATA3%20Card
The only downside of that device, is that it doesn't seem to be compatible
with PCI-E 1.0, it is 1.1/2.0 only, so will not work on many old motherboards.

-- 
With respect,
Roman

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Recommended pci-e 1x SATA cards.

[Tapani Tarvainen]

On Thu, Apr 14, 2011 at 04:54:33PM +0600, Roman Mamedov (rm@romanrm.ru) wrote:

> I suggest that you avoid Silicon Image 3132, they have data corruption issue
> (on some board designs?), triggered or amplified by transferring via both
> ports at the same time, at full speed.

I'd appreciate more details or references on that.

I've got two Sil3132 cards on my home server, with RAID1 setup
across them, and they've never given me any trouble whatsoever.

-- 
Tapani Tarvainen

Re: Recommended pci-e 1x SATA cards.

[Roman Mamedov]

[-- Attachment #1: Type: text/plain, Size: 539 bytes --]

On Thu, 14 Apr 2011 14:05:59 +0300
Tapani Tarvainen <raid@tapanitarvainen.fi> wrote:

> On Thu, Apr 14, 2011 at 04:54:33PM +0600, Roman Mamedov (rm@romanrm.ru)
> wrote:
> 
> > I suggest that you avoid Silicon Image 3132, they have data corruption
> > issue (on some board designs?), triggered or amplified by transferring via
> > both ports at the same time, at full speed.
> 
> I'd appreciate more details or references on that.

See this thread: http://comments.gmane.org/gmane.linux.raid/30629

-- 
With respect,
Roman

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Recommended pci-e 1x SATA cards.

[Tapani Tarvainen]

On Thu, Apr 14, 2011 at 05:25:28PM +0600, Roman Mamedov (rm@romanrm.ru) wrote:

> > > I suggest that you avoid Silicon Image 3132, they have data corruption
> > > issue (on some board designs?), triggered or amplified by transferring via
> > > both ports at the same time, at full speed.

> See this thread: http://comments.gmane.org/gmane.linux.raid/30629

Thanks! Based on that, it'd seem it's not so much Sil3132 per se,
but some (too cheaply?) boards built around it, and the obvious
conclusion is that they should only be bought with plan to
test them before real use and return flakey ones.

I _have_ had similar data corruption problems with one Sil3124
card, none with two more, and also none with a 3114 card
(in heavy use for several years by now).

On the other hand I've had similar trouble with a JMicron card
and a Marvell-based card. So planning to test and return
bad cards is probably a good idea with them as well.

-- 
Tapani Tarvainen

Re: Recommended pci-e 1x SATA cards.

[Roman Mamedov]

[-- Attachment #1: Type: text/plain, Size: 2200 bytes --]

On Thu, 14 Apr 2011 14:41:18 +0300
Tapani Tarvainen <raid@tapanitarvainen.fi> wrote:

> On Thu, Apr 14, 2011 at 05:25:28PM +0600, Roman Mamedov (rm@romanrm.ru)
> wrote:
> 
> > > > I suggest that you avoid Silicon Image 3132, they have data corruption
> > > > issue (on some board designs?), triggered or amplified by transferring
> > > > via both ports at the same time, at full speed.
> 
> > See this thread: http://comments.gmane.org/gmane.linux.raid/30629
> 
> Thanks! Based on that, it'd seem it's not so much Sil3132 per se,
> but some (too cheaply?) boards built around it

Maybe, but in my opinion a chip design should be 'well-rugged' against this
kind of board-design blunders, under no circumstance it should pass silent
corruption of user data just like that. There are checksums built in into the
SATA protocol, and parity in PCI-E, so the data path itself is protected quite
well against a flaky trace or solder or whatever. Silicon Image, on the other
hand, had multiple data corruption reports with 3112/3114/3124 in the past,
and many of those were fixable by turning off PCI performance optimizations,
i.e. likely logic-level chip design problems. So I suspect 3132 has some less
than ideal design decisions as well. And maybe some of those can be worked
around by adding more external components (e.g. bigger capacitors etc), which
are not present on cheaper board designs. So to me this seems to be a split
blame, both the board designers and the chip designer are at fault.

> and the obvious conclusion is that they should only be bought with plan to
> test them before real use and return flakey ones.

That's a good idea at all times.

> On the other hand I've had similar trouble with a JMicron card
> and a Marvell-based card. So planning to test and return
> bad cards is probably a good idea with them as well.

I have one semi-bad JMicron based card - its problem is manifesting itself as
one SATA port at first throwing a lot of CRC errors then switching itself down
to 1.5GBps from 3.0GBps, and continuing to work perfectly at the slower speed.
To me this is an example of error-handling done right.

-- 
With respect,
Roman

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Recommended pci-e 1x SATA cards.

[Steven Haigh]

On 14/04/2011 8:54 PM, Roman Mamedov wrote:
> On Thu, 14 Apr 2011 20:31:26 +1000
> Steven Haigh<netwiz@crc.id.au>  wrote:
>
>> I'm looking at adding a couple of SATA ports to my home server and
>> figure its probably better to get a 1x pci-e card instead of using the
>> PCI bus.
>>
>> Has anyone got any recommendations on cheap pci-e SATA cards that work
>> well with linux?
>>
>> Any comments on these?
>> http://cgi.ebay.com.au/PCI-E-PCIE-RAID-Express-Card-SATA-II-Controller-A24-/220762966195?pt=AU_Computer_Components_Controller_Cards&hash=item33667f88b3
>>
>> http://cgi.ebay.com.au/PCI-E-PCIE-RAID-Express-Card-2-SATA-II-2-eSATA-A25-/300543913241?pt=AU_Components&hash=item45f9d02d19
>>
>> This is a jMicron adapter - but I haven't heard much good or bad about them.
>> http://cgi.ebay.com.au/PCI-E-PCIE-Express-Card-2-SATA-RAID-II-Controller-/260745100434?pt=AU_Computer_Components_Controller_Cards&hash=item3cb59e7c92
>>
>> Any others that people recommend?
>>
>
> JMicron is okay, it's an AHCI compatible controller, so does not even require
> its own driver module, it just uses 'ahci'. Keep in mind that it is not a
> PCI-E 2.0 device, and the transfer speed will be limited to 250 MB/sec over
> both ports (and because of various overheads, something closer to 200 MB/sec).
> A couple of modern disks can easily reach and become limited by that.

Hmmm - this might be an idea... I have two 1x slots on the mainboard - 
and I have 8 x hotswap bays. The current setup has 5 drives (2 x 80Gb 
RAID1, 3 x 1Tb RAID5) plus one eSATA drive off the mainboard. Speed is 
great, but I can't see myself getting more than 8 drives in this unit - 
ever.

> I suggest that you avoid Silicon Image 3132, they have data corruption issue
> (on some board designs?), triggered or amplified by transferring via both
> ports at the same time, at full speed.

Thanks for the pointer!

> If you can spend a bit more, try finding a controller based on Marvell 9123,
> which is a 2-port SATA3 chip for PCI-E 2.0. For example "ASRock SATA3 Card"
> which is sold separately (from motherboards) for about $33.
> https://encrypted.google.com/search?hl=en&source=hp&q=ASRock%20SATA3%20Card
> The only downside of that device, is that it doesn't seem to be compatible
> with PCI-E 1.0, it is 1.1/2.0 only, so will not work on many old motherboards.

Hmmm - I tried looking for these anywhere in Australia and came up 
empty. ebay is probably my preferred source for any card - as a lot of 
the time stuff like this is just not available in Australia at any kind 
of decent cost.

--
Steven Haigh

Email: netwiz@crc.id.au
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
Fax: (03) 8338 0299

There are no great men, only great challenges that ordinary men are 
forced by circumstances to meet. -- Admiral William Halsey

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Roman Mamedov put forth on 4/14/2011 5:54 AM:

> I suggest that you avoid Silicon Image 3132, they have data corruption issue
> (on some board designs?), triggered or amplified by transferring via both
> ports at the same time, at full speed.

Stating this opinion based on a very limited number of such reports is
irresponsible.  If the issue is a bad PCB from a couple of vendors,
state the vendors and the card model and rev.  DO NOT paint the SATA
chip itself as BAD and something to avoid.  Case in point:

http://blog.backblaze.com/2009/09/01/petabytes-on-a-budget-how-to-build-cheap-cloud-storage/

BackBlaze uses three Sil3132 based 2 port Syba PCIe x1 SATAII cards in
each storage 45 drive storage pod server, and one 4 port Addonics
Sil3124 PCI card.  Odd balance WRT bandwidth, but if you read the doc
it'll be clear why they did this.

They've had HUNDREDS of these Syba and Addonics Silicon Image chip cards
in production for multiple years now and report nothing but great
success.  They use 9 SATAII backplanes per chassis each with a 1:5
Silicon Image 3726 PMP, for a total of 45 drives per chassis.  They
create 3 x 15 drive mdadm RAID6 arrays per chassis, as the drives are
physically laid out in 3 rows of 15.

These systems are greater in scale and complexity than anything I've
seen discussed on this list in the year+ I've been subbed.  Again, the
backbone of their 45 drive pod is the Sil3132 chip, each 3132 managing
10 drives, 5 each connected to BOTH card ports via PMPs, 30 drives total
per chassis controlled by the Sil3132 chips.  The Addonics Sil3124 card
manages 15 drives across 3 ports and 3 PMPs.

Silicon Image has the most compatible, tested, and field proven reliable
SATA chips on the planet, ditto its PMP chips.  BackBlaze is but one
example of this.  Also, Silicon Image PMP chips ship in more products
than all other PMPs combined.

Again, there is nothing wrong with the Silicon Image chips.  They have a
very good reputation.  There is apparently a problem with some Chinese
really cheap badly designed/manufactured Silicon Image based  cards
going into the Russian market.  From the comments I read it appears the
"sample set" of bad cards consisted of less than a dozen.  This could
have simply been a bad batch of cards.  From the comments, this actually
seems like the most likely scenario.

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Brad Campbell]

On 15/04/11 03:37, Stan Hoeppner wrote:
> Roman Mamedov put forth on 4/14/2011 5:54 AM:
>
>> I suggest that you avoid Silicon Image 3132, they have data corruption issue
>> (on some board designs?), triggered or amplified by transferring via both
>> ports at the same time, at full speed.
>
> Stating this opinion based on a very limited number of such reports is
> irresponsible.  If the issue is a bad PCB from a couple of vendors,
> state the vendors and the card model and rev.  DO NOT paint the SATA
> chip itself as BAD and something to avoid.  Case in point:

Valid point. The card referenced in that thread is a generic unbranded 
card made in China and bought off the shelf from a computer store in 
Dubai, UAE. It has CE, FCC & C-tick marks on it but no approval numbers 
and the only code on the device is PI43132-8X2A written under the boot rom.

Re: Recommended pci-e 1x SATA cards.

[Roman Mamedov]

[-- Attachment #1: Type: text/plain, Size: 3647 bytes --]

On Thu, 14 Apr 2011 14:37:23 -0500
Stan Hoeppner <stan@hardwarefreak.com> wrote:

> > I suggest that you avoid Silicon Image 3132, they have data corruption
> > issue (on some board designs?), triggered or amplified by transferring via
> > both ports at the same time, at full speed.
> 
> Stating this opinion based on a very limited number of such reports is
> irresponsible.

Irresponsible to chip and board makers maybe, but I prefer to be responsible
with regard to protecting my own data from possibly bad components first.

> If the issue is a bad PCB from a couple of vendors, state the vendors and
> the card model and rev.

Yes, after the initial report I was thinking the problem is limited to one
particular vendor of the boards or even just a batch, but since it was
confirmed on this list, by person from a different continent, who bought card
from a different place/vendor/batch and still had the exactly same issue,
that's when it was enough for me to blacklist all cards on 3132 in general.

> Case in point:
> 
> http://blog.backblaze.com/2009/09/01/petabytes-on-a-budget-how-to-build-cheap-cloud-storage/

They also recommend using Seagate 7200.11 1.5TB drives, is that a
recommendation you would also support? :)
http://www.desktopreview.com/default.asp?newsID=593
http://hardware.slashdot.org/story/09/01/17/0115207/Seagate-Hard-Drive-Fiasco-Grows
http://it.slashdot.org/comments.pl?sid=1098793&cid=26542735

> BackBlaze uses three Sil3132 based 2 port Syba PCIe x1 SATAII cards in
> each storage 45 drive storage pod server, and one 4 port Addonics
> Sil3124 PCI card.  Odd balance WRT bandwidth, but if you read the doc
> it'll be clear why they did this.

The Syba card also costs 4 times what the cheaper cards on the same chip cost,
so is that what it takes to make a non-data-corrupting card with SiI3132? At
that price point, why not just get a Marvell 9123 anyway.
Oh wait, there are reports that even Syba corrupts data in the same way!
I even made a screenshot in case this goes away (can't find a permanent link
other than http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027)
http://ompldr.org/vODl1eg/2011-04-15T045338Z-syba.png
So it is just more expensive, not more reliable in any way. And it is after
all the chip's fault.

> Again, there is nothing wrong with the Silicon Image chips.  They have a
> very good reputation.

I do not agree, they have an terrible reputation and an awful track record
regarding reliability with the earlier 3112/3114/3124 chips. And their
problems most often manifest in the worst way possible, as silent corruption
of user data. See for yourself:
https://encrypted.google.com/search?hl=en&q=3112+data+corruption
https://encrypted.google.com/search?hl=en&q=3114+data+corruption

"Sil3112/3114 are now virtually the only controllers with occassional and
unresolved data corruption issues."
http://comments.gmane.org/gmane.linux.ide/44799

"there are known major problems with Silicon Image controllers on FreeBSD,
Linux, and Windows.  The most common problem is silent data
corruption. ...JMicron controllers are known to behave OK"
http://lists.freebsd.org/pipermail/freebsd-hackers/2008-September/026152.html

So it is not surprising to me AT ALL, that 3132 would also have similar kind
of 'surprises'. And with presence of a readily-available cheap alternative,
namely JMicron JMB363 (with an untarnished reputation, and being a nice
standard 'ahci' controller), just excluding any Silicon Image from future
choices forever, seems to be the right thing to do to me.

-- 
With respect,
Roman

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Recommended pci-e 1x SATA cards.

[Roman Mamedov]

[-- Attachment #1: Type: text/plain, Size: 332 bytes --]

On Thu, 14 Apr 2011 20:31:26 +1000
Steven Haigh <netwiz@crc.id.au> wrote:

> Has anyone got any recommendations on cheap pci-e SATA cards that work 
> well with linux?

See this article:

"From 16 to 2 ports: Ideal SATA/SAS Controllers for ZFS & Linux MD RAID" -
http://blog.zorinaq.com/?e=10

-- 
With respect,
Roman

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Recommended pci-e 1x SATA cards.

[Steven Haigh]

On 14/04/2011 10:53 PM, Roman Mamedov wrote:
> On Thu, 14 Apr 2011 20:31:26 +1000
> Steven Haigh<netwiz@crc.id.au>  wrote:
>
>> Has anyone got any recommendations on cheap pci-e SATA cards that work
>> well with linux?
>
> See this article:
>
> "From 16 to 2 ports: Ideal SATA/SAS Controllers for ZFS&  Linux MD RAID" -
> http://blog.zorinaq.com/?e=10
>

That is a very nice writeup... I searched for the HighPoint Rocket 620 - 
listed as $24.99 on NewEgg - but Australian suppliers seem to have it 
above $100AUD. The cheapest price I found was $54AUD - and they were out 
of stock.

It seems that even ebay are over the $100AUD mark for the card - however 
I can import one from the USA for less than $40AUD. Sometimes I hate the 
Australian technology market!

--
Steven Haigh

Email: netwiz@crc.id.au
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
Fax: (03) 8338 0299

There are no great men, only great challenges that ordinary men are 
forced by circumstances to meet. -- Admiral William Halsey

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Steven Haigh put forth on 4/14/2011 8:16 AM:

> That is a very nice writeup... I searched for the HighPoint Rocket 620 -
> listed as $24.99 on NewEgg - but Australian suppliers seem to have it
> above $100AUD. The cheapest price I found was $54AUD - and they were out
> of stock.

This is the card you should get:
http://www.sybausa.com/productInfo.php?iid=536
http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027

It's the same Syba Sil3132 based card BackBlaze uses in their 45 drive
mdadm managed pods.

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Steven Haigh]

On 15/04/2011 5:55 AM, Stan Hoeppner wrote:
> Steven Haigh put forth on 4/14/2011 8:16 AM:
>
>> That is a very nice writeup... I searched for the HighPoint Rocket 620 -
>> listed as $24.99 on NewEgg - but Australian suppliers seem to have it
>> above $100AUD. The cheapest price I found was $54AUD - and they were out
>> of stock.
>
> This is the card you should get:
> http://www.sybausa.com/productInfo.php?iid=536
> http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027
>
> It's the same Syba Sil3132 based card BackBlaze uses in their 45 drive
> mdadm managed pods.

I did look at these, and they do seem to be what I am after - however 
they don't seem to exist in Australia :(

--
Steven Haigh

Email: netwiz@crc.id.au
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
Fax: (03) 8338 0299

There are no great men, only great challenges that ordinary men are 
forced by circumstances to meet. -- Admiral William Halsey

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Steven Haigh put forth on 4/14/2011 6:50 PM:
> On 15/04/2011 5:55 AM, Stan Hoeppner wrote:
>> Steven Haigh put forth on 4/14/2011 8:16 AM:
>>
>>> That is a very nice writeup... I searched for the HighPoint Rocket 620 -
>>> listed as $24.99 on NewEgg - but Australian suppliers seem to have it
>>> above $100AUD. The cheapest price I found was $54AUD - and they were out
>>> of stock.
>>
>> This is the card you should get:
>> http://www.sybausa.com/productInfo.php?iid=536
>> http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027
>>
>> It's the same Syba Sil3132 based card BackBlaze uses in their 45 drive
>> mdadm managed pods.
> 
> I did look at these, and they do seem to be what I am after - however
> they don't seem to exist in Australia :(

Sure they do.  Just not the Syba branded 3132 card.  The Addonics
branded 3132 card is available down under:

http://www.techbuy.com.au/p/127742/HARD_DISK_CONTROLLERS_SERIAL_ATA_RAID_CONTROLLERS/Addonics/AD2SA3GPX1.asp

Addonics is probably a more trusted brand than Syba anyway, being based
in Silicon Valley, not China or Taiwan.

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Steven Haigh]

On 15/04/2011 2:06 PM, Stan Hoeppner wrote:
> Steven Haigh put forth on 4/14/2011 6:50 PM:
>> On 15/04/2011 5:55 AM, Stan Hoeppner wrote:
>>> Steven Haigh put forth on 4/14/2011 8:16 AM:
>>>
>>>> That is a very nice writeup... I searched for the HighPoint Rocket 620 -
>>>> listed as $24.99 on NewEgg - but Australian suppliers seem to have it
>>>> above $100AUD. The cheapest price I found was $54AUD - and they were out
>>>> of stock.
>>>
>>> This is the card you should get:
>>> http://www.sybausa.com/productInfo.php?iid=536
>>> http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027
>>>
>>> It's the same Syba Sil3132 based card BackBlaze uses in their 45 drive
>>> mdadm managed pods.
>>
>> I did look at these, and they do seem to be what I am after - however
>> they don't seem to exist in Australia :(
>
> Sure they do.  Just not the Syba branded 3132 card.  The Addonics
> branded 3132 card is available down under:
>
> http://www.techbuy.com.au/p/127742/HARD_DISK_CONTROLLERS_SERIAL_ATA_RAID_CONTROLLERS/Addonics/AD2SA3GPX1.asp
>
> Addonics is probably a more trusted brand than Syba anyway, being based
> in Silicon Valley, not China or Taiwan.
>

Keep in mind that this card is over double the cost of the Syba brand 
from NewEgg. For some reason, it seems that the cards I've located in 
retail stores are double what they are in the US. Taking into account 
the exchange rate, they should be about the same cost - is $1AUD is 
currently around $1.05USD.

At this stage, I think my best bet for a new card is to get a mate in 
the US to get one and ship it over. It is probably still cheaper than 
buying locally.

--
Steven Haigh

Email: netwiz@crc.id.au
Web: http://www.crc.id.au
Phone: (03) 9001 6090 - 0412 935 897
Fax: (03) 8338 0299

There are no great men, only great challenges that ordinary men are 
forced by circumstances to meet. -- Admiral William Halsey

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Steven Haigh put forth on 4/14/2011 11:35 PM:

> Keep in mind that this card is over double the cost of the Syba brand
> from NewEgg. For some reason, it seems that the cards I've located in
> retail stores are double what they are in the US. Taking into account
> the exchange rate, they should be about the same cost - is $1AUD is
> currently around $1.05USD.

I'm not familiar at all with product pricing (gauging) in AU.  The
Addonics card is only $8 more than the Syba card at Newegg, $35 vs $27
USD.  Isn't all PC hardware overpriced in AU, just like in the UK?  I've
seen numerous mobos and what not priced at over 300 pounds, making them
equivalent to over $1000 USD.  These were run of the mill upper midrange
Asus etc type boards that sell for less than $200 USD.

> At this stage, I think my best bet for a new card is to get a mate in
> the US to get one and ship it over. It is probably still cheaper than
> buying locally.

If international shipping on a less than 1 lbs package is less than $30
then yeah, you'd probably save a little.

Sorry I couldn't get you better options, cost wise.  I recommended the
lowest price highest quality/compat cards available, in the US that is.

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Roman Mamedov]

[-- Attachment #1: Type: text/plain, Size: 1998 bytes --]

On Thu, 14 Apr 2011 14:55:40 -0500
Stan Hoeppner <stan@hardwarefreak.com> wrote:

> Steven Haigh put forth on 4/14/2011 8:16 AM:
> 
> > That is a very nice writeup... I searched for the HighPoint Rocket 620 -
> > listed as $24.99 on NewEgg - but Australian suppliers seem to have it
> > above $100AUD. The cheapest price I found was $54AUD - and they were out
> > of stock.
> 
> This is the card you should get:
> http://www.sybausa.com/productInfo.php?iid=536
> http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027
> 
> It's the same Syba Sil3132 based card BackBlaze uses in their 45 drive
> mdadm managed pods.
> 

Did you read the second link you posted, specifically the Feedback section?
That's one more confirmation of corruption issue we discussed in this thread.

"Cons: Corrupts data when reading from two drives at once (!)
Other Thoughts: I bought this to add 2 SATA ports to my desktop. The cheap
software-based PCIe --> 2x SATA cards seem to be based on SIL3132 or JMB363
chipsets (this is a SIL3132), but this was more highly rated so I went with
it. My intention was to add two individual hard drives to my system, with no
RAID involved. Things seemed fine at first, until I used TeraCopy, which does
a CRC after file copy operations. It revealed occasional mismatches when
copying files to either drive.  Then I noticed MD5 hashes of files on the
drives, taken sequentially, differing without any changes to the files'
contents. The final straw was that reading from each of the two separate
drives simultaneously would result in silent read errors,  to the point where
the file would be heavily corrupted but the OS didn't care.   I stopped using
the controller and bought a 3ware one instead, and it's been working
flawlessly.   I don't know if this is a problem specific to this controller or
its chipset, but I would not recommend it."
http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027

-- 
With respect,
Roman

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Roman Mamedov put forth on 4/14/2011 11:58 PM:
> On Thu, 14 Apr 2011 14:55:40 -0500
> Stan Hoeppner <stan@hardwarefreak.com> wrote:
> 
>> Steven Haigh put forth on 4/14/2011 8:16 AM:
>>
>>> That is a very nice writeup... I searched for the HighPoint Rocket 620 -
>>> listed as $24.99 on NewEgg - but Australian suppliers seem to have it
>>> above $100AUD. The cheapest price I found was $54AUD - and they were out
>>> of stock.
>>
>> This is the card you should get:
>> http://www.sybausa.com/productInfo.php?iid=536
>> http://www.newegg.com/Product/Product.aspx?Item=N82E16815124027
>>
>> It's the same Syba Sil3132 based card BackBlaze uses in their 45 drive
>> mdadm managed pods.
>>
> 
> Did you read the second link you posted, specifically the Feedback section?
> That's one more confirmation of corruption issue we discussed in this thread.

The overall Newegg rating for the Syba card you deride is 4/5 eggs
across 109 reviews.  If the card was as horrible as you make it out to
be, the rating would be 1/5 not 4/5, and people would stop buying it.

Note that the user who posted the read corruption issue referred to a
single application in Windows 7 where this error occurred, TeraCopy.
It's far more likely he was up against an application or driver issue
than a hardware issue with the Syba card.  He did not state whether a
Windows Explorer copy would also cause the problem, nor xcopy, nor
Robocopy, etc.

BTW, did you even read the BackBlaze blog I posted?  They run hundreds
of this exact Syba 3132 card, with Linux, with mdraid, and have reported
zero problems.  And they're using a 5:1 PMP on each 3132 port.  If these
cards, or the 3132 were junk, as you state, surely BackBlaze would have
run into problems in 2+ years of full production, no?

Again, you're taking isolated incidents and assuming they are the norm,
when they most certainly are not.

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Roman Mamedov]

[-- Attachment #1: Type: text/plain, Size: 3198 bytes --]

On Fri, 15 Apr 2011 16:31:08 -0500
Stan Hoeppner <stan@hardwarefreak.com> wrote:

> The overall Newegg rating for the Syba card you deride is 4/5 eggs

I am not deriding anything... but from the way you react to it, I get a
feeling I am talking to a Silicon Image or Syba employee. :)

> across 109 reviews.  If the card was as horrible as you make it out to
> be, the rating would be 1/5 not 4/5, and people would stop buying it.

What is the probability most simply won't notice it? Or they don't use 2 disks
on this card, or don't use them both at full speed. Also I agree that most
likely the bug isn't present on 100% of boards, if you remember the initial
reporter (in Russia) had the problem only on 2 boards of 5 identical ones he
bought.

> Note that the user who posted the read corruption issue referred to a
> single application in Windows 7 where this error occurred, TeraCopy.
> It's far more likely he was up against an application or driver issue
> than a hardware issue with the Syba card.  He did not state whether a
> Windows Explorer copy would also cause the problem, nor xcopy, nor
> Robocopy, etc.

Yes, and still, this is an extremely valuable report. The problem described is
exactly the same what was reported on previous two occasions: "two ports used
at full speed => silent corruption in files". And now we know the problem is
not limited to the Linux driver, it also appears in Windows 7. But waving it
off as "just a bug in TeraCopy" or whatever, against two previous
confirmations of exactly this type of corruption on exactly this hardware (the
3132 controller) in the same circumstances, seems unreasonable to me.

> BTW, did you even read the BackBlaze blog I posted?  They run hundreds
> of this exact Syba 3132 card, with Linux, with mdraid, and have reported
> zero problems.  And they're using a 5:1 PMP on each 3132 port.  If these
> cards, or the 3132 were junk, as you state, surely BackBlaze would have
> run into problems in 2+ years of full production, no?

Again, can you quote where I used the word "junk"...
Regarding the Backblaze blog - yes, I read it. Dunno - did they post any
follow-up if they had to replace some cards? Would they even publish something
like that? Would they actually pinpoint some mysterious corruptions now and
then to the controller card? Also if we're talking a hardware bug here(and I
think we do) -- maybe it doesn't surface when PMPs are used instead of disks
directly? Or maybe it doesn't happen with the server-grade Intel motherboard
(which AFAIR they use), but only when the card is combined with PCI-E
implementation of more common chipsets?

> Again, you're taking isolated incidents and assuming they are the norm,
> when they most certainly are not.

"Significant percentage of 3132 cards" is enough to me. Where I'd define
"significant" even as 0.01% of cards. But actually it is likely to be more
widespread than that, given this exact problem was already reported by 3
people on 3 continents with 2 different OSes and 4 controller boards,
including a brand-name one. If this is not a wide sampling, I don't know what
is...

-- 
With respect,
Roman

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

Re: Recommended pci-e 1x SATA cards.

[Brad Campbell]

On 16/04/11 13:15, Roman Mamedov wrote:
> On Fri, 15 Apr 2011 16:31:08 -0500
> Stan Hoeppner<stan@hardwarefreak.com>  wrote:


>> Again, you're taking isolated incidents and assuming they are the norm,
>> when they most certainly are not.
>
> "Significant percentage of 3132 cards" is enough to me. Where I'd define
> "significant" even as 0.01% of cards. But actually it is likely to be more
> widespread than that, given this exact problem was already reported by 3
> people on 3 continents with 2 different OSes and 4 controller boards,
> including a brand-name one. If this is not a wide sampling, I don't know what
> is...
>

That it happens at all is enough for me to stay _well_ clear of them.

I can't recommend anybody go within a mile of a product that works well 
for 99.999% of people, when if you happen to be the 0.001% sample you 
don't know you are until the device has silently eaten all your data.

When there are alternatives available that have 0 reports of silently 
eating data, why would you even chance it?

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Brad Campbell put forth on 4/16/2011 3:57 AM:

> I can't recommend anybody go within a mile of a product that works well
> for 99.999% of people, when if you happen to be the 0.001% sample you
> don't know you are until the device has silently eaten all your data.

Then you may as well not recommend any board to anyone, because few
manufacturers of relatively inexpensive boards have QC that high or an
in the field failure rate that low, not matter whose control IC they
mount on the PCB.  Margins on such products are razor thin, thus these
manufacturers cut corners where they can.  Some cut corners more
sharply, such as the in the case where 2/5 boards demonstrated the flaw
and 3 didn't.  This directly points to a QC problem, not an IC design
flaw in the 3132.

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Brad Campbell]

On 17/04/11 04:40, Stan Hoeppner wrote:
> Brad Campbell put forth on 4/16/2011 3:57 AM:
>
>> I can't recommend anybody go within a mile of a product that works well
>> for 99.999% of people, when if you happen to be the 0.001% sample you
>> don't know you are until the device has silently eaten all your data.
>
> Then you may as well not recommend any board to anyone, because few
> manufacturers of relatively inexpensive boards have QC that high or an
> in the field failure rate that low, not matter whose control IC they
> mount on the PCB.  Margins on such products are razor thin, thus these
> manufacturers cut corners where they can.  Some cut corners more
> sharply, such as the in the case where 2/5 boards demonstrated the flaw
> and 3 didn't.  This directly points to a QC problem, not an IC design
> flaw in the 3132.
>

Now as I have a board here that demonstrates the exact same problem and 
is physically different from those boards then I'd suggest that there is 
more to this than bad QC on the part of one manufacturer.

While I understand what you are saying. What part of "silent data 
corruption" are you not getting?

Find me an example of an LSI or Marvell based board that causes *silent* 
data corruption. They all have their flaws, sure and some of those can 
be traced back to cheap dodgy manufacturing. The critical issue with all 
these other products is that when they have issues you know about it. 
The LSI boards used to lock up a channel when you sent an unaligned CDB 
to them, the Marvell boards have been known to drop channels completely 
(total hardware failure). Both these issues are highly visible, and 
whilst annoying you at least *know* there is a problem.

The whole premise of storage that has no end to end verification 
strategy is that you can trust your data path. The SIL chips (and I 
don't care how much you bang on about cheap hardware) have a bug that in 
certain circumstances invalidates all guarantees about data integrity 
silently.

Hells bells, there is even a case that Roman pointed to where the 
precise controller you recommend as being of suitable quality 
demonstrates the exact fault we are talking about, and you try to blame 
the copy program or the Windows driver.

There appears to be an insidious flaw in these chips that only manifests 
itself under "perfect storm" conditions, but when it does it silently 
eats your data.

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Brad Campbell put forth on 4/17/2011 12:44 AM:

> Hells bells, there is even a case that Roman pointed to where the
> precise controller you recommend as being of suitable quality
> demonstrates the exact fault we are talking about, and you try to blame
> the copy program or the Windows driver.
> 
> There appears to be an insidious flaw in these chips that only manifests
> itself under "perfect storm" conditions, but when it does it silently
> eats your data.

I've been in the hardware game a long time, and all the evidence I'm
seeing WRT this silent data corruption issue points simply to QC, not a
chip design flaw.  If the problem were a chip design flaw, we'd see far
more widespread reporting, as millions of this chip have shipped into
the marketplace.

The fact that the Russian with 5 cards purchased in Dubai could
routinely demonstrate this problem with 2 of 5 identical cards, and
never on the other 3, points directly to a board QC isue.  Again, if the
problem were a design flaw in the IC itself, all 5 cards would have
exhibited the problem.

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Mikael Abrahamsson]

On Sun, 17 Apr 2011, Stan Hoeppner wrote:

> The fact that the Russian with 5 cards purchased in Dubai could 
> routinely demonstrate this problem with 2 of 5 identical cards, and 
> never on the other 3, points directly to a board QC isue.  Again, if the 
> problem were a design flaw in the IC itself, all 5 cards would have 
> exhibited the problem.

I think you're missing the point.

A proper hardware design should detect errors in the chip, and should 
offline itself (or at least complain loudly). This is robustness in 
design, that if something goes wrong in the production process that QC 
doesn't find, card notices this and goes offline.

I personally have two of these controllers and they're working well, but I 
have to agree that it's very likely that these low-cost products are not 
as resilient to problems as the more expensive devices out there.

-- 
Mikael Abrahamsson    email: swmike@swm.pp.se

Re: Recommended pci-e 1x SATA cards.

[Stan Hoeppner]

Mikael Abrahamsson put forth on 4/17/2011 1:45 PM:
> On Sun, 17 Apr 2011, Stan Hoeppner wrote:
> 
>> The fact that the Russian with 5 cards purchased in Dubai could
>> routinely demonstrate this problem with 2 of 5 identical cards, and
>> never on the other 3, points directly to a board QC isue.  Again, if
>> the problem were a design flaw in the IC itself, all 5 cards would
>> have exhibited the problem.
> 
> I think you're missing the point.

/me laughs.  I'm guessing that's because you're very late to this thread.

> A proper hardware design should detect errors in the chip, and should
> offline itself (or at least complain loudly). This is robustness in
> design, that if something goes wrong in the production process that QC
> doesn't find, card notices this and goes offline.

NOBODY in the PCB add-in card biz does this, except maybe in the
mainframe space.  You seem to be missing the fact that the SATA IC cost
$50 million (example) to bring to market in R&D and manufacturing cost.

The amount of money an individual board maker puts into his product to
which he solders this SATA IC is maybe $50k.  He uses the chip
producer's reference diagram to reduce his own R&D costs to the minimum
required.  THEN, he starts cutting corners, using cheap caps, resistors,
etc that don't meet the specifications set forth by the chip maker.
This is why you have 2/5 identical cards with a problem, and 3/5 that
don't.  Some of the components meet voltage/current specs, some don't.

> I personally have two of these controllers and they're working well, but
> I have to agree that it's very likely that these low-cost products are
> not as resilient to problems as the more expensive devices out there.

Not once have I stated that cheap cards are a "good" solution, whether
based on SiI, JMicron, or Marvell ICs.  The OP in Australia "demanded" a
cheap 2 port PCIe SATA card, and I recommended a SiI 3132 based card.
SiI was blanketly attacked, by one poster anyway, I defended it, and
rightfully so.  The chips themselves (current product anyway) are fine.
 Many cards based on them have QC issues, which is not news to anyone in
this game more than a few years.  Many cheap cards based on Marvell and
JMicron chips have QC issues as well.  The problem here isn't the SATA
IC but the QC of the $15-25 USD card.

Again, I'm not sure what "point" you believe I'm missing.  My hardware
experience likely puts me in the 90th percentile or above of members of
this list.  I've seen the manufacturing supply chain of both ultra cheap
and more expensive product for over 15 years.  I've seen disk
controllers, video cards, etc, from two manufacturers based on the same
IC.  The ultra cheap ones often experience problems the higher dollar
units due not.  This has been going on in this industry for 30 years.

The problem is almost never a design flaw with the main IC (whether
SCSI, SATA, GPU, etc), but almost always the board implementation.  The
big money is in the ICs, and they have high QC.  Board (PCB)
manufacturing cost is often $0.50 USD/board or less.  If all $0.50 of
that board making was QC, is that sufficient? (rhetorical)

-- 
Stan

Re: Recommended pci-e 1x SATA cards.

[Sven Eschenberg]

Are you sure about this?

All those nifty specification updates for CPUs tell quite a different
story. And yes IC producers usually have such things too, they just
prefer to not publish them openly but rather decide to gag their
customers (like OEMs) by non disclosure agreements, if those plan on
getting a hold of those spec updates to work around the issues instead
of throwing away all the chips they bought.

But that just on a side note.

Bad PCB design and bad quality components are usually the more common
things causing problems. On the other hand, what was that story about
the ICHs in quite some chipsets having problems with the SATA Ports?

http://newsroom.intel.com/community/intel_newsroom/blog/2011/01/31/intel-identifies-chipset-design-error-implementing-solution

BTW: Big OEMs decided to get those broken chips nevertheless for
notebooks, because the problems usually do not occur with the first two
ports and the chips will probably be VERY cheap *eg*.

Marvell had issues (design) with the legacy PATA Interface on some
PATA/SATA chips earlier, then bumped the rev. and fixed the silicon.

Design errors and flaws are extremely common and there are certainly
enough in any IC, question is, if you hit 'em and if they cause
problems. And don't forget all the possible problems that arise from
production flaws (bad wafers etc.).

Anyway, concerning the original issue a bad PCB is probably rather the
issue. And SIL is as good a choice as Marvell or JMicron in most cases.

Regards

-Sven

On Sun, 2011-04-17 at 15:26 -0500, Stan Hoeppner wrote:

> The problem is almost never a design flaw with the main IC (whether
> SCSI, SATA, GPU, etc), but almost always the board implementation.  The
> big money is in the ICs, and they have high QC.  Board (PCB)
> manufacturing cost is often $0.50 USD/board or less.  If all $0.50 of
> that board making was QC, is that sufficient? (rhetorical)
> 

RE: Recommended pci-e 1x SATA cards.

[Guy Watkins]

} -----Original Message-----
} From: linux-raid-owner@vger.kernel.org [mailto:linux-raid-
} owner@vger.kernel.org] On Behalf Of Stan Hoeppner
} Sent: Sunday, April 17, 2011 2:37 PM
} To: Brad Campbell
} Cc: Roman Mamedov; Steven Haigh; linux-raid@vger.kernel.org
} Subject: Re: Recommended pci-e 1x SATA cards.
} 
} Brad Campbell put forth on 4/17/2011 12:44 AM:
} 
} > Hells bells, there is even a case that Roman pointed to where the
} > precise controller you recommend as being of suitable quality
} > demonstrates the exact fault we are talking about, and you try to blame
} > the copy program or the Windows driver.
} >
} > There appears to be an insidious flaw in these chips that only manifests
} > itself under "perfect storm" conditions, but when it does it silently
} > eats your data.
} 
} I've been in the hardware game a long time, and all the evidence I'm
} seeing WRT this silent data corruption issue points simply to QC, not a
} chip design flaw.  If the problem were a chip design flaw, we'd see far
} more widespread reporting, as millions of this chip have shipped into
} the marketplace.

I disagree.  In fact, Intel recalled their new chip (sandy bridge I think)
that had a design flaw that was very unlikely to cause any problems.
However, Intel did not want the risk, so recalled them all.

} 
} The fact that the Russian with 5 cards purchased in Dubai could
} routinely demonstrate this problem with 2 of 5 identical cards, and
} never on the other 3, points directly to a board QC isue.  Again, if the
} problem were a design flaw in the IC itself, all 5 cards would have
} exhibited the problem.

I would bet the cards were based on the reference design.  Maybe the
reference design has the flaw?  Chip or board design, signal timing could be
really close to some limit, add a bit extra solder here or there and you
have a bad card.  No 2 cards are identical.  If they were identical, you
would only need to test the first card made, and know the rest are
identical, so no extra testing is needed.

Guy

} 
} --
} Stan
} --
} To unsubscribe from this list: send the line "unsubscribe linux-raid" in
} the body of a message to majordomo@vger.kernel.org
} More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: Recommended pci-e 1x SATA cards.

[Brad Campbell]

On 18/04/11 02:36, Stan Hoeppner wrote:
> Brad Campbell put forth on 4/17/2011 12:44 AM:
>
>> Hells bells, there is even a case that Roman pointed to where the
>> precise controller you recommend as being of suitable quality
>> demonstrates the exact fault we are talking about, and you try to blame
>> the copy program or the Windows driver.
>>
>> There appears to be an insidious flaw in these chips that only manifests
>> itself under "perfect storm" conditions, but when it does it silently
>> eats your data.
>
> I've been in the hardware game a long time, and all the evidence I'm
> seeing WRT this silent data corruption issue points simply to QC, not a
> chip design flaw.  If the problem were a chip design flaw, we'd see far
> more widespread reporting, as millions of this chip have shipped into
> the marketplace.


I did a cursory search on the 3132 last night. There are a *staggering* 
number of reports of silent data corruption on the Mac. It turns out 
these cards are a pretty common add-on in the Mac community. It's so bad 
some vendors of expansion chassis explicitly state that using a Sil3132 
card to drive their chassis will eventually lead to corrupt data.

> The fact that the Russian with 5 cards purchased in Dubai could
> routinely demonstrate this problem with 2 of 5 identical cards, and
> never on the other 3, points directly to a board QC isue.  Again, if the
> problem were a design flaw in the IC itself, all 5 cards would have
> exhibited the problem.
>

Actually Stan, I purchased my card in Dubai. The Russian cards were 
purchased in Eastern Europe.

I'm going to drop this thread now as the horse is well and truly dead.

Friends don't let friends use SIL3132 controllers.