2.5TByte SATA drive to pi?

I have a model B sitting around doing nothing, and a 2.5TB drive I use for backup which I'd like to offload from the machine it's in to a standalone system. But all the USBSATA enclosures and docking stations I've found on fleabay have limits of 2TB or below. Anybody know one that'll handle a bigger drive?

--
John Stumbles 
(.sig temporarily unavailable)
Reply to
John
Loading thread data ...

Oops, there are some devices that handle 4TB drives. My bad.

Reply to
John

For this purpose a Banana Pi is much more suitable than a Raspberry Pi. Onboard SATA, fast network interface.

Reply to
Rob

software support ;-)

(If I were after fast NAS it would be worth it but this only has to do and rsync snapshot thing once a day so performance isn't really an issue.)

Reply to
John

Are the limits genuine or just stated because that was all that was available at the time?

I doubt that there is any real physical limit except those imposed by the SATA specification (if there is one) & your operating system (which you can guarantee wont be subject to the same limits as windoze).

--
I always had a repulsive need to be something more than human. 
		-- David Bowie
Reply to
alister

When you have no knowledge about the matter it is better not to post things like that, because you lead readers into stray paths.

YES THERE IS a limit to the capacity that USB disk adapters can handle, and YES YOU DO need to verify that they work above 2.2 TB.

Reply to
Rob

There is more than just a different cable on each end...

--
Everything you read in newspapers is absolutely true, except for the  
rare story of which you happen to have first-hand knowledge. ? Erwin Knoll
Reply to
The Natural Philosopher

I am surprised but happy to stand corrected.

--
It's difficult to see the picture when you are inside the frame.
Reply to
alister

That's a reasonable question when you remember the 400GB? 'limit' implied by the specs for external IDE to USB enclosures over a decade ago which proved to be just simply due to the largest available drives the manufacturers were able to test their product against.

That would seem to clarify the _possible_ confusion between supposition and fact _this_ time around. :-)

--
J B Good
Reply to
Johny B Good

You really have to check your USB2SATA controler. I have one that works ok on 2TB HDD but if I try to use it on 3TB HDD it reports different size... so partition could be unreadable. Once I asked the seller to confirm that it works with 4TB HDD... but it did not work. I do not know if there is a way to check this without having a partitioned HDD to check what your device reports.

Reply to
Nikolaj Lazic

I keep a SATA disk docking station attached to my win2k desktop PC. It has both USB2 and e-SATA interfaces. Naturally, I'm using the e-SATA connection. You might reasonably expect that the USB on such a docking station would support >2TiB disk drives but in view of what has been said in this thread, I'm not going to take that for granted.

The next NAS box HDD upgrade I'm planning on making sometime within the next 6 months should give me the opportunity to test with a 6TB drive on a Linux setup to solve this particular conundrum. Until then, I'll remain in 'blissful ignorance'. :-)

--
J B Good
Reply to
Johny B Good

I don't think Windows 2000 can use GPT partitioning, which is a prerequisite to using discs beyond the 2TiB mark.

Note that ATA's thresholds at 8GiB (LBA24) and 128PiB (LBA48) aren't relevant here, the hardware limit is one of SCSI's READ10/WRITE10 commands which give a 2TiB limit. (USB Mass Storage is SCSI in disguise). SCSI's READ16/WRITE16 have a 8ZiB limit, which should be enough for a little while.

Theo

Reply to
Theo Markettos

Hmmm. I've got a USB/SATA adaptor somewhere. I wonder if there's a way to tell if its chipset supports this READ16/WRITE16 if I haven't got a

2+ TiB drive to hand?
Reply to
Dave Farrance

I'm afraid you're correct. There's not even a workaround to use >2TiB disks just for storage as I think there may have been for winXP (I have a hazy memory that a driver update became available in winXP to use such disks purely for data storage purposes - XP certainly couldn't be made to boot from such a large disk without pointlessly[1] limiting the LBA range to within the 2TiB limit.

Assuming winXP can be made to access such large disks, purely as large data storage volumes, I doubt such a method would be available to win2k (at least without some 'reverse engineering' of the winXP driver if such a driver exists).

It's not a major issue since Gbit connected NAS drives of 4TB present no problems using the SMB sharing mechanism and I expect the same will also apply to the newly available 6TB drives when I get around to my next disk capacity upgrade later this year (I'll be pulling the last of the 2TB drives out of the NAS box for that upgrade).

If the SATA docking station does suffer the 2TiB limit via its USB2 port, I'll still be ok with the e-SATA connection (but just not with win2k as the host OS).

[1] Not entirely pointlessly if the only drive you have to hand to boot from is a 3TB disk in which case, the "short stroking" exercise will offer a small performance boost by way of consolation for the significant capacity sacrifice involved.
--
J B Good
Reply to
Johny B Good

This problem has occurred several times in the past, and solutions like OnTrack, SpeedStor and EZ-drive were developed to make large disks work with existing sytems.

If I remember correctly, several different products were released under the OnTrack name that worked for different systems and limits, starting with MS-DOS.

No idea if such things still exist today.

Reply to
Rob

The fact that software and firmware are written with such limitations is ridiculous. The capacity increase of inexpensive drives has been steady and predictable.

--
-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon
Reply to
Michael J. Mahon

There is a tradeoff between effeciency and limits. Of course the difference is usually quite small, but it is understandable that the SCSI commands were extended from 6 to 10 bytes to raise the disk size limit from 2GB to 2TB. In those days 2TB was a very large disk (with 2GB disks just becoming available).

Reply to
Rob

Standards should always aim to be useful for at least the next decade, with escapes to allow easy expansion.

The steady increase in RAM and disk capacities is one of the most predictable advances of the digital age.

--
-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon
Reply to
Michael J. Mahon

So when did MBR based partitioning and its 32bit LBA aka 2.2TiB limit first appear? Oh that would be around 1983.

When did 2TiB drives become silly money cheap? Say 2012 for argument's sake.

Well that gives us 3 decades.

What were you saying about useful for a decade? ;-)

Reply to
mm0fmf

So you're saying it's a dead standard?

If a standard is in widespread, expanding use, then it is reasonable to expect that it has sponsors who will ensure that it continues to anticipate technology evolution (so as to maintain its economic benefit).

This problem was 1) easily foreseen and 2) easily solved years before its limits are reached by including "versioning".

We have known how to deal with backward compatibility and open-ended expandibility since the 1960s, if not longer.

I have managed the backward-compatible evolution of a computer architecture from 32 bits to 64, and, while having complexities, it is quite doable.

Certainly the data structures that create these limits can be extended to overcome them, particularly since the standard can be extended years before the limits are reached.

(Of course, at some point a file system with a smallest allocation unit of a gigabyte will be pretty wasteful. ;-)

--
-michael - NadaNet 3.1 and AppleCrate II: http://home.comcast.net/~mjmahon
Reply to
Michael J. Mahon

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.