Transfer data from USB HDD to Freenas box

Status
Not open for further replies.

ives31

Dabbler
Joined
Oct 17, 2014
Messages
33
Hi,
Can someone give me some pointers on how to transfer the contents of my USB hdd to my freenas box?
The USB hdd is NTFS or fat, not sure which. Is there a way to do this via the gui or is it a command line job?

thanks for any help.
 

gpsguy

Active Member
Joined
Jan 22, 2012
Messages
4,472
You can't do it via the webGUI.

The best option would be to connect it to your client computer and copy it via the network, using CIFS, FTP, or whatever protocol you are using.


Sent from my phone
 

Ericloewe

Server Wrangler
Moderator
Joined
Feb 15, 2014
Messages
20,194
You can use an NTFS drive attached to the server without risking serious data corruption only in read mode, but it has to be done manually. FAT is out of the question as it's not supported.
 

ives31

Dabbler
Joined
Oct 17, 2014
Messages
33
You can't do it via the webGUI.

The best option would be to connect it to your client computer and copy it via the network, using CIFS, FTP, or whatever protocol you are using.


Sent from my phone
Thanks.
This will be time consuming, first copying to my client and then copying again to the NAS, but if there's no way round it that's what I'll have to do.
 

depasseg

FreeNAS Replicant
Joined
Sep 16, 2014
Messages
2,874
Why copy it to the client?

Just connect it to the client and copy from the attached drive to the NAS.
 

Ericloewe

Server Wrangler
Moderator
Joined
Feb 15, 2014
Messages
20,194
Why copy it to the client?

Just connect it to the client and copy from the attached drive to the NAS.
It's easier, and many cases even faster, since USB 3.0 is probably a no-go on the server.
 

9C1 Newbee

Patron
Joined
Oct 9, 2012
Messages
485
I just
Why copy it to the client?

Just connect it to the client and copy from the attached drive to the NAS.

Just transferred 10TB of data using this method. I would recommend a copy utility such as TeraCopy on a windows machine. It uses CRC to verify the copy is correct on the server.
 

depasseg

FreeNAS Replicant
Joined
Sep 16, 2014
Messages
2,874
It's easier, and many cases even faster, since USB 3.0 is probably a no-go on the server.

I was referring to why perform the the extra step of copying it from the USB drive to the client drive. Copying it from the client makes sense across the network to the NAS.
 

Ericloewe

Server Wrangler
Moderator
Joined
Feb 15, 2014
Messages
20,194
I was referring to why perform the the extra step of copying it from the USB drive to the client drive. Copying it from the client makes sense across the network to the NAS.
It's typically faster, but by no means necessary.
 

9C1 Newbee

Patron
Joined
Oct 9, 2012
Messages
485
How is it faster? I have no idea how going through the usb to the hdd. Then from hdd through the nic to the server, would end up being faster. Can you break it down for me?
 

mjws00

Guru
Joined
Jul 25, 2014
Messages
798
Depending on the device it can be slow transferring small files as we are adding network latency on top of USB latency when waiting for acks etc. Not as noticeable on fast disks and USB 3. But stick a large slower USB stick in, and dumping to fast local storage can make a difference. Plus if we're moving lots of data we may want a sync program that benefits by fast local storage as a buffer and to use for a working set.

Truthfully a modern USB 3 enclosure and a fast disk is quicker than 1Gbe. Mine just saturates and writing direct is cool. If I grab an older portable drive with a slow laptop drive in it. I'm often better off to dump local and then transfer over the wire. Ymmv.
 

cyberjock

Inactive Account
Joined
Mar 25, 2012
Messages
19,525
I have to disagree with this argument. There's two possible outcomes:

1. Copy data from USB to local drive, then again from local drive to the server. You're copying the data twice. Even in best case, you're looking to copy every byte twice.
2. Copy data from USb directly to the server. Only copying the data once, and so you save on quite a bit of time. Not sure what acks you are going to argue about, but the acks delay is going to be the same reguardless, the question is whether you do the acks in 2 parts for #1 or 1 part for #2. But the time lost while actually writing the data in #1 makes it a slower solution.
 

mjws00

Guru
Joined
Jul 25, 2014
Messages
798
If things go perfectly... cool. If things go sideways off slow media. I'd rather read once to local fast storage and not have to start from scratch. Too easy to hit an error or drop a connection the middle of a large transfer and have to restart. Remember your love of USB ;) It is also an old school problem imho. Both speed and reliability is much much better thus my "direct writing is cool".

You need more time in HELL, cyber ;). It wasn't long ago reading and writing over the wire to a remote detachable storage could be nasty and unreliable. If you've never had to copy locally to get crap hardware or something causing an error or timing out of the loop you are a lucky man. Can't tell you how many times windows has puked on me and I got to start all over.
 

Ericloewe

Server Wrangler
Moderator
Joined
Feb 15, 2014
Messages
20,194
I have to disagree with this argument. There's two possible outcomes:

1. Copy data from USB to local drive, then again from local drive to the server. You're copying the data twice. Even in best case, you're looking to copy every byte twice.
2. Copy data from USb directly to the server. Only copying the data once, and so you save on quite a bit of time. Not sure what acks you are going to argue about, but the acks delay is going to be the same reguardless, the question is whether you do the acks in 2 parts for #1 or 1 part for #2. But the time lost while actually writing the data in #1 makes it a slower solution.
With SATA, I'd wholeheartedly agree.

With USB 2.0 and 3.0 using old protocols, my experience has been less than ideal. Maybe I'm just unlucky or this might be a problem caused by high-latency networks (my desktop is connected via 802.11ac - throughput is somewhere between 100Mb/s Ethernet and GbE, but latency is good ol' WiFi latency (never really bothered to measure the internal speed)). In any case, in my experience, direct file transfers are often slower than an initial copy and transferring that copy.

Maybe some of those fancy new SCSI over USB protocols solve this, but I haven't had the chance to try that out.
 

9C1 Newbee

Patron
Joined
Oct 9, 2012
Messages
485
If things go perfectly... cool. If things go sideways off slow media. I'd rather read once to local fast storage and not have to start from scratch. Too easy to hit an error or drop a connection the middle of a large transfer and have to restart. Remember your love of USB ;) It is also an old school problem imho. Both speed and reliability is much much better thus my "direct writing is cool".

You need more time in HELL, cyber ;). It wasn't long ago reading and writing over the wire to a remote detachable storage could be nasty and unreliable. If you've never had to copy locally to get crap hardware or something causing an error or timing out of the loop you are a lucky man. Can't tell you how many times windows has puked on me and I got to start all over.

Maybe you should look into TeraCopy as well. During said transfer of said 10TB, things did go Tango Uniform a few times. TeraCopy picks up where it left off. You also have the ability to add jobs to a que. Really neat stuff. Doing a large transfer using just the stock windows tools? Now THAT is just crazy talk.
 
Status
Not open for further replies.
Top