NV+ Slow Transfers

[icgbob]

Like many I've seen here I have an NV+ (256mb) with terrible transfer rates. Previously I had 4 Seagate 7200.1 320gb drives. Yesterday I installed 4 x Seagate ST2000DL003-9VT166 from HCL (I have a memory upgrade on order) and did a complete factory reset as I've seen recommended in the forums due to block size. I set the drives to X-RAID and all are running fine. The issue is the horrific transfer speeds. Like many I 'm getting speeds that would be appropriate if I had a 100mb LAN, however, I'm on a wired Gigabit LAN with CAT6 cables all through the network.

I've tried every trick posted here and around the net with no improvement. Tweaking the NIC after installing latest drivers, changing cables, MTU speed changes, turn this on or that off. I even installed an second Intel Pro1000 NIC with no improvement. You name it, I've tried it.

I use the NAS here at the house for both business and pleasure - On all the PCs the My Documents folder (and others) defaults to the NAS so I always everything configured the same no matter where I'm at in the house. I had it running for years with the old drives and simply needed more space due to DVDs, hence the upgrade.

The PCs are high end with SSDs OS/Program drives and fast work drives using fastest AMD processors all with over 8gb or ram. The particular on in my office has 16gb with 2TB WD and 2TB Seagate work drives.

Prior to the upgrade I got so-so speeds to and from the NV+ - I'd say on average about 40MB/s. I never concerned myself with the speed issue as that is more than fast enough to use anything from it including music and movies or typical file transfers.

However, since the upgrade, my transfer rates are never above 12MB/s - usually around 5MB/s. While I can actually work with that, putting all the DVDs, music and such back on is rediculously slow as you can imagine.

I've done test transfers (huge ones) from one pc to the other across the network (which has Netgear Gigagit Switches and a Motorola Cable Modem with 4 Gigabit ports). The speeds are well over 100MB/s - typically around 111MB/s. That's awesome and I'd be done loading my NV+ in an hour or less at those speeds. However, at the 5MB/s I'm getting to the NV+ it will take me days and days to load everything.

As I stated, I've tried everything posted here and around the net. However, just now I noticed in the Fadiator Home page it says 1000MB network speed. But, when I went to the Network management page and checked the Speed options (under Network/Interfaces) it only offers options for Auto Negotiate, 100MB Full Duplex and 100MB Half Duplex. There are no settings for 1000MB (1GB). I have RAIDiator 4.1.7 [1.00a043] installed. I tried the Update tab and it says I have the most current version.

Settings wise Jumbo is turned off (though I've had it on across the network during testing), I have OpLocks on for some shares and Off for others to see how things change, Fast CIFS writes on, and Full Data Journaling Off.

If anyone has an idea other than the common ones posted please let me know. I'm going to call Tech Support on Monday but, hope to have it resolved before then. I need to get this handled as the Daughters and Grand Kids will be here in a week.

Speaking of that. Only one PC is typically used at a time as it's just me, unless a daughter and her laptop stop by.

Cheers and thanks in advance for the assistance.

Updated: I forgot to post that I have Windows 7 (Pro / Home) 64b on all machines, clean logs (few errors), no Smart issues. Transfers are smooth and consistent, it's simply the speed that sucks.

Update 2: I just threw a very fast USB 3.x drive on the NV+ and get only 5MB/s transfer that way, too. Previously USB was about 30MB/s

This is what I found in the Network Log. Nothing to write home about.
eth0 Link encap:Ethernet HWaddr 00:0D:A2:01:2D:5E
inet addr:192.168.0.3 Bcast:192.168.0.255 Mask:255.255.255.0
UP BROADCAST RUNNING MULTICAST MTU:1430 Metric:1
RX packets:37065923 errors:0 dropped:0 overruns:0 frame:0
TX packets:3585098 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:3106651402 (2.8 GiB) TX bytes:1069654480 (1020.1 MiB)
Interrupt:41

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:83558 errors:0 dropped:0 overruns:0 frame:0
TX packets:83558 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:6406664 (6.1 MiB) TX bytes:6406664 (6.1 MiB)

JUMBO_FRAMES_eth0=0
MTU_eth0=1430
SPEED_DUPLEX_eth0=AUTO_NEGOTIATION

[TeknoJnky]

strange, I'd suggest a direct connect between the nas and a PC (use static ips) and see if there is any change.

[icgbob]

I'll do that tomorrow and report. I wish I had a previous version of Radiator to try out.

[icgbob]

On a direct connection I get all the way up to a whopping 15MB/s. Other ideas. Could it be an issue with this Radiator version. I've been looking for some older ones to try with no luck.

[TeknoJnky]

4.1.7 works fine for me.

do you have any other spare drives? you could power down, remove the current drives (lable their slot), factory default with a spare disk and see if the performance is any different.

[icgbob]

I don't have an identical spare. I do have another 2gb (Seagate) but that model isn't on the HCL. I'll have to see about getting another.

Anyone know why the only network interface options are Auto, 100mb FD & 100mb HD? Seems like it ought to have 1000gb choices also.

[mdgm]

Auto should give you gigabit provided you use Cat5e or newer ethernet cables and gigabit network hardware.

[stencorp69]

I am experiencing the exact same issue. PC to PC 111MB transfer, PC to NV+ 5-10MB. I have changed cables to device and changed ports with no improvement. Was there a resolution to this issue?

thanks

Stephen

[mdgm]

stencorp69 would have been better to start a new thread rather than reviving a dead one.

What brand and model disks are you using?
What version of RAIDiator?
What OS are you using on your client machine (e.g. Windows XP Professional).

Have you tried a direct-connection?

What does your partition.log (Status > Logs > Download all logs, then extract zip contents) look like?

[stencorp69]

Hi,

Sorry my bad, the issues I'm having are with the same device and appear to be the same issues. So I was following this thread up to see if there was a resolution for the issue.
What brand and model disks are you using? NV+
What version of RAIDiator? 4.1.8
What OS are you using on your client machine (e.g. Windows XP Professional). Windows 7 32bit
Have you tried a direct-connection? Yes
What does your partition.log (Status > Logs > Download all logs, then extract zip contents) look like?
Disk /dev/hdi doesn't contain a valid partition table

Disk /dev/hdc: 2000.3 GB, 2000388448256 bytes
255 heads, 63 sectors/track, 243200 cylinders, total 3907008688 sectors
Units = sectors of 1 * 512 = 512 bytes
Disk identifier: 0x00000000

Device Boot Start End Blocks Id System
/dev/hdc1 32 4096031 2048000 83 Linux
Partition 1 does not end on cylinder boundary.
/dev/hdc2 4096032 4608031 256000 82 Linux swap / Solaris
Partition 2 does not end on cylinder boundary.
/dev/hdc3 4608032 3906992335 1951192152 5 Extended
/dev/hdc5 4608040 3906992335 1951192148 8e Linux LVM

Disk /dev/hde: 2000.3 GB, 2000388448256 bytes
255 heads, 63 sectors/track, 243200 cylinders, total 3907008688 sectors
Units = sectors of 1 * 512 = 512 bytes
Disk identifier: 0x00000000

Device Boot Start End Blocks Id System
/dev/hde1 32 4096031 2048000 83 Linux
Partition 1 does not end on cylinder boundary.
/dev/hde2 4096032 4608031 256000 82 Linux swap / Solaris
Partition 2 does not end on cylinder boundary.
/dev/hde3 4608032 3906992335 1951192152 5 Extended
/dev/hde5 4608040 3906992335 1951192148 8e Linux LVM

Disk /dev/hdg: 2000.3 GB, 2000388448256 bytes
255 heads, 63 sectors/track, 243200 cylinders, total 3907008688 sectors
Units = sectors of 1 * 512 = 512 bytes
Disk identifier: 0x00000000

Device Boot Start End Blocks Id System
/dev/hdg1 32 4096031 2048000 83 Linux
Partition 1 does not end on cylinder boundary.
/dev/hdg2 4096032 4608031 256000 82 Linux swap / Solaris
Partition 2 does not end on cylinder boundary.
/dev/hdg3 4608032 3906992335 1951192152 5 Extended
/dev/hdg5 4608040 3906992335 1951192148 8e Linux LVM

Disk /dev/hdi: 2000.3 GB, 2000388448256 bytes
255 heads, 63 sectors/track, 243200 cylinders, total 3907008688 sectors
Units = sectors of 1 * 512 = 512 bytes
Disk identifier: 0x00000000

disks are on the HCL list

[mdgm]

That looks good. So what difference did a direct-connection make (if any)?

You might wish to open a tech support (http://www.readynas.com/support) case.