UNtil this morning, my UNRAID box was working perfectly. Now I am getting painfully slow read/write speeds. Normally I get around 80 MBps with my desktop(ethernet cable) and 12 MBps with my laptop(wireless N connecting at 150mbps). Today I was having problems deleting some files. The read/write speeds were normal (I am about 95% sure anyway) before I deided to reboot. I stopped sickbeard and sabnzbd, but the array could not unmount the disks. I was telneted in, so I used that to force a reboot. The sickbeard/sabnsbd installs were not via the unmenu package manager, instead I am using the .plg files, managing them under the UNRAID settings. It seems that I need to stop them there, rather than logging in and shutting sab/sick down with thier shutdown options, in order to stop the array.

After rebooting, I stopped the parity check that automaticly started, and then deleted the problem files. I tried to back up some files and that is when I got the slow write speeds with my laptop. Writing was about 1.8 MBps, reading was under 500 KBps. I thought maybe stopping the parity check had messed things up, so I rebooted again(after I safely stopped the array). I did not stop the parity check, and I tested the speed again after the parity check finished, and I was getting the same results. I stopped the array, removed the parity drive from the array, started, no change. Stopped array and added parity drive back.

I then tested with my desktop. 3MBps write, less than 500k read. I tested read/write to my laptop and speeds were normal. I tested speeds to shares, and to individual drives as well, and there was no change. I did a soft reboot on the router(3 month old linksys E4200) and that resulted in no change. Hard reset on the router, no change. I changed ethernet cables, no change(both cables I used were less than 6 months old and in good condition) I tried all the ethernet ports on the router, and there was no change.Sabnzbd still downloads and transfers the files like usual. Telnet is slow as well - when I type something in, it takes a second or two to catch up. Like it is lagging.

My hardware is a ALiveNF6G-DVI motherboard with AMD x2 3800+ cpu, 2g DDR2 RAM, gigabit ethernet, 4x sata2. My drives are a Hitachi 750 GB 7200 RPM cache drive, parity and disk 1 are both WDC_WD20EARS, disk 2 is WDC_WD10EADS. Flash drive is a 8g sandisk.

I have seached the forums and found one thread with a smiliar problem, but it was not able to help. I dont think that is is the onboard network card, but I dont know what else it could be. I dont think I can understand the info below about it to know. The disk information looks good to me.

I am running UNRAID Server Plus, version 5.0-beta13. Syslog is attached. Thanks in advance.

Below is the info from ethtool, ifconfig, and hdparm -tT on all drives, except for the flash drive.

root@NAS:~# ethtool eth0

Settings for eth0:

        Supported ports: [ MII ]

        Supported link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Supports auto-negotiation: Yes

        Advertised link modes:  10baseT/Half 10baseT/Full

                                100baseT/Half 100baseT/Full

                                1000baseT/Full

        Advertised pause frame use: No

        Advertised auto-negotiation: Yes

        Speed: 1000Mb/s

        Duplex: Full

        Port: MII

        PHYAD: 1

        Transceiver: external

        Auto-negotiation: on

        Supports Wake-on: g

        Wake-on: d

        Link detected: yes

 

root@NAS:~# ifconfig eth0

eth0      Link encap:Ethernet  HWaddr 00:13:8f:e7:5f:b2

          inet addr:192.168.1.105  Bcast:192.168.1.255  Mask:255.255.255.0

          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

          RX packets:994933 errors:0 dropped:12 overruns:0 frame:0

          TX packets:673245 errors:0 dropped:0 overruns:0 carrier:0

          collisions:0 txqueuelen:1000

          RX bytes:1470095892 (1.3 GiB)  TX bytes:73254754 (69.8 MiB)

          Interrupt:43 Base address:0x8000

 

 

Disk 1(WDC_WD20EARS)

root@NAS:~# hdparm -tT /dev/sda

 

/dev/sda:

Timing cached reads:  1444 MB in  2.00 seconds = 722.20 MB/sec

Timing buffered disk reads:  328 MB in  3.02 seconds = 108.76 MB/sec

 

Disk 2(WDC_WD10EADS)

root@NAS:~# hdparm -tT /dev/sdd

 

/dev/sdd:

Timing cached reads:  1470 MB in  2.00 seconds = 735.36 MB/sec

Timing buffered disk reads:  296 MB in  3.01 seconds =  98.21 MB/sec

 

Parity (WDC_WD20EARS)

root@NAS:~# hdparm -tT /dev/sdc

 

/dev/sdc:

Timing cached reads:  1438 MB in  2.00 seconds = 718.62 MB/sec

Timing buffered disk reads:  376 MB in  3.02 seconds = 124.45 MB/sec

 

Cache(ST3750528AS)

root@NAS:~# hdparm -tT /dev/sde

 

/dev/sde:

Timing cached reads:  1400 MB in  2.00 seconds = 700.17 MB/sec

Timing buffered disk reads:  302 MB in  3.01 seconds = 100.42 MB/sec

root@NAS:~#

 

syslog-2012-01-15.txt

The syslog looks ok.

 

1. Test with a different client first. Make sure that the client is not the problem.

 

2. Edit your go file:

 

#!/bin/bash
# Start the Management Utility
/usr/local/sbin/emhttp &
#blah blah
#blah blah

 

Make all lines start with a # except for the one with "emhttp". Then restart and test.

By try a different client, do you mean something like putty? Tried that, no change.

Made the changes to the go file, no change.

A different client is a different computer.

I tested the read/write from two different computers. My laptop wirelessly, and my desktop, connected via ethernet cable.

Was speed normal from the laptop? If so then the problem is in the desktop PC.

I stated in the OP that I tested with both the desktop and the laptop. If it were only so easy that it were not the server.

I'm not understanding. You said,

I tested read/write to my laptop and speeds were normal.

I then tested with my desktop. 3MBps write, less than 500k read. I tested read/write to my laptop and speeds were normal.

The part that you quoted was right after I stated that I tested the desktop and got the slow speeds, then I tested the speeds to my laptop, from the desktop, and they were normal.

So, desktop to server = crappy. Desktop to laptop = good. Laptop to server = crappy. Laptop to desktop = good.

Please post a new syslog with the go file edited as described; all add-ons should be disabled. Perform several large transfers before collecting the syslog.

I did not figure this was the problem, but I put in a new gigabit NIC today, and for the last several hours, it has been working just fine.