[RESOLVED: I found out after much testing that the onboard gigabit LAN (Atheros AR8131 chipset I believe) on the Asus M4A78LT-M LE motherboard IS NOT COMPATIBLE WITH UNRAID. I've swapped it out with a separate NIC and everything works fine now]

 

Hi guys,

 

I just built my UnRAID array and I was thrilled when I was hitting 50mb/s speeds when copying files into the array.

 

However my joy was short lived when I soon found out that after I copied stuff in, I couldn't read them, explorer just jammed up. I can hardly even browse the directories without explorer slowing to a crawl.

 

I've been up for 2 hours trying to find info on the forums but I'm pretty new at this and I've run out of ideas.

 

Here's my syslog. I'm running an Asus M4A7BLT-M LE and currently running 2 brand new WD Caviar Green 1.5TB drives (without parity). I'm running unRAID 4.5.3.

 

Apr 19 02:50:28 ELhub ifplugd(eth0)[1313]: Program executed successfully.
Apr 19 02:50:28 ELhub rpc.statd[1359]: Version 1.1.4 Starting
Apr 19 02:50:28 ELhub emhttp: unRAID System Management Utility version 4.5.3
Apr 19 02:50:28 ELhub emhttp: Copyright (C) 2005-2009, Lime Technology, LLC
Apr 19 02:50:28 ELhub emhttp: Unregistered
Apr 19 02:50:28 ELhub emhttp: Device inventory:
Apr 19 02:50:28 ELhub emhttp: pci-0000:00:11.0-scsi-0:0:0:0 host1 (sda) WDC_WD15EARS-00Z5B1_WD-WMAVU2783927
Apr 19 02:50:28 ELhub emhttp: pci-0000:00:11.0-scsi-1:0:0:0 host2 (sdb) WDC_WD15EARS-00Z5B1_WD-WMAVU2938717
Apr 19 02:50:28 ELhub emhttp: pci-0000:00:11.0-scsi-2:0:0:0 host3 (sdc) WDC_WD15EARS-00Z5B1_WD-WMAVU2784587
Apr 19 02:50:28 ELhub emhttp: pci-0000:00:11.0-scsi-3:0:0:0 host4 (sdd) ST3500320AS_9QM4J77N
Apr 19 02:50:28 ELhub emhttp: pci-0000:00:14.1-ide-1:0 ide1 (hdc) ST3500630AS_5QG0X9W4
Apr 19 02:50:28 ELhub emhttp: pci-0000:00:14.1-ide-1:1 ide1 (hdd) WDC_WD5001AALS-00E3A0_WD-WCATR0566648
Apr 19 02:50:28 ELhub emhttp: shcmd (1): rmmod md-mod >>/var/log/go 2>&1
Apr 19 02:50:28 ELhub emhttp: _shcmd: shcmd (1): exit status: 1
Apr 19 02:50:28 ELhub emhttp: shcmd (2): modprobe md-mod super=/boot/config/super.dat slots=0,0,8,32,8,16 >>/var/log/go 2>&1
Apr 19 02:50:28 ELhub kernel: xor: automatically using best checksumming function: pIII_sse
Apr 19 02:50:28 ELhub kernel: pIII_sse : 11443.200 MB/sec
Apr 19 02:50:28 ELhub kernel: xor: using function: pIII_sse (11443.200 MB/sec)
Apr 19 02:50:28 ELhub kernel: md: unRAID driver 0.95.4 installed
Apr 19 02:50:28 ELhub emhttp: Spinning up all drives...
Apr 19 02:50:28 ELhub kernel: md: disk0 removed
Apr 19 02:50:28 ELhub kernel: md: import disk1: [8,32] (sdc) WDC WD15EARS-00Z WD-WMAVU2784587 offset: 63 size: 1465138552
Apr 19 02:50:28 ELhub kernel: md: import disk2: [8,16] (sdb) WDC WD15EARS-00Z WD-WMAVU2938717 offset: 63 size: 1465138552
Apr 19 02:50:28 ELhub kernel: mdcmd (2): set md_num_stripes 1280
Apr 19 02:50:28 ELhub kernel: mdcmd (3): set md_write_limit 768
Apr 19 02:50:28 ELhub kernel: mdcmd (4): set md_sync_window 288
Apr 19 02:50:28 ELhub kernel: mdcmd (5): set spinup_group 1 0
Apr 19 02:50:28 ELhub kernel: mdcmd (6): set spinup_group 2 0
Apr 19 02:50:28 ELhub kernel: mdcmd (7): spinup 1
Apr 19 02:50:28 ELhub kernel: mdcmd (: spinup 2
Apr 19 02:50:29 ELhub emhttp: shcmd (3): /usr/local/sbin/set_ncq sdc 1 >/dev/null
Apr 19 02:50:29 ELhub emhttp: shcmd (4): /usr/local/sbin/set_ncq sdb 1 >/dev/null
Apr 19 02:50:29 ELhub kernel: mdcmd (10): start STOPPED
Apr 19 02:50:29 ELhub kernel: unraid: allocating 18220K for 1280 stripes (3 disks)
Apr 19 02:50:29 ELhub kernel: md1: running, size: 1465138552 blocks
Apr 19 02:50:29 ELhub kernel: md2: running, size: 1465138552 blocks
Apr 19 02:50:29 ELhub emhttp: shcmd (5): mkdir /mnt/disk1
Apr 19 02:50:29 ELhub emhttp: shcmd (5): mkdir /mnt/disk2
Apr 19 02:50:29 ELhub emhttp: shcmd (6): mount -t reiserfs -o noacl,nouser_xattr,noatime,nodiratime /dev/md1 /mnt/disk1 >/dev/null 2>&1
Apr 19 02:50:29 ELhub emhttp: shcmd (7): mount -t reiserfs -o noacl,nouser_xattr,noatime,nodiratime /dev/md2 /mnt/disk2 >/dev/null 2>&1
Apr 19 02:50:29 ELhub kernel: mdcmd (12): check 
Apr 19 02:50:29 ELhub kernel: md: recovery thread woken up ...
Apr 19 02:50:29 ELhub kernel: REISERFS (device md1): found reiserfs format "3.6" with standard journal
Apr 19 02:50:29 ELhub kernel: REISERFS (device md1): using ordered data mode
Apr 19 02:50:29 ELhub kernel: REISERFS (device md2): found reiserfs format "3.6" with standard journal
Apr 19 02:50:29 ELhub kernel: REISERFS (device md2): using ordered data mode
Apr 19 02:50:29 ELhub kernel: REISERFS (device md2): journal params: device md2, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
Apr 19 02:50:29 ELhub kernel: REISERFS (device md2): checking transaction log (md2)
Apr 19 02:50:29 ELhub kernel: REISERFS (device md1): journal params: device md1, size 8192, journal first block 18, max trans len 1024, max batch 900, max commit age 30, max trans age 30
Apr 19 02:50:29 ELhub kernel: REISERFS (device md1): checking transaction log (md1)
Apr 19 02:50:29 ELhub kernel: md: recovery thread has nothing to resync
Apr 19 02:50:29 ELhub kernel: REISERFS (device md2): Using r5 hash to sort names
Apr 19 02:50:29 ELhub kernel: REISERFS (device md1): Using r5 hash to sort names
Apr 19 02:50:29 ELhub emhttp: shcmd (9): rm /etc/samba/smb-shares.conf >/dev/null 2>&1
Apr 19 02:50:29 ELhub emhttp: _shcmd: shcmd (9): exit status: 1
Apr 19 02:50:29 ELhub emhttp: shcmd (10): cp /etc/exports- /etc/exports
Apr 19 02:50:29 ELhub emhttp: shcmd (11): killall -HUP smbd
Apr 19 02:50:29 ELhub emhttp: shcmd (12): /etc/rc.d/rc.nfsd restart | logger
Apr 19 02:53:22 ELhub login[2992]: ROOT LOGIN on `tty1'
Apr 19 02:53:31 ELhub kernel: usb 3-2: USB disconnect, address 2

 

Do let me know if you have any ideas as to what might be wrong...

 

Thanks a lot!

Ervine

Just to add on..

 

I've fired up unRAID Plus and now have 6 drives, 4 storage, 1 parity, 1 cache.

 

However the problem is universal regardless of which drive/user share I put files into.

 

I can copy stuff over ONCE, after which explorer just has trouble accessing the share and slows down to a crawl. Even copying over a small 200kb JPEG jams up everything and I have trouble accessing any share for a few minutes. (During this time the //tower interface works perfectly, as does unmenu).

 

I even tried accessing the shares with my laptop running windows XP, exactly the same problem.

 

Anyone has any idea what's going on?

Here's the full syslog.

 

I hope someone has an idea of what's wrong...  :'(

syslog.txt

Here's the full syslog.

I hope someone has an idea of what's wrong...  :'(

 

Apr 19 08:52:08 ELhub kernel: ACPI Error (psargs-0359): [ECEN] Namespace lookup failure, AE_NOT_FOUND
Apr 19 08:52:08 ELhub kernel: ACPI Error (psparse-0537): Method parse/execution failed [\] (Node c14720c8), AE_NOT_FOUND
Apr 19 08:52:08 ELhub kernel: ACPI Error (dswload-0781): [PRID] Namespace lookup failure, AE_ALREADY_EXISTS
Apr 19 08:52:08 ELhub kernel: ACPI Exception: AE_ALREADY_EXISTS, During name lookup/catalog (20090903/psloop-230)
Apr 19 08:52:08 ELhub kernel: ACPI Error (psparse-0537): Method parse/execution failed [\] (Node c14720c8), AE_ALREADY_EXISTS
Apr 19 08:52:08 ELhub kernel: ACPI: Marking method \___ as Serialized because of AE_ALREADY_EXISTS error

 

You may have buggy ACPI in the BIOS.  Check if there are any BIOS updates for your mobo, or try booting with the acpi=off option in your syslinux.cfg

 

---

Thanks for the help!  ...  There was a BIOS update dated 3/1/10, and now the ACPI errors are gone.  

I'll be keeping an eye on the syslog.  Thank you guys again for your help.

Well i've disabled ACPI and so far my "reads" are synced up. It hasn't failed yet so i'm crossing my fingers

 

Thanks for the help Purko!

 

Unfortunately, I've tried both updating the bios and adding the acpi=off in the syslinux.cfg but the situation is still the same (although the error in the syslog is no longer there).

 

I've attached the new syslog.

 

I'm just taking a stab in the dark here, but could it be something to do with my router or my onboard gigabit LAN?

syslog_afterflashbios.txt

Interestingly, when I run File Browser via UnMenu, I can access some of the files.

 

Via regular Windows Explorer I can't even open a JPEG in the shares without jamming everything up!! I don't have this problem with File Browser via UnMenu, but once it's a bigger file (say 3MB mp3) then it too will jam up.

I see you have a few of the WD15EARS in the server.  Did you put the jumper on pins 7-8?

 

(CAUTION: Putting that jumper AFTER you've copied files to the disk will lead to the loss of those files!)

 

Did you use the preclear tool on those disks?

 

Try to isolate if it's a network problem or a disk problem.

Telnet into your sever and try copying stuff around with the mc tool. What speeds do you see?

 

Also, how about this command...

hdparm -tT /dev/[sh]d?

 

Can you get a SMART report on the disks?

smartctl -a /dev/sdX

(replace "X" with a disk letter)

 

 

Thanks again Purko,

 

I didn't know about that jumper thing! But I tried it and it still doesn't work.

 

But I have to admit, I got lazy and skipped the pre-clearing. I'm going back and doing everything again from scratch. Hopefully that will resolve these issues.

 

Preclearing is running at almost 90MB/s, is this normal?

I didn't know about that jumper thing! But I tried it and it still doesn't work.

How do you mean you tried it?  What doesn't work?

 

Shut down the array, put jumper in, start array, reformat.

 

Still had problems reading the drives after..

OK, finish the full preclear, and pay attention to what it says at the end.

After that, do the other things I suggested above.

 

Ok sure thing, I'll try running just 2 completely precleared HDs (without parity) first.

 

If there are still problems then I'm assuming the issue lies elsewhere?

 

By the way, I realised that even writing to the flash drive results in the same problem.... Which is why I was guessing it's nothing to do with the HDs...

Precleared my 2 x 1.5TB drives, remounted, reformated, same thing...

 

Here's a screen shot from one of the drives after the pre-clear.

 

I really doubt it's a hard drive issue, what can go wrong with the network that might cause this?

Precleared my 2 x 1.5TB drives, remounted, reformated, same thing...

 

Here's a screen shot from one of the drives after the pre-clear.

 

I really doubt it's a hard drive issue, what can go wrong with the network that might cause this?

What are you seeing in that screen shot that looks wrong to you. 

 

All I see is that the disk heads were loaded twice onto the platters during the course of the pre-clearing operation.  It would have been an issue if the heads were not loaded onto the disk platters.

 

Joe L.

 

 

Yes I'm saying that there doesn't look like there's any problems with the hard drive.

 

Thus what else could be causing my problems. Even trying to open a JPEG file from the server jams it up.

Actually, looking closer, I see a disk temperature of 60C.  

 

That is blistering hot.  (60 degrees Celsius = 140 degrees Fahrenheit)

 

Many disks shut themselves down above 55C.

I suggest lots more air movement across the drive.  You may have a fan that has stopped spinning.

 

Joe L.

Hm... yeah, things do get a little hot here in Singapore, 60deg was at full load on the drive, ambient runs around 40

Hm... yeah, things do get a little hot here in Singapore, 60deg was at full load on the drive, ambient runs around 40

 

At full load you should be looking to keep the drive no higher than 50C and preferably no higher than 45C.

 

When i am doing a full parity check the hottest one of my drives gets is around 44C. Get some cooling in there or you will kill the drive ina  very short amount of time.

Ooo.. ok roger that... 

 

But that still doesn't explain why the array jams up the moment I try accessing any file in it.... 

For fun I tried copying a 31GB folder into the array... no problem at all...

 

 

Unfortunately the same can't be said about reading info off the array...

If you re-flashed the BIOS, you will want to remove the "acpi=off", not that it is causing your current problems.

 

Your system looks OK (except for that heat!), but as I believe others have said, you will want to confirm which machine is the source of the issue.  Try accessing the files from another machine on your network.  Then access and copy/move them on the server itself, see Transferring Files Within the unRAID Server.  That should isolate the source of the issue, whether it is a particular machine or not, and whether the network and router are at issue.

 

I would also log into your router and check the connected devices.  Make sure there are no conflicts with MAC addresses, machine host names (any other device named ELhub?), and IP addresses.

Hi again...

 

I did the transfer internally in the server, no problem, moved a few GBs between the array's harddrives.

 

I also tried accessing the server via my laptop, same problem, array jams up the moment I try to access files.

 

I'll be trying to use a gigabit NIC tomorrow to see if the onboard one is giving me problems. Any other suggestions?

 

Thanks again for the help so far!

Hi again...

 

I did the transfer internally in the server, no problem, moved a few GBs between the array's harddrives.

 

I also tried accessing the server via my laptop, same problem, array jams up the moment I try to access files.

 

I'll be trying to use a gigabit NIC tomorrow to see if the onboard one is giving me problems. Any other suggestions?

 

Thanks again for the help so far!

Different network cable, different router.

RESOLVED!!!!

 

I tried another router I had lying around, as well as swapping the cables around, nothing worked. The last thing I tried was to use a seperate NIC. After swapping to the NIC everything works as it should!  

 

For the record, the onboard gigabit LAN (Atheros AR8131 chipset I believe) on the Asus M4A78LT-M LE motherboard IS NOT COMPATIBLE WITH UNRAID.

 

Thanks so much for the help! Really appreciate it!

 

Also, I've replaced the slower chassis fan and added another one to further cool the system, thanks for the tip! (Drives now run at just over 40deg at full load)