I have read the other threads and also want to report that i am losing my network connection and have to reboot the server when trying to copy more than a couple of GB at a time.  I tried the fix to the cfg file, and that did not work.  Here are my specs:

 

unRAID v4.2beta4

MB - Gigabyte 965p-ds3 -

RAM - 1gb DDR2 Ram

HDD - 4x500GB Seagate SATA

NIC - onboard NIC 10/100/1000

 

I have heard external NIC cards can help, so i will try one of those.  Just wanted to let ya'll know in 4.2beta4, this is still an issue.

 

Ruby

I have read the other threads and also want to report that i am losing my network connection and have to reboot the server when trying to copy more than a couple of GB at a time.  I tried the fix to the cfg file, and that did not work.  Here are my specs:

 

unRAID v4.2beta4

MB - Gigabyte 965p-ds3 -

RAM - 1gb DDR2 Ram

HDD - 4x500GB Seagate SATA

NIC - onboard NIC 10/100/1000

 

I have heard external NIC cards can help, so i will try one of those.  Just wanted to let ya'll know in 4.2beta4, this is still an issue.

 

Ruby

 

Hi Ruby,

 

I had the exact same issue and was resolved by adding nolapic in syslinux.

 

This is what the syslinux should look like in 4.2.4b

default menu.c32

menu title Lime Technology LLC

prompt 0

timeout 50

label unRAID OS

  menu default

  kernel bzimage

  append initrd=bzroot rootdelay=10 nolapic

label Memtest86+

  kernel memtest

 

 

could you confirm you're seeing this issue and you have the same append line?

 

Regards,

Mola

Yes, i am using the nolapic in the syslinux file, to no avail.  We will see what the Netgear NIC card does to the situation.  I have another 120 minutes until the parity sync completes. I will test after that.

 

 

Thanks!

ok...i did the cardinal sin in trying to figure out root cause.  Changed too many variables.  However, i have copied over 75GB without fail, so problem seems solved.  Here is what i did.

 

 

Modified the syslinux file as above

Added more RAM- 2gb total

got the GA311 Netgear NIC card

Turned off parity sync before copying data.  Will turn on after copy is complete.

 

 

Thats it, will report if more errors occur or will report success.

 

 

Ruby

The common factor with at least one other member is the Marvell 8053 GigE chip on your motherboard.

 

 

Bill

I'm new to unRAID and yesterday when i attempted to copy about 70GB over to the test server it lost net connection.  I installed a TredNET gigabit PCI network card and it seemed fine after that.

 

I have a sneeking suspicion that the problem was if not caused by, then exacerbated by the fact that one of the hard drives that i was testing with was logging errors.  I saw that, after switching NIC's. I then stopped it and installed a different HD. I didn't try going back to the onboard NIC though to verify this.  I was intending to upgrad anyway because the onboard one is 100mbit

ok...i did the cardinal sin in trying to figure out root cause.  Changed too many variables.  However, i have copied over 75GB without fail, so problem seems solved.  Here is what i did.

 

 

Modified the syslinux file as above

Added more RAM- 2gb total

got the GA311 Netgear NIC card

Turned off parity sync before copying data.  Will turn on after copy is complete.

 

 

Thats it, will report if more errors occur or will report success.

 

 

Ruby

 

 

Update:

 

After a week or two of copying files, i have had zero errors and zero lost network connections.  I have copied over 1TB of files in the last week.  So, problem seems solved.  I would say at this point, it would be the onboard Marvel NIC chipset that was the culprit as Bill suggested above.

 

Thanks all!!!!

 

Ruby

There is also an issue with the SAMBA "smbd" program in all versions of 4.2 up through and including beta4.

 

It was isolated and identified in this thread: http://lime-technology.com/forum/index.php?topic=1021.0

 

Now, this does not mean the "nolapic" option had no effect, or your network card was not also an issue, they might have been your solution, depending on your symptoms.

 

The "smbd" symptoms were intermittent loss of connectivity to the SAMBA shares when doing concurrent reads and writes to the unRaid shared drives.  We never lost connectivity via telnet.

 

All that said, you might still have a "lost network connection" issue on occasion until Tom either goes to samba 3.0.25c (not yet released), patches the buggy samba 3.0.25b version, or reverts to the older samba 3.0.24.

 

Several of us have extracted the 3.0.24 version of "smbd" from unRaid 4.1 for use in unRaid 4.2.  Details are in the thread listed above.

 

Joe L.