brand new md-1510 keeps locking up

[res0nat0r]

I just got this a couple of days ago and it seems to have locked up 3x sofar, I think it is when NFS traffic is occuring.

 

I still have my monitor plugged into this box and havent been able to catch any error messages when i tail -f syslog, is there any persistent syslog file

stored somewhere i could look at that survives reboots? i am not too familiar with the unraid os layout yet to know where to look. the box seems to hardlock and i cant even

switch between virtual terminals when this happens and have to power cycle it.

 

i dont think it is heat or anything such, as last night i ran 3 parallel pre_clears on the 3 drives i have installed and it was up fine this morning.

 

i bought a cheap netgear gig router to put between this and my ubuntu based XBMC machine, and when i tried to rsync my initial data everything seemed

to lock up after a bit. i have seen a few posts related to this, but am getting this error on my xbmc box:

 

eth0: too many iterations (6) in nv_nic_irq

 

i think this is more of an ubuntu problem and not the unraid box so i can figure that out later. but i took out the router from the picture and everything is just

currently connected at 100/full, but i tried to load up a movie over nfs on my xbmc box from my unraid and when i skipped around in the movie it looks like it caused the unraid box lock up.

 

any guesses where to begin looking? i have all of the smb options off in my config and am using nfs for my shares, and am using 4.4.2 pro

 

thanks

[NAS]

Immediately I would say that you should contact Limetech via email. Whilst this is the official support forums IRL there has been very little official support here.

 

I mean nothing by this other than you paid good money which includes commercial support and you can get that using email.

 

By all means use this forum as well but Limetech should be your first port of call.

[res0nat0r]

I am trying once more to just do a big initial copy of all of my data to the box. if it messes up at all i am going to send an email...

[RobJ]

eth0: too many iterations (6) in nv_nic_irq

 

This particular line is reported by the forcedeth driver, used by the onboard nVidia nForce NIC.  I get it too if I am pushing a large amount of traffic through the network interface.  I have never seen a problem such as you are reporting, it has simply seemed to indicate an internal limit to how much the chipset or driver could handle.

 

What I would suggest is to try adding a higher performance network card, such as an Intel PRO/1000 PCIe card (and disable the onboard NIC).  (No guarantees that this will fix the lockups though.)

[res0nat0r]

I copied about 1TB of data to the box over the weekend and it didnt lock up at all. this was not using my gig switch though. i am copying about 300 more GB using the gig switch now to the unraid box and it seems ok sofar.

 

i added this to /etc/modprobe.d/forcedeth.conf

 

options forcedeth max_interrupt_work=20

 

i updated my initramfs on my ubuntu box and rebooted, and now rsyncing the data isnt causing the error message above. also i dont have any TX/RX errors

via netstat -i

 

i am not sure what happened before, if this network issue caused the network just to crap and i thought it was the unraid box or not, but i did have a kbd and

mouse hooked up to the thing before it stopped responding, but i dunno....but things look ok as of now.