Lost access to freenas when transfer large files

[Sagitario]

Hello,

I use ntbackup from windows 2003 server to back directly to my freenas 8.0.2 server. The total size file is 1.2TB. Everything is running when till 1 month when the ntbackup stops at +/- 200GB receiving a lost of access to freenas.

I'm not very family with the 8 version, in 7 I have a system log in the GUI to see the error. How can I see it on 8?

I have six 2TB disks and the volume RAID5 (ZFS) status is HEALTHY.

Can someone give me some directions?

Regards,

EDIT: At the time that the lost of connection occurs I see in the report tab a 10 minutes clean space in all graphics.

 

[ProtoSD]

Sagitario,

Please search the forums better, this topic has been discussed several times already. Also you need to post your hardware and FreeNAS version information.

Thanks

 

[Sagitario]

Sorry...

FreeNAS-8.0.2-RELEASE-amd64 (8288)
CPU E5300 @ 2.60GHz
Gigabyte EP45T-UD3RL
2GB DDR3 1333
160GB IDE (System)
6x 2TB SATA (RAID)

I search but dont find nothing related....

 

[ProtoSD]

Here is something related, not with a solution. It sounds like a similar problem with large files failing to finish copying.

http://forums.freenas.org/showthread.php?3470-The-problem-with-the-message-by-the-rsync-files-larger-than-10-GB&highlight=backup+large+file+deltacopy

 

[Sagitario]

I have see that post title before but I'm not using rsync, that why I don't think that is related.

I don't understand why this is happen... it work fine in the past...

Thanks,

 

[Digidoc]

It may be because you don't have enough memory. 2GB really isn't enough for when you're using ZFS. I could be wrong, but FreeNAS7 is less demanding compared to FreeNAS8.

For the record, I have run FreeNAS8 on as little as 4GB of RAM and it did work, but it didn't run as smoothly as how it now runs on my newer system with 16GB.

 

[ProtoSD]

Sagitario,

I don't think the problem is related to rsync or ntbackup. The post I remember seeing said deleting the destination file and trying again helped. Sorry, I see so many posts I don't remember all of the details only that I saw them here. Here is another thread that could help with tuning ZFS for your limited memory, @digidoc is correct, you really need more RAM for ZFS, however since it was working before that makes it more difficult to diagnose.


http://forums.freenas.org/showthread.php?3606-How-I-did-ZFS-through-FreeNAS-on-a-32bit-proc-with-low-memory.

 

[Sagitario]

Hi,

Problem solve last night and running stable till now.

You right, the problem was low memory that reboot my server. I don't see it reboot because it is running remotely (yes yes... I know that to run a diagnose and troubleshoot I need to be seat in front of the server... but the location is not local).

Like you say since it was working before that makes it more difficult to diagnose and is that why I don't related to low memory.

For now zfs tuning work good with this config:

# /boot/loader.conf
vm.kmem_size="1024M"
vm.kmem_size_max="1024M"
vfs.zfs.arc_max="256M"

# /etc/sysctl.conf
kern.maxvnodes="400000"

By: http://blog.herbertm.ca/archives/315

Next week I will upgrade the freenas up to 6GB RAM. Hope this way can have file transfer more fast that current.

I have only one more question, with the 6GB I need to tune up ZFS or left it by default?

Thank you for the help.

 

[Sagitario]

Hello Again,

Alter 16h of transferring the freenas freeze!!! Now I have put myself in front of the server to see what happen. First I upgrade to 6GB RAM, then try transfer the files... and after a while I got a kernel panic. :\

I let the zfs tuning at same... is this the problem? With values I need to change?

Thanks in advance...

 

[William Grzybowski]

Post a photo of the kernel panic

 

[Sagitario]

Sorry for this late reply, but suddenly my motherboard die... I will change it to a new one and see if the kernel panic remains.

Thanks for the support.

 

[Sagitario]

Hello again,

My motherboard was back from repair... try to tranfer large files and the system reboot with the kernel panic.

When I came back here I see hat a 8.0.3 was out and try it.

Well... now the system is dead, after instalation the system stop with this error:

Fatal trap 12: page fault while in kernel mode
cpuid = 1; apic id = 01
fault virtual address = 0x4000000030
fault code = supervisor read instruction, page not present
instruction pointer = 0x20:0xffffffff805d3b2
stack pointer = 0x28:0xffffff8000034860
frame pointer = 0x28:0xffffff8000034890
code segment = base 0x0, limit 0xfffff, type 0x1b


Any help?

EDIT: The problem is in the RAM module, change it and everything work well...