Unexpected Shutdown when initializing iSCSI Volume

[nsc]

Hi,

I have 2 OpenE DSS, only one is affected by this problem. Both DSS have exactly the same hardware. This problem appeared in all version (3278, 3511 32bit or 64 Bit) and it's now easy to reproduce :
create a big iSCSI volume (1.5To) with initializing and wait... At random %, the serveur shutdown. No screen error, no message, nothing jut a big shutdown. Whern i restart server, initializing restart from current pourcentage not zero...
And there is no log in HTTP GUI about this problem.

I have a ticket opened to support but no response at this time.

I think it an hardware problem, but which one ?

Motherboard are Tyan, Controler Areca 1680 with 11 SAS disks.

NSC

[To-M]

You said that you have 2 DSS servers but only one does this shutdown and not both of them?

What is the ticket #?

[nsc]

Hi Tom,

Ticket is 1009172, i've got an answer now, i answered too.

Yes i have 2 DSS servers with exactly the same hardware.

With the second one, i never had this problem but may be there are less load because it was used as backup (but like my defective one, i updated it with USB Stick and i created atleast 3 times iSCSI volume on it).-

[To-M]

Just checked the logs and it does look like a slow raid and the cpu load is very high - check the cache settings on the RAID controller.

[nsc]

I check ARECA GUI of my 1680, i don't have a lot of option :

Volume Option:
Volume Stripe Size : 64 KB (or 4, 8 16, 32, 128)
Volume Cache Mode : Write Back (or Write Through)
Tagged Queuing : Enabled

System Option:
SATA NCQ Support : Enabled
HDD Read Ahead Cache : Enabled (or disable Maxtor, Disabled)
Volume Data Read Ahead : Normal (or aggressive, convervative, disable)
HDD Queue Depth : 32 (or 1 2 4 8 16)
Disk Write Cache Mode : Auto (or Enabled Disabled)

And some others options, but i don't think they are "Cache" related.

I have to play with Write Back / Write Through first ?

[To-M]

Use the Write Back feature first.

[nsc]

Write Back is my current config (the one with shutdown...)

May be i could "stress" my second DSS to be sure that is a problem hardware first ?

And to remember, i got a shutdown when applying update 3511, i can suppose that while copying to USB Stick to intern USB the RAID Unit is not used...

[To-M]

When updating with DSS 3511 the system is down so there is no effect on the RAID controller on on the USB DOM.

[nsc]

I got the same shutdown, at 92% of copy, shutdown...

So i could suppose that i have another problem right ?

After this shutdown i made a 4 hours Memtest without any errors...

I could contact my vendor for garanty, but first i need to know which part of the server is deffective...

[To-M]

Did you test with a smaller size, maybe 100GB to see if it crashed again?