I just order ADD-3800CU from Laptops4me.com here is the link move fast guys before they all gone! http://www.laptops4me.com/advanced_sear ... 1&x=9&y=10

_________________
MSS 4TB,2GB, 3800+,CFI port Multiplier 2TB ,APC
MV2120 1TB,D-Link DNS-323 1TB

So did it again, same ol' same ol'... After 8 hours and 15 minutes I took another printscreen, this time nothing strange happened.

I feel sorry for those who experiment problems, and I don't quite understand how come my system stays smooth while others shutdown. It is the same machine, isn't it. I know, that particular configurations make the MSS quite different among us, but tests have been made by some while removing their add-ins and the sort. I hope my experience can shade (not sure of this expression ) some light on the problem. I have 3 disks in the MSS, plugged on an APC, external enclosure etc. I use McAfee (no problem), and many add-ins.

I just performed two computer back-ups while Orthos was still running (more than 8½ hours now). I don't input pictures but I have them if someone wants to examine.

_________________
HP-ProLiant N40L 8GB WHS2011
HP-EX470 4GB Athlon-X2-BE-2350-(Stepping G1) Windows 8 (server build)

Ordered. Disregard the entirety of my BE2350 thread. This probably the best thing I've done all year. I can sleep again!




Edit: Got mine from: http://www.ewiz.com/detail.php?p=ADD-38 ... 7a6d89cb11

Mine will be stable on Monday!

Well, Orthos may have failed at 2.5 hours, but the first incremental back-up went off without a hitch this morning. I'll let her run all day today and see how it holds up.

I'd like to solicit thoughts on whether or not a Server Restore or a Factory Rest would help with stability? As VieuxJules noted, there is really no good explanation why one system can handle the upgrade whereas another would not. Sure there are possible variabilities in components/manufacturing, but software/configuration are variables that stand a better chance of being significantly different from one user to the next. My temp values look very good compared to what others have reported and they seem very stable. As an indicator of hardware health, they seem to be saying things are okay - albeit only for the limited parameters we are measuring.

The expression is "shed some light".

How did you determine that your error was on core #1?

I have run Orthos twice with the same result.
Test #1
At 2 hours 29 minutes 36 seconds I got this:
FATAL ERROR: Final result was B6510E36, expected 3AC73ABD
Hardware failure detected, consult stress.txt file

Test #2
At 2 hours 23 minutes 21 seconds I got the same error and msgs.

I could not find a file named stress.txt on my system. Also, I noticed that in the lower window Orthos had this;
Torture Test ran 2 hours 29 minutes 36 seconds - 0 errors, 0 warnings.

I was running CPUID Hardware Monitor during the tests and it indicates that the high temps were;
TMPIN0 94 C (201 F)
TMPIN1 114 C (237 F)
TMPIN2 114 C (237 F)

THRM 63 C (145 F)

core #0 58 C (136 F)
core #1 55 C (130 F)

I'll post pictures later in necessary.

Take a look at zxdavb's earlier posts... he was running into (what looks like) the same thing but somehow found a way to get it stable...

My error message was the exact same (i.e. Final result was XXXX , Expected YYYY, Consult stress.txt for results) and at the same time (~2hrs, 30 min). When you open the log, you'll see the reference to the core that cause the problem. I did not have CPUID Hardware Monitor running, but I did have SpeedFan up. Nothing was higher than 60C and that was the Northbridge - CPU was around 54C.

I'd say this is the EXACT same error that srx71 and I are experienceing. Can't wait to get back home to reattempt!

I'll give his solution (could it be called that) i.e. running multiple instances of Orthos a try this weekend. I wonder if there is something wrong with our processors that we should pursue with AMD. Any thoughts? I haven't had a chance to Google this yet so I don't know what that might yield in the way of information.

Most of zxdavb's posts concerning this are on pages 3 & 4 of this thread for those looking for them.

Weird.

I'm kind of at a loss for the issues you guys are encountering. My chip just worked. I think BullHead's BE-2300 & X2 3800 EE SFF 'just worked' too. Zero issues. Zero Shutdowns. 10Hrs Small-FFTs stable on the first try.

The OS is SMP proven... I think the talk of 'cpu-to-cpu-handshaking' issues or the suggestion that the software behaves unpredictably now that there are two CPUs available is a little far fetched....

So.. what could it be?... shouldn't be heat(sink)/paste related because your temps are normal... How's the quality of your electricity?

Maybe try OCCT?... They say its 'quicker' / more effective than Orthos (at detecting an unstable system)

I have not had any shutdowns. I'll check out OCCT as well.

Likewise, I've not had any random shutdowns since reconnecting the PCs yesterday. I had one incremental back-up (my Vista box) and one original back-up (my daughter's XP Home Edition box) complete this morning as scheduled with no hiccups. I was gone all day and have returned home to a seemingly happy MSS. It sure is quick on Console Connects and RDP - wicked fast as they say.

I agree that the temps of my system give no clues to as to the shutdowns (i.e. normal or within expectations.) Power is a non-issue for me as I have the MSS on an AVR UPS. As for the OS being SMP proven, it also had a data corruption bug....

I'm trying the Orthos variation that zxdavb used now. Edit: Scratch that... I'm using OCCT instead.

Well, the MSS shutdown after nearly 56 minutes of OCCT, just as it was going into Idle/Monitoring Mode. I tried again and it lasted for about 40 minutes before shutting down teh second time. I quickly changed the processor back to the Sempron and tried again, this time it made it through. I'm going to start ORTHOS tonight to see what it yields with a stock processor.

So at this point, it seems that the CPU upgrade path is not meant for my MSS and I'm not sure why. It may be OS/Software related, but most likely is in some obscure hardware component(s) that is/are beyond the reach of our monitoring. For sure it is nothing obvious like high temps.

For ymboc's records, my 3800+ failed with Orthos at 2-2.5 hours average and it failed OCCT at ~50 minutes as an average. I guess the question remains, "was it stable enough to have been functional?" Given that my demands are light (i.e. back-up, media streaming, light webshare), I may not have ever had any problems. I feel I've invested enough time for now and must get back to other projects. Maybe during the next rainy day I'll have time to tinker some more.