C2750 Configuration Question

[FloridaDan]

Greetings and salutations!

First off I want to say thanks for the awesome guides and "how-tos". Between you guys and the manual, there's a ton of good stuff available to those of us who are just starting out with FreeNAS.

So what brings me here to today is a bit of a head-scratcher with my C2750. If I set my signature up right, the rest of the configuration is there. The drives are not yet in a pool, they're still being tested.

The history on this is that my current board is my second copy of this part. I had to RMA the first copy as it seemed intent on filling up the Event Log every time the system booted. The second copy behaves much more "normally" in this respect, but I'm still seeing a few things with which I'm not completely at peace. I've been reading over in the ASRock forums, but have found that the discussion is sketchy at best, so I figured I'd try and picking the brains of those here who use this board (either in their own builds or in the FreeNAS Mini, as I understand it's built around this part also) to see how their experiences compare with mine.

What I have found is that there are a couple of disturbing events in the Event Log. Both appear to be tied up with BIOS settings surrounding CPU speed and sleep.

The first is:

Code:

System Reconfigured,OEM System Boot Event,Undetermined System Hardware Failure,Entry Added to Auxiliary Log,PEF Action

This message shows up while the board is POSTing, and appears to be related to the BIOS setting for Intel C States. If I disable C States, this message goes away. I'm not really concerned about not having the deeper states as apparently FreeNAS doesn't use them (for reasons that make sense). I'm including this here as a part of my basic sanity check.

The second message is the one that has me scratching my head a bit.

Code:

FRB2/Hang in POST Faliure

Yes, that's how it is spelled in the log, and this also occurs during POST.

The board does complete POST, boots, and runs in a stable way.

The only things that I have been able to come with on this one are that some guys trying to run Debian were seeing this related to a kernel panic. In picking through what there was on this, it looks like disabling SpeedStep took care of this one.

I'm in the middle of a badblocks run on my drives at the moment so I can't go test that for myself, yet. I'll be doing that as soon as I can. However, I'm much more concerned about turning this off...

In the meantime, I wanted to see what experiences you've had with these settings and messages. Does this all line up with what folks here have been seeing, or have I hit the RMA jackpot and gotten two bad copies of this part in a row? I'm skeptical of that last one, as in 25+ years of building systems, that first copy was the first RMA I've ever had to initiate where there was no obvious physical damage to the part when I received it, so I find it very unlikely that I got two bad copies in a row.

Thanks in advance for any input that you may have!

 

[johnnyTragic]

FloridaDan, it seems I am have having the same issue. At this point i can't figure out if it is simply bad build Quality or just a new board that they are still trying to refine. I have also submitted an RMA as well. my confidence in this MOBO is not very good at this time. the customer support is pretty good though. in going through several weeks of customer support they do not seem to recognize any other OS outside of windows, so it seems the other "supported" OSs are not something they test, at least that is my experience in my various communications with them. to this point i have installed Free BSD 10.x, FreeNAS 9.x, Unbuntu Server 11.x, CentOS 6.5.x. all run very well on this with most of the BIOS options disabled. It seems i am also having an issue with the Ethernet ports as well ( dropping ) hopefully the new one will be more stable, if not; I'm cutting my losses and moving to another manufacturer.

122605/28/2014 21:04:13UnknownSystem EventSystem Reconfigured,OEM System Boot Event,Undetermined System Hardware Failure,Entry Added to Auxiliary Log,PEF Action - Asserted
122505/28/2014 21:04:13UnknownSystem EventSystem Reconfigured,OEM System Boot Event,Undetermined System Hardware Failure,Entry Added to Auxiliary Log,PEF Action - Asserted
122405/29/2014 01:04:12UnknownSystem EventSystem Reconfigured,OEM System Boot Event,Undetermined System Hardware Failure,Entry Added to Auxiliary Log,PEF Action - Asserted
122305/29/2014 01:04:12UnknownSystem EventSystem Reconfigured,OEM System Boot Event,Undetermined System Hardware Failure,Entry Added to Auxiliary Log,PEF Action - Asserted
122205/29/2014 01:03:58UnknownProcessorFRB2/Hang in POST Faliure - Asserted

 

[FloridaDan]

So I know the feeling. I was pretty disgusted with my first copy of the board, and frankly you've had a very different experience with ASRock support than I did. The automated e-mails from their "contact us" system showed up something like a week after I put the tickets in, and only after I found them on Face Book of all places and PM'd them there. I never have heard anything from any actually in tech support with them.

Reading their forums, it does seem that they're very Windows centric, but there are a smattering of folks who stray off into other realms. They were able to provide the advice that got rid of "Unknown System Event". (Turn off C states.)

Turning off speed stepping seems to work for some, but not for others on the "Hang in POST". However, the symptoms really are different, when shutting that off makes the message go away.

Given that both copies of the board I've had have generated that message. (At one point I ploughed through the piles of the logs the first one created and I was able to see those messages in there too.)

That all said, I'm not really displeased with the board. The reality is that even though the board whines about POST getting stuck, it does in fact POST and run in a stable way. It correctly reports all the hardware it should, at least that I can see and tell, and runs as described. I'm still trying to track down whether I'm comfortable with my observations and am willing to trust my system to them.

If that message is spurious, this is an excellent example of where just a little more care on the part of a programmer would have made all the difference in the world; if that message wasn't there, we wouldn't be having this conversation.

I'd really be interested in hearing from someone who has a Mini based on this board, or from one of the guys running the Supermicro variant. I still find it hard to believe that I got two bad copies in a row.

Please share how your experiences come out.