Solved BC f4 error when heavily using my new HDD

[tritous]

Edit function seems to dislike me

Well my problems aren't over. Running something from my new disk killed the system, set all fans to max, and froze it that way. Took a power off/on to get control back. next to nothing on the event logs, but there was a message from 3 minutes before that from the atapi again: event 11 "The driver detected a controller error on \Device\Ide\IdePort2". from this point I don't know the best way forward, although I'll be keeping researching to see if this could be a driver error or not

 

[F5ing]

I take it you got the error after removing the problematic 250GB?

 

[tritous]

aye, this is after. I'm trying to research the fault but everything I seem seems to be either related to someone trying to use the disk as a boot drive (master Boot Record doesn't support large partitions, but as it's an aux this is automatically a GUID partition table, which supports way more). I've checked drivers, the driver for "disk drive" for windows hasn't changed since 2006, there are no BIOS updates for my mobo, there isn't a new firmware for my disk, no new drivers for any of my SATA controllers either.

I've seen a couple of stories about ASUS motherboards and Seagate drives having bit of a tiff, but I don't seem to have the problems others are having, Frankly I'm lost on this one.

 

[tritous]

update. Had this issue with most of my steam applications, and still getting the occasional ATAPI controller event too, although it seems less related to the crashes. Updating my graphics driver, I've been joyously reminded of just how bad Sapphire are for making cards incompatible with most ATI drivers and catalyst versions. I am currently trapped on safe mode playing the old trial and error step by step roll back to find the last stable version. with my connection and with 160mb driver packs...that's a 30 minute download per version. For reference, the crash I get from the wrong version is:

Problem signature:
Problem Event Name: BlueScreen
OS Version: 6.1.7601.2.1.0.256.1
Locale ID: 2057

Additional information about the problem:
BCCode: 116
BCP1: FFFFFA8006C874E0
BCP2: FFFFF880044086D8
BCP3: FFFFFFFFC0000001
BCP4: 0000000000000003
OS Version: 6_1_7601
Service Pack: 1_0
Product: 256_1

I've also found the following new errors:

sptd: "Driver detected an internal error in its data structures for ."
- <Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
- <System>
<Provider Name="sptd" />
<EventID Qualifiers="49156">4</EventID>
<Level>2</Level>
<Task>0</Task>
<Keywords>0x80000000000000</Keywords>
<TimeCreated SystemTime="2012-04-29T11:22:59.868007700Z" />
<EventRecordID>816215</EventRecordID>
<Channel>System</Channel>
<Computer>Millimaximaeous</Computer>
<Security />
</System>
- <EventData>
<Data />
<Binary>000000000100000000000000040004C0EF0000000000000000000000000000000000000000000000</Binary>
</EventData>
</Event>


I also get the following from the Service Control Manager
"The following boot-start or system-start driver(s) failed to load:
AsIO
AsUpIO
Avgldx64
Avgmfx64
discache
spldr
sptd
Wanarpv6"
The time for that looks like it might be from just after starting safe mode though


I would be really grateful if anyone has any ideas on root causes. Meanwhile...graphics driver downdates here I come

 

[F5ing]

Have you run memory tests?

 

[tritous]

no, I've not yet seen any reason to suspect my memory and it's not particularly old. I didn't think the RAM was likely to have any influence over the SATA bus? Do you have a recommended test in mind?

 

[F5ing]

Memtest86+, from www.memtest.org is included in the Ultimate Boot CD: Ultimate Boot CD - Overview. Burn the UBCD iso to CD, boot your suspect machine from it, and run the memory tests.

 

[tritous]

believe it or not...i dont have any blank cds in spite of having had 2 dvd dual layer writers. I don't think i've written one in about 8 years, not since flash drives.

I'm becoming increasingly suspect this has all been caused by steam doing an update that breaks it. After completely uninstalling it (it will take me about 3 weeks 24/7 to recover the apps on my connection) it now wont reinstally, claiming it doesnt have access rights to itself. I'm taking that up with steam support. Everything else seems to be working except for that, I still wish I knew for sure that this has been causing all the errors, but it's always hard to tell. I have heard stories recently of steam overwriting driver dlls though, if it's true it's feasible.

For now I'm going to keep things on the old windows default driver for my graphics card: it's always worked. I'll just have to keep an eye on the disk and make sure it really is stable

 

[F5ing]

Well, if you do feel the need to test the memory, and if you've got a spare USB flashdrive available, UBCD can be install on it and you can boot it from there. There are instructions on how to do that in the FAQ (link on the left side of that page).

 

[karlsnooks]

tritous,
What god-forsaken country are you in that you can not buy a CD? They are cheap and readily available all over europe.

For USB (from the memtest86+ site)
http://www.memtest.org/download/4.20/memtest86+-4.20.usb.installer.zip

 

[tritous]

I live in a country where the shops open after i start work and close before I finish. Boot USB is a preference to me, I'd been considering putting together a generic boot USB of tools for helping friends out but never got around to looking into it. ultimate boot cd (or usb) looks like exactly what I had been thinking of.

Only down side is these abnormal crashes are stopping me from being able to complete the download on my slow connection. Hardware monitors say my CPU hasn't passed 40C and GPU is a stable 58C, all voltages look stable enough to build a city on, and the ATAPI errors are completely desynchronised to the crashes. I've also had 2 crashes now where steam wasn't running, in one case where it wasn't even installed, in these cases when I was watching video.

I've tried a couple of graphics drivers now, old and new, it doesn't seem to have stopped the crashes. A tool I very much would be interested in: do you know of any utilities that can do an integrity check on GDDR?

 

[tritous]

Ok, finally get it run and find the bad news: I looks like I have memory issues. 3 addresses, single bit error, systematically (as the 3 addresses are very close I assume it's a single point error). Royal pain since it means a new ram pair.

I assume that this is very very capable of causing the sorts of errors i've seen without even getting a BSOD depending on what is writing to that area. It doesnt sound like the sort of thing that coulid relate to the SATA controller events though. I guess I should be grateful it's fairly high up in my RAM and I can get enough operation to order the new parts in.

 

[tritous]

Just had another crash, and I was particular to check the RAM pair I kept was the correct pair. I have no doubts there was an issue with the RAM, just like i was sure there was an issue with corrupt sectors in my 250GB disk, but I guess there is a 3rd issue somewhere too

 

[tritous]

I just ran sfc (system file check) and all my system files pass their verification checks. This means that all my drivers are as they were before: versions I know for fact worked before. My hardware hasn't changed except for the disk which seems to work, so I'm really lost now for an explanation

 

[tritous]

Update to my updated update (since edit still doesn't work for me). This time very interesting find:

Had a cut out, started up and got the flashing red alert saying one of the power adaptors for my graphics cards was not attached. Not something I'd expect to unplug from rebooting and a very worrying sign that it could be one of the expensive hardware parts.

The power adaptors were in on the gfx card, but were only 1/2 in and not clicked in place at the PSU end. It's a possibility that while plugging in the new hard disk my cable management took a knock or two.

I don't know if this could be a cause. It's possible that the couple of pins used to detect if the connector is attached were in, but two of the power pins were not, leading to the chip stalling every time it tried to do anything graphically intensive, it may also be unrelated to my random crashes. Does anyone have any thoughts on this hypothesis? Granted teh only proof it works is if it works, but if this is a viable reason I'll put my efforts on seeing if it's fixed, otherwise I'll keep trying to find a new point of analysis.

 

[F5ing]

Sounds reasonable.

Back to the RAM for a sec: Were you sure the problem was with the sticks versus the slots?

 

[tritous]

I can't think of any reason why the slots would fail on just 3 bits in side 3 particular addresses. If it happened to all addresses on that stick maybe.

I'm not sure if I hope the cause was my failure to utilise a basic 8 pin plug or not. as embarassing problems go this is between facepalm and facedesk.

 

[F5ing]

Yeah, you're probably right about the memory. And I take it all the contacts for it were in good shape? Just trying to rule out any more issues due to memory.

So everything seems in order now that the graphics card is properly powered?

 

[tritous]

the contacts were fine, in fact it visually appeared perfect apart from just 3 erroneous bits out of 2gb. Worst of all, matched pairs means it's two sticks I lose, although given that my 3 working sticks (2 more coming too) are all identical make and model I really question at time about the "matched pair" thing. Still, I can't seem to buy them separately and it's not expensive.

Embarassing, but it seems my gfx is working properly again. Along with your helps with the memory and disk analysis tools I think this closes all my problems. Thanks for the support

 

[F5ing]

You're welcome! Glad to hear your battleship is all sorted out.