LPB SATA/hang problems???

[RGone]

:confused: back to the beginning and after a complete re-read of your post will point out the topic that raises a red flag with me.

have I been unlucky and picked the problematic HDD for my o/s?

Maybe! I see many causing trouble.

then system eventually refused to boot with corrupted system32/config/system file.

Certainly sounds like memory related!

Normally run RAM at 11,2,2,2 but have tried much slacker and lower FSB With 3.1v usually

At least 3.2V with 3 sticks of memory and 11, 3, 2, 2 & CAS2.0 or Cas2.5 might be much better with the 3 memory slot load.

Oh yeah, running CAT4.9 and Nforce3.13

Don't know about the CAT's but the 3.13's probably need to be the 4.24's.

 

disk - an error was detected on device Deviceharddisk0D during a pagin operation.

 

chkdsk still shows no bad sectors.

Are both these drives faulty or is the system unstable (prime runs fine tho)

Disk error at during paging certainly sounds like memory errors and not scandisk would find "nothing".

 

Update - memory and 11,3,3,2 and CPC off seems to help. Running 11x200 and appears better than 166x11 (touch wood)

I imagine it does do better and is why I recommended it above. You have to remember I am DOING my post from the very front of the thread.

Weird tho have been running at 200x12 & 11,2,2,2 CPC on for months before this.

Possible failing stick or climate conditions changed and hotter now or the north and south bridge are collecting dust and running hotter. Lot of "what ifs" here.

 

 

The SW_IDE drivers are not recommended, uninstall and switch to the MS default drivers.

 

Have you actually ran the DIAG utility for your drives (made by the drive manufacturer)?

 

The last drives to do this to me were my Fujitsu's... as you can see they are no longer in my config. Apparently they had bad servo-motors. Check your Event Log (I think System log), and look for entries like "HD not responding within timeout period" or the likes.

Now that was a good question about your setup and how some drives may act up. AND you answer with this>

Don't think Samsung do diag for SATA.....wonder why!

Which would be enough reason for them to be "not" in *my* personal case.

 

Damn.....is there no cure!!!!!

In deference to youth and ladies> Deeyam there might be. GO TO overclocking section and find out which HellFire bioses have 3114 SATA raid 50.39 or above in his bios and try it!!! Bios savior is a must in my mind to flash BETA bios. Then run the 1.0.0.7 SiL Drivers or higher.

 

stay away from the NV IDE drivers.

 

are you using SP1a or SP2?

Stick with SP1a for sure until it is sorted as SP2 is iffy. Now do search by user name and find the Microsoft KB "knowledge base" articles and 'fixes' for failure to flush cache before shutdown and read what he says about the useage of the KB articles.

Will try the memory sticks thang soon. Cheers. I'll be vexed if it cures it tho U need 1Gb RAM these days for games. I guess it will mean new ram and u can't buy another 512 winbond stick the same ne more. mmmm

Yep but if new ram fixed it > oh what a relief and if lowering to two sticks pointed the "thing" to do > Oh really what a relief.

Still looking for some guru to come and shed the light

We all wish somebody would come by and wave a magic wand and unburden us from our labors and plight. In this case I have pointed out enough red flags that I have seen to allow you to become your very own guru !

 

Sincerely, RGone...

[Poisonsnak]

Hmm try downloading the samsung regular diagnostic anyway. I downloaded Seagate's onto a floppy and it required no drivers or prodding in any way and it detected the SiI controller and my two drives on it.

 

Also RGone's RAM ideas are good I have heard of HD corruption due to RAM issues quite often. Maybe before you reinstall windows try memtest86 for a few (more like 12 hehe) hours with all 3 sticks and see if you make it angry. If the board is having trouble driving that much RAM you could back off to the one stick of 512 and see where that gets you.

[**FIZZY**]

Update time again

 

The SATA drives aren't listed as supported in any of Samsung's diag proggies.

 

Now running 11x200 (1.75v)

and now the important bit...

RAM at 11,2,2,2 CPC on at........3.3v :nod:

 

I've done a couple of loops of memtest (2hrs ish)

Prime for a couple fo hours...and all seems cool.

 

Touch wood, no more lock ups :cool:

 

Seems like this RAM wants loads of V up its @ss. Weird tho how slightly less volts and much slacker timings/FSB isn't the cure!

 

Thought I would erradicate RAM during trouble shooting by backing off - seems like not the case - just needed a power boost.

 

I'll keep you posted if things start to go pears.

 

many many thanks all :shake:

 

edit/ changed sig to suit /

[uwackme]

INPUT:

 

That powersupply is NOT gonna cut it going any higher. Cranking an Athlon along with 1G ddr, 2x80G drives and a 9800pro..... combined with pushing the OC envelope... you WILL need a bigger supply.

 

Fortron530W, Superflower550W, OCZ520W,Antec510W there is a short list of stuff that JUST PLAIN WORKS. Think of it as the basement, you cant build a good house on a crappy basement.

 

200x11 (2.2Ghz) and near stock clocks on 9800pro will probably be fine. just keep it in mind for the future.

 

One nice boost can come from using a single IDE drive ALSO and placing the XP pagefile on that instead of on the raid array. This allows simultaneous accessing and is one of the "perks" of the NForce2 system.

[**FIZZY**]

Cheers for the headsup on the PSU uwackme

 

It's something that I've been watching for a while now. My Enermax was a good investment (about 4years ago). I keep an eye on the rails (particularly during load) and have checked with a multimeter. They appear quite stable. I will change it eventually.

 

The only spare IDE drive I have is ATA66 5400rpm 10Gb - will this give a suitable boost? or be too slow? Better than nothing I guess :nod:

 

I went out yesterday and left Memtest running - did 6hrs no errors = YAY

 

Once again cheers gurus :angel:

[**FIZZY**]

Just thought - I've not got RAID anymore after my grief encountered.

 

I got 2 separate 80Gb SATA drives now - if i put page file on non-o/s drive is that as good as using IDE? or is the fact that both using the same controller the bottleneck?

 

Cheers

[deejayz]

I sorta agree with the PSU opinon, sounds like a PSU / Cable Connctors related issuse. I had this exact prob when I first setup raid and made a post sometime ago about it. Since putting in a better psu with tighter molex / sata power conections and checking to see if the sata data leads are in before a boot after maintanace seems to av solved the problems.

 

I could force it to messup with a little nudge of the wires, the noise I would get is a click then same one these drives make when powering up, so i thought is was a power problem. In my case this turned out to be true...

 

A Few Tips:

 

Keep you drives on there own power lead if pos and dont split power between the Graphics card and HardDrives.

 

After new bios install delete and recreate raid array, my 120 Gb drives show as 114 Gb and 111Gb depending on Sata Bios used, this may also lead to some unforseen problems.

 

Get a Higer Rated PSU that one aint cutting the mustard in its curren configuration.

[twink]

Just a note that I thought might be handy, I had similar problems when running a 400 Watt PSU on a LP B board, with a 2600+, Radeon 9800 and 4x seagate 120GB PATA hdds. Swapping the video card over to the other floppy power adapter seems to either fixed or reduced the problem.

[**FIZZY**]

I always try to keep load on power leads even.

GFX on 1 lead, HDD's on another,etc

 

3.3v through the RAM is still very stable :nod:

 

Can't see why lower fsb and slacker timings at 3.1v was being problematic....but hey it's sorted now.

 

Cheers and all that.

 

Any view on the page file issue re: using the other SATA drive?