Please register or login. There are 0 registered and 1066 anonymous users currently online. Current bandwidth usage: 326.30 kbit/s May 18 - 06:31pm EDT 
Hardware Analysis
      
Forums Product Prices
  Contents 
 
 

  Latest Topics 
 

More >>
 

    
 
 

  You Are Here: 
 
/ Forums / Other Hardware /
 

  "delayed Write failure" in WIN XP 
 
 Author 
 Date Written 
 Tools 
Continue Reading on Page: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, Next >>
Vadim Rapp Jan 30, 2004, 01:01pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
ATI mentioned this problem in their installation notes for the recent several releases of the drivers. Here's their text:

============

Data corruption may occur if the Large System Cache feature is enabled in Windows XP.

For complete details see; http://www.ati.com/support/infobase/4217.html

This symptom does not occur on all systems. The key ingredients that lead to this symptom may include:

- System Memory greater than 512 Meg. (1 gigabyte of RAM is common)
- Large NTFS disk volumes. And multiple large volumes. (60-100 gigabyte hard drives possibly in RAID arrays)
- AGP graphics with large AGP resource requirements (AGP aperture greater than default)
- Large file transfers

To avoid this symptom ensure the System Cache option is NOT selected for Memory usage. Use the Programs option.

- right-click on the My Computer Icon then Properties to open the System Properties dialog box
- select the Advanced Tab then the Settings button under Performance
- select the Advanced tab on the Performance options tab
- in the Memory Usage area, ensure the Programs radio button is selected
- Apply the changes and select OK

Want to enjoy less advertisements and more features? Click here to become a Hardware Analysis registered user.
Everett Anderson Feb 01, 2004, 03:09pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
It seems odd that ATI would take an issue that isn't ATI-specific. I updated my closed Microsoft problem report with the link, though. woo. I'm afraid they'll think it's an ATI problem, though.

- Everett

Corvus Raven Feb 01, 2004, 10:27pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
You mentioned something that is not M$.. so... :)

.." AH.. well that is not OUR problem.." :) hehe

--------------------------
ASUS A7N8X Deluxe PCB 2.00 w/ BIOS 1005 (cause? Dunno.. ATM)
AMD XP 2800+ @ 0.0 GHz 0MHz FSB (overheating for no appearent reason)
(2) - Corsair XMS512-3500C2 (5-2-2-2T) @ 0MHz
ATI Radeon 9700 Pro @0
Enermax 500W p/s (ok.
Jean-Paul Satyre Feb 11, 2004, 04:52pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
Well, on Saturday the Delayed Write issue finally got me too.
In the mean time it has destroyed 70 gig of data, much of which I wont be able to recover.

Up until Saturday I was running:

WinXP
AMD 2200
ATI 9700
2xMaxtor 60 gig (Raid 0 on Abit KX7-333)

Everything was fine. However my HDD was full.

So I lashed out and bought a shiney new Seagate 120 Gig drive. With that sexy 8 Mb cache....

Well. I installed that and it broke my computer. Or more precisely: It ruined windows.
I get the Delayed Write error on booting. Can save the data for the $MFT... goodbye windows.
Now, overwriting the drive partition with an image didn't help. Rearranging the partitions didn't help. Reinstalling windows didn't help. Disabling the new Seagate HDD didnt help.
The Seagate drive is fine.
My HDDs seem to be fine.

Currently I have just installed windows on the new IDE drive and it's working ok. Have not changed the memory settings to system cache this time. I have disabled the drive cachine for the Seagate IDE drive. I have installed a patch to address the caching issue anyway. There is a fix out there for the HDD precaching issue.
I am going to recover as much data from my HDD as possible and then destroy my raid array, maybe do a low-level format, rebuild the raid array and see if i can convince winXP to run from my C: drive on raid again.

At the moment it dies every time and the poor partitions are in a mess.

So I am going to start from the very bottom and work upwards once I have as much of my data back as I can recover.

Corvus Raven Feb 11, 2004, 05:33pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
Most commont solution that I have seen was just the power supplu to the video card. (delayed loop)

I am not certain (and doubt it anyway) that it is the same problem or same fix, for all cases.

--------------------------
ASUS A7N8X Deluxe PCB 2.00 w/ BIOS 1005 (cause? Dunno.. ATM)
AMD XP 2800+ @ 0.0 GHz 0MHz FSB (overheating for no appearent reason)
(2) - Corsair XMS512-3500C2 (5-2-2-2T) @ 0MHz
ATI Radeon 9700 Pro @0
Enermax 500W p/s (ok.
Brian Murphy Feb 12, 2004, 04:23pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
This has got to be an MS problem.

Google search shows this happens with a number of drive vendors and system vendors. I've tried four drives (120GB Western Digital firewire/USB2, two different 120GB Iomega firewire/USB2, 80GB Iomega USB2) on two separate systems (gateway PII-133MHz USB2, Dell PIII-1.2GHz firewire) and get the problem. WD has given up on trying to solve the problem, so that drive goes back to the store. Iomega has advanced the problem to their 'advanced research group'. MS support is still going through their KB recommendations (which I have already tried).

Have tried almost all voodo recommendations
- looked for specific firmware updates (none available for these systems/drives)
- looked for specific USB/firewire driver updates from controller mfgrs (none available).
- change drive policies (hardware manager) to both 'optimized for performance' and 'optimized for quick removal'
- eliminate any IRQ sharing with device connected to drive
- change PCI latency to give firewire more PCI bus time (from 32 to 248)
- go through steps of changing drive letters, but don't actually changed them (procedure suggested by BOTH WD and Iomega)
- change XP system / memory preformance to be optimized for "programs"
- change page file max size to 1.5x recommended size
- change page file to 'managed by system'
- change HKEY_LOCAL_MACHINE\SYSTEM\CurrentControllSet\Control\Session Manager\Memory Management\System Pages to 0xffffffff
- tried Fat32 vs NTFS
- disable all startup programs with msconfig
- tested to see if 'large chunk' IO transfers caused the problem (thanks BusTrace for the test app); they aren't
- happens with generic file cut/paste (Iomega 'doesn't support' MS backup, which was one way to trigger the problem)

Some things of note:
- doesn't occur on any of the drives when I/O crawls along via a USB 1 connection
- only seems to occur during intensive file WRITE (not read), e.g., copying a 1-30GB file or directories/contents of that aggregate size.
- does not seem to occur during 'routine' WRITE scenarios (100s of files, < 1GB in size)
- Accompanied variously by system event IDs
[9] sbp2port "The device, , did not respond within the timeout period."
[14] Volsnap: "The shadow copy of volume C: was aborted because of an IO failure."
[26] Application Popup: "Application popup: Windows - Delayed Write Failed : Windows was unable to save all the data for the file [drive:filename; filename is usually the name of the file being transfered or else something like: D:\$Mft or D:\$BitMap]. The data has been lost. This error may be caused by a failure of your computer hardware or network connection. Please try to save this file elsewhere.
[34] disk: "The driver disabled the write cache on device \Device\Harddisk2\DR6.
[50] ntfs: "{Delayed Write Failed} Windows was unable to save all the data for the file . The data has been lost. This error may be caused by a failure of your computer hardware or network connection. Please try to save this file elsewhere."
[51] disk: "An error was detected on device \Device\Harddisk2\D during a paging operation."
[57] ftdisk: "The system failed to flush data to the transaction log. Corruption may occur."

So, lets pick a disk and see what drivers are in common when it is placed on two different systems where the failure occurs (info Using device manager "devices by connection" to walk up the device tree from the failing drive):

On the Dell Inspirion 8100
*Disk:
Driver:
disk.sys 5.1.2600.1106 (xpsp1.020828-1920)
PartMgr.sys (5.1.2600.0) (XPClient 010817-1148)
*SBP2 Compliant IEEE 1394 device:
Driver:
sbp2port.sys 5.1.2600.1106 (xpsp1.020828-1920)
*TI OHCI Compliant IEEE 1394 Host Controller
Driver: MS 7/1/2001 v5.1.2535.0
1394bus.sys 5.1.2600.1106 (xpsp1.020828-1920)
arp1394.sys 5.1.2600.1106 (xpsp1.020820-1920)
enum1394.sys 5.1.2600.0 (xpclient 010817-1148)
nic1349.sys 5.1.2600.1106 (xpsp1.020828-1920)
ohci1394.sys 5.1.2600.1106 (xpsp1 020828-1920)
*Intel(R) 82801BAM/CAM PCI Bridge - 2448
Driver: MS 7/1/2001 v5.1.2600.0
pci.sys 5.1.2600 (xpsp1.020828-1920)
*PCI bus
Driver: MS 7/1/2001 v5.1.2600.0
pci.sys 5.1.2600.1106 (xpsp1.020828-1920)
*Microsoft ACPI-Compliant System
Driver: MS 7/1/2001 v5.1.2535.0
acpi.sys 5.1.2600.1106 (xpsp1.020828-1920)
*Advanced Configuration and Power Interface (ACPI) PC
Driver: MS 7/1/2001 v5.1.2600.0
hal.dll 5.1.2600.1106 (xpsp1.020828-1920)
ntkrnlpa.exe 5.1.2600.1151 (xpsp2.030422-1633)
ntoskrnl.exe 5.1.2600.1151 (xpsp2.030422-1633)

On the Gateway
*Disk:
Driver: MS 7/1/2001 v5.1.2535.0
disk.sys 5.1.2600.1106 (xpsp1.020828-1920)
PartMgr.sys 5.1.2600.0 (XPClient.010817-1148)
*USB Mass Storage Device
Driver MS 7/1/2001 v5.1.2600.0
UBSTOR.SYS 5.1.2600.1106 (xpsp1.020828-1920)
*ALi PCI to USB Enhanced Host Controller
Driver: ALi Corporation 9/9/2003 v5.1.2600.116
AliEhci.sys 5.1.2600.0116
AliGP.sys 5.1.2600.0116
AliHub.sys 5.1.2600.0116
AliRtHub.sys 5.1.2600.0116
*PCI Bus
Driver: MS 7/1/2001 v5.1.2600
pci.sys 5.1.2600.1106 (xpsp1.020828-1920)
*Standard PC
Driver: MS 7/1/2001 v5.1.2600.0
hal.dll 5.1.2600.1106 (xpsp1.020828-1920)
ntkrnlpa.exe 5.1.2600.1151 (xpsp2.030422-1633)
ntoskrnl.exe 5.1.2600.1151 (xpsp2.030422-1633)

Is it the drive?
No - happens on different drives from both Iomega and Western Digital. Doesn't happen when drives are connected to a MAC (testing is going to extreemes, eh?)

Is it the drive cable?
Not likely unless the drives don't work with the cables they're shipped with AND Belkin generic (firewire) cabling

Is it the system hardware?
Not likely given the large hardware differences in the two systems used for testing.

Is it the MS OS / drivers? Well, what is in common across all testing strategies that produce a failure?

1) intense file writing at USB 2 or Firewire (400Mb) speeds.
2) Microsoft's disk.sys and partmgr.sys at the low end, and the general XP stuff at and above PCI bus at the higher level.

Brian Murphy Feb 12, 2004, 04:29pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
Correction to the above. The Gateway PII is a 333Mhz machine (so it meets disk mfgr system requirements).

Jean-Paul Satyre Feb 12, 2004, 05:06pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I have had a very embarassing and severely belated revelation cum-flash-of-brilliance.

My delayed write failure upon booting most often came with the following pretty message:

Error 0xc0000147

This is the error code for no page file.

This has been mercilessly gnawing away at my subconscious for days now.
Suddenly, after spending the last 24 hours recovering data and transferring the entire 120 Gig to my new HDD, it hit me.

Windows Recovery console. Command: Map.

Remember what the booting command for windows looks like Multi (0) etc...

It tells windows to boot from drive 0 volume 0 partition 1
The IDE drive was identified as drive 0.
It has nothing to do with the drive letters you see in windows per se.
No.. windows was vainly attempting to boot from drive 0 volume 0 partition 1
When I should have simply changed the boot command to drive 1 volume 0 partition 1

Once I have the second copy of xp installed on my newly formatted raid array I am going to back everything up with drive image and test my theory.

I bet I can turn drive caching back on (after applying the patch) for the IDE HDD, making it run faster.

However I am oging to leave the memory settings alone for now, just to be on the safe side.

Corvus Raven Feb 14, 2004, 11:25am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
It would figure if it were just some problem with M$ and they have no idea how to fix.

Another common denominator is the 2.0
Did you say it worked with the 1.0/1.1 USB?

The few time I had that problem it was 'fixed' with re-instal. (My HDs are IDE though)

--------------------------
ASUS A7N8X Deluxe PCB 2.00 w/ BIOS 1005 (cause? Dunno.. ATM)
AMD XP 2800+ @ 0.0 GHz 0MHz FSB (overheating for no appearent reason)
(2) - Corsair XMS512-3500C2 (5-2-2-2T) @ 0MHz
ATI Radeon 9700 Pro @0
Enermax 500W p/s (ok.
Brian Murphy Feb 18, 2004, 10:37am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
Correct, the problem does not happen under USB 1.x (on either machine). Drives also work fine under "light" I/O on USB 2 & firware. Failure only occurs under heavy I/O (seems to be writes but I can't be positive) on USB 2 and firewire under XP (heavy I/O on MAC with these drives hums on with no problems). The PII system is a bit old, but the Dell system as well as the drives I'm testing are "Certified for windows XP" and listed in the MS windows catalog as such (which has shut down MS support tech from saying it is a hardware vendor problem alone - first place they went, naturally). Both systems have recently been re-installed clean from the ground up, using all of the latest vendor bios/firmware/driver updates. Note that on the USB 2 system, I spoke with the vendor and they said I could remove their drivers (from ALi, noted above) and use the native XP drivers. I did this. Problem still occurs, so now it seems pretty clear that it is an XP driver problem. I"m slowly getting the MS support tech to that realization in a current ticket I have open with them. His recent suggestion is "corrupt drivers?". I responded with "they're MS drivers - how do I determine if they're corrupt?". Once we verify they aren't corrupt, that leaves us with flawed MS drivers PERIOD ;-).

Oh yeah, add to the list of things tried changing the cables - that didn't fix the problem either.

Corvus Raven Feb 19, 2004, 11:26am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I have USB 2.0 as well.. so ..I was kina guessing that it could also be a party to the problem.

Surprise.. Something new is causeing a new problem :)

I haven't had the problem since. (delayed write..) but I do seem to have some other problem lately. UT2003/uted drops out for no appearent reason. The errors seem to indicate lost data (memory leak?) or something is otherwise misplaced that is expected to be there. At first I thought it might be heat related, but it has happened within a few minutes of powerup or an hour later.. XP isn't saying too much about it. other then asking if I would like to send the report off.. Umm no. :)

"...[50] ntfs: "{Delayed Write Failed} Windows was unable to save all the data for the file . The data has been lost. This error may be caused by a failure of your computer hardware or network connection. Please try to save this file elsewhere."..."

Like how this line says that the data is lost. but at the end it asks for you to save it. :)
No wounder windoze users get confused ;)

--------------------------
ASUS A7N8X Deluxe PCB 2.00 w/ BIOS 1005 (cause? Dunno.. ATM)
AMD XP 2800+ @ 0.0 GHz 0MHz FSB (overheating for no appearent reason)
(2) - Corsair XMS512-3500C2 (5-2-2-2T) @ 0MHz
ATI Radeon 9700 Pro @0
Enermax 500W p/s (ok.
Gabriel Klein Feb 20, 2004, 04:26am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
>> This is the reason of the delayed write failure on my Maxtor (SATA 160) drive.

When you go on some corrupted sectors, the hard disk seams to go into corrupted mode.
(Failed mode).

This is the reason why when you load Windows, everything seams OK but suddenly the system seams to "freeze".
The solution to recover your datas is not to go on these sectors. (Why don't all these recovers utilities have a "skip" some sector feature or auto-reboot feature.

What I have done is first to delete track 0-63 using maxtor utility. So you can boot the hard disk and it doesn't goes in failed mode immediately. (Don't do it if it doesn't goes in failed mode!!)
(A better solution would be to disabled the partitions with a DOS soft.)
Then find the bad sectors (Each time you find a bad sector... reboot!)
Then I have copied all "good" sectors (99.99% of the disk...) to another disk.
Then I have used an utility to recover my partitions... And CheckDisk will do the rest.

The Solution:
> Send the hard disk back to your reseller or Maxtor as this is a physical error. I have tryed a low-level format after my first crash, but another crash happend again two months after.
> If you are not under garantie, try to low-level the hard disk!
> Don't put any importants data on the start of the disk! (Use C:\ only to store program files.)

Don't hesitate: gab**riel*.kle*in**@*ep*fl.c*h



Best regards

Michael Yeang Feb 25, 2004, 06:18am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I'm a newbie to this forum, but I've got the same Delayed Write Failure experience using USB 2.0. Here's the URL to another forum where you can read about these discussions and hopefully get some useful info there.

http://forums.viaarena.com/messageview.cfm?catid=16&thread...TARTPAGE=1

I hope you can access it, otherwise I shall have to repeat the postings here.

Michael Yeang Feb 25, 2004, 08:56am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I'm a newbie to this forum, but I've got the same Delayed Write Failure experience using USB 2.0. Here's the URL to another forum where you can read about these discussions and hopefully get some useful info there.

http://forums.viaarena.com/messageview.cfm?catid=16&thread...TARTPAGE=1

I hope you can access it, otherwise I shall have to repeat the postings here.

Incidentally, I think I have solved the problem by doing these two things:
1. I changed HKEY_LOCAL_MACHINE\SYSTEM\CurrentControllSet\Control\Session Manager\Memory Management\System Pages to 0xffffffff
2. In System Properties - Advanced - Performance Options - Visual Effects, I selected the option "Let Windows choose what's best for my computer" instead of "Adjust for best performance"

I'm not sure which one did the trick, but I think it worked, after one successful, error-free defrag test on the same USB hard disk with 68GB of files. Never had success before.

Michael Yeang Feb 25, 2004, 09:15am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I'm a newbie to this forum, but I've got the same Delayed Write Failure experience using USB 2.0. Here's the URL to another forum where you can read about these discussions and hopefully get some useful info there.

http://forums.viaarena.com/messageview.cfm?catid=16&thread...TARTPAGE=1

I hope you can access it, otherwise I shall have to repeat the postings here.

Incidentally, I think I have solved the problem by doing these two things:
1. I changed HKEY_LOCAL_MACHINE\SYSTEM\CurrentControllSet\Control\Session Manager\Memory Management\System Pages to 0xffffffff
2. In System Properties - Advanced - Performance Options - Visual Effects, I selected the option "Let Windows choose what's best for my computer" instead of "Adjust for best performance"

I'm not sure which one did the trick, but I think it worked, after one successful, error-free defrag test on the same USB hard disk with 68GB of files. Never had success before.

Corvus Raven Feb 25, 2004, 02:38pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
Gabriel Klein:
That sounds like the error code was genorated properly. :) If the HD tries to save the data and comes accross a bad sector it may try to write and endup timing out for the operation.

Gabriel & Michael:
After looking at the other forum it would appeare that Maxtor drives may be part of the problem. Alot had same manufacture.


There are times I am glad I don't have a particular problem. And there are othertimes I weouldn't mind.. Can be entertaining.

--------------------------
ASUS A7N8X Deluxe PCB 2.00 w/ BIOS 1005 (cause? Dunno.. ATM)
AMD XP 2800+ @ 0.0 GHz 0MHz FSB (overheating for no appearent reason)
(2) - Corsair XMS512-3500C2 (5-2-2-2T) @ 0MHz
ATI Radeon 9700 Pro @0
Enermax 500W p/s (ok.
Michael Yeang Feb 25, 2004, 07:18pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I got the "0xffffffff" tip from Ed R on page 2 of this discussion thread. Thanks, Ed!

This kind of entertainment, I don't need. :) I'd much rather sit in front of my PC TV munching nuts!

These forums are really valuable in helping each other brainstorm and come up with the best solutions, and even if we've not encountered such problems, we can learn from others' mistakes / misadventures to take steps to avoid them.

By the way, my earlier duplicate posts happened because the first time I got a "page not found" error, so I thought it was not posted, and I clicked on Post a second time.

Corvus Raven Feb 25, 2004, 10:29pm EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I just thought it was my glassess :) (One for each eye. ;) )


___________________________
It's funny before my machine crashed, I had no problems. :)

--------------------------
ASUS A7N8X Deluxe PCB 2.00 w/ BIOS 1005 (cause? Dunno.. ATM)
AMD XP 2800+ @ 0.0 GHz 0MHz FSB (overheating for no appearent reason)
(2) - Corsair XMS512-3500C2 (5-2-2-2T) @ 0MHz
ATI Radeon 9700 Pro @0
Enermax 500W p/s (ok.
Michael Yeang Feb 26, 2004, 04:57am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
I think my elation was a bit premature. Today, I tried to defrag and delayed write failed again :(. So in conclusion, I'm doing this:

1. Synchronize files in USB 2.0.
2. Defrag, check disk or compress using a USB 1.1 hub between the external drive and the PC. This works fine, except for the speed. I don't know whether it's worth investing on a USB 2.0 hub or a cable that's certified for hi-speed.

BTW, is there a way to edit a message here without posting a new one?

porkchop Feb 26, 2004, 07:17am EST Report Abuse
>> Re: "delayed Write failure" in WIN XP
at the top of your message where your name is and the date
on the right hand side there should be an option to edit post
this appears on your messages only
and i believe that after a message has been edited
the edit date will show up at bottom of your message

see what i mean

jp


asus a7n8x-e
2500 barton(11x200)
ati 9800 pro
1gb hynix pc3200


Continue Reading on Page: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, Next >>

 

    
 
 

  Topic Tools 
 
RSS UpdatesRSS Updates
 

  Related Articles 
 
 

  Newsletter 
 
A weekly newsletter featuring an editorial and a roundup of the latest articles, news and other interesting topics.

Please enter your email address below and click Subscribe.