Solved

"Delayed Write Failed : Windows was unable to save all the data for the file" upon user logout

Posted on 2004-04-21
10
23,738 Views
Last Modified: 2011-08-18

Most of the time (about 75%) when my users logout they get the following error when trying to save their profile;
"Application popup: Windows - Delayed Write Failed : Windows was unable to save all the data for the file \\vicky\UserHome\harmanc\Profile\prf41A.tmp. 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.  " (this is the error from the computer system)

and sometimes there is the error that "the semaphore timeout period expired"

This is causing me no end of grief, and I can't seem to isolate it as a network problem.

Cheers

Tom H
0
Comment
Question by:tommyh
[X]
Welcome to Experts Exchange

Add your voice to the tech community where 5M+ people just like you are talking about what matters.

  • Help others & share knowledge
  • Earn cash & points
  • Learn & ask questions
10 Comments
 
LVL 7

Accepted Solution

by:
IceRaven earned 500 total points
ID: 10877070
Hi tommyh,

Does your server have the latest service pack?
http://support.microsoft.com/default.aspx?scid=kb;EN-US;321733
http://support.microsoft.com/?kbid=327194



Cheers,
IceRaven
0
 

Author Comment

by:tommyh
ID: 10877572

I located the key specified in those KB articles, but it is the hotfix value already.

Also both machines are automatically udpdated at 9am, so will be current on service packs and critical updates as of today,
0
 

Expert Comment

by:acguess123
ID: 10880346
Tommyh,

Is your server using RAID? The reason I ask is that about a month ago I had a customer that had almostm the same problem. It wasn't on the user profiles, but it was giving the same "Delayed Write Failed" error in the event viewer. I ended up having to recreayed the RAID array and restore from backup the partition that I was having trouble with.


Thanks,

Adam
0
Best Practices: Disaster Recovery Testing

Besides backup, any IT division should have a disaster recovery plan. You will find a few tips below relating to the development of such a plan and to what issues one should pay special attention in the course of backup planning.

 

Author Comment

by:tommyh
ID: 10882289

Yes. the server does use raid 5 for that volume. It would be an absolute nightmare to try to restore the data as it also contains all the user exchange mailboxes and about 50GB of assorted other files. (In this case the cure would be very much worse than the disease)

I tried moving some of the users home directories to another share, which is on a raid 1+2 (mirrored and striped), but that made no odds. I also deleted a few of the profiles and recreated them empty, which for a while stopped the error occuring, but as the profiles started getting bigger again as the data was put back the error returned with a vengance.

I'm thinking that it is a microsoft error of the type suggested by that KB article, but something that has not been picked up yet.
0
 
LVL 18

Expert Comment

by:crissand
ID: 10891549
It looks like a Raid failure. Boot into raid bios and see if one of the disks is down, and if so, replace the disk.
0
 

Author Comment

by:tommyh
ID: 10943792
After much trial and error I have isolated the problem to the
"HKEY_LOCAL_MACHINE\System\CurrentControlSet\Services\Lanmanworkstation\Parameters\enablesecuritysignature" entry.

According to microsoft the problem was fixed with windows 2000 server SP4, however my W2K server and all my XP pro workstations have current updates and still had this problem.

I wrote a batch script that sets lanmanworkstation\enablesecuritysignature to value 0 using the reg tool, and run it as a startup script under GPO computer configuration. This eliminated the problem from all the workstations. And as a final confirmation of this, I had cause to reinstall an XP laptop from scratch, this machine immediately had the problem I described above upon logout, but as soon as I rebooted it, and it picked up the GPO and the startup script the problem disappeared.

There is a discussion of the problem here;
http://www.tangent-systems.com/support/delayedwrite.html

Which actually provides a fairly detailed discussion of the issue, and the correct fix. Though I didn't find this article until after much searching and testing, and having already found it myself. But that's just my luck. ;->







0
 
LVL 7

Expert Comment

by:IceRaven
ID: 10948586
Glad your systems are now working!

Cheers,
IceRaven.
0
 

Author Comment

by:tommyh
ID: 10953659

interestingly, I have been monitoring the event logs for the specific error, which is Event ID 50, Source MRxSMB, and it is still occuring on occasion. ie it happened twice today from about 50 logouts, rather than about 100 times a day (each logout generated more than 1 of these errors in the event log)

so its not totally fixed, but these I can possibly put down to more general networking problems. its a shame as I thought I had cracked it.
0
 

Expert Comment

by:dandorricott
ID: 13935991
IF you go to "My Computer" right click "C" drive - "Properties" "Hardware" select Local HDD and "Properties" select "Policies" and remove the tick from "Enable Write caching on the disk"....
I have done this on a few machines and it has not returned. I haven't seen the problem since. I have not had this confirmed 100% so please leave more feedback
0
 

Author Comment

by:tommyh
ID: 13936140

Some time later after this issue I discovered that there was a duplex setting inconsistency between 2 switches, in the intervening network, which cleared up that problem and some others as well.

I no longer work on those systems, so this one is defnintely closed.

However I would point out that if you are experiencing this problem, then I would make 110% sure that the network is fine. Look for dropped packets and errors on switches. try and find something like that before messing about with the windows networking settings.

There is a tool for testing local network throughput, which I can't remember offhand, if you are not getting something like 80% of the expected max throughput for the network, then I would look to solve that before doing the windows hacks stuff
0

Featured Post

Simplifying Server Workload Migrations

This use case outlines the migration challenges that organizations face and how the Acronis AnyData Engine supports physical-to-physical (P2P), physical-to-virtual (P2V), virtual to physical (V2P), and cross-virtual (V2V) migration scenarios to address these challenges.

Question has a verified solution.

If you are experiencing a similar issue, please ask a related question

NTFS file system has been developed by Microsoft that is widely used by Windows NT operating system and its advanced versions. It is the mostly used over FAT file system as it provides superior features like reliability, security, storage, efficienc…
The goal of this blog is: - To define the incident management process - To go over the key elements of an incident management system - To look into incident alert management tools that integrate with ConnectWise.
Are you ready to implement Active Directory best practices without reading 300+ pages? You're in luck. In this webinar hosted by Skyport Systems, you gain insight into Microsoft's latest comprehensive guide, with tips on the best and easiest way…

739 members asked questions and received personalized solutions in the past 7 days.

Join the community of 500,000 technology professionals and ask your questions.

Join & Ask a Question