I recently installed a second hard disk and put the pagefile on it, taking care to leave a small pagefile on the OS disk as a receptacle for minidumps. Well, I had a BSOD -- but no minidump. And the reason appears to be that, despite the settings in Virtual Memory, there is no pagefile on the OS partition. In Explorer I can see pagefile.sys on the second disk (F) but nothing on the OS partition (C). The latest development is that I tried setting "No paging file" for C in Virtual Memory, followed by a reboot and setting 4MB. Another reboot and this time the pagefile.sys was there. Unfortunately, on the next startup it disappeared again. Can somebody help me get it back? Thanks1 person got this answerI do too

Well, I'm happy to have found someone who knows the quirks of XP so well. The people I bumped into on the web who were giving advice about moving the pagefile seemed unaware of Session Manager's aversion to a pagefile of less than 20 MB. So I was in the dark. Now thanks to you, I've got a pagefile welded to the OS partition and I can look forward eagerly to my next BSOD, secure in the knowledge that the minidump will find a home to go to. I feel like a properly armed troubleshooter. Thanks for the on-target solution and the illuminating explanation.Johnnino

You're welcome. Thanks for the follow-up.John

Here's some more follow-up for you, John John, which I hope you will find interesting. Everything's been going fine on my computer -- there's a pagefile on F and a pagefile on C -- so fine in fact that the problem I was having with a driver leading to a BSOD has not reappeared and so I can't even test the dumpfile capture. BUT...I've noticed in the Event Log that on most startups -- not all -- there is a series of warnings "Error in the peripheral \Device\Harddisk0\D during a paging operation". There can be more than 20 of these warnings. Since they all occur in the same fraction of a second, I can't imagine they're causing much delay. But is this normal? I'm assuming that Harddisk0 is the main C drive. The Utility "Pagefile Usage Monitor" shows 19 MB of pagefile usage on C (pagefile size=20MB). I thought Windows was supposed to recognize the larger pagefile on F and do its paging there. Or does it have to knock 20 times on C before it realizes there's no more room? Is this functioning as it should -- or do I need to make some adjustments? Thanks.Johnnino

Are there any other drive related errors referred to in Event Viewer? The type of error you mention can be caused by failing hardware. Hard drive connecting cable can start to give intermittent problems. I have encountered this type of problem with SATA cable, which is cheap to replace. I also bought a new hard drive recently and the firmware, which came with the drive, needed updating.You can check harddrives with HD Tune (freeware)..http://www.softpedia.com/get/System/Hard-Disk-Utils/HD-Tune.shtml Hope this helps, Gerry Cornell

Thanks Gerry,No other drive related errors in the Event Log. Both disks check out OK with Sea Tools for Windows. I'll have to keep open the possibility of changing the Sata cables. Johnnino

I assume the error you are seeing repeats regularly. Can you please post a copy of the actual report.You can access Event Viewer by selecting Start, Control Panel,Administrative Tools, and Event Viewer. When researching the meaningof the error, information regarding Event ID, Source and Descriptionare important.A tip for posting copies of Error Reports! Run Event Viewer and doubleclick on the error you want to copy. In the window, which appears is abutton resembling two pages. Click the button and close EventViewer.Now start your message (email) and do a paste into the body ofthe message. Make sure this is the first paste after exiting fromEvent Viewer.Have you checked for firmware updates for the drive?Hope this helps, Gerry Cornell

Gerry,The warning is repeated 20 times. The data appear to be slightly different each time. Here's the first one, with the data indicated as Bytes: Tipo evento: AvvisoOrigine evento: DiskCategoria evento: NessunoID evento: 51Data: 01/11/2010Ora: 21.49.31Utente: N/DComputer: GMEDIACENTERDescrizione:Errore nella periferica \Device\Harddisk0\D durante un'operazione di paging.Per ulteriori informazioni, consultare la Guida in linea e supporto tecnico all'indirizzo http://go.microsoft.com/fwlink/events.asp.Dati:0000: 03 00 68 00 01 00 b6 00 ..h...¶.0008: 00 00 00 00 33 00 04 80 ....3..0010: 2d 01 00 00 9a 00 00 c0 -.....À0018: 00 00 00 00 00 00 00 00 ........0020: 00 00 00 00 00 00 00 00 ........0028: 47 04 00 00 00 00 00 00 G.......0030: ff ff ff ff 01 00 00 00 ÿÿÿÿ....0038: 40 00 00 30 00 00 00 00 @..0....0040: 00 20 0a 12 40 03 20 40 . ..@. @0048: 00 00 02 00 14 00 00 00 ........0050: 00 00 00 00 f0 28 d4 84 ....ð(Ô0058: 00 00 00 00 c0 26 d4 84 ....À&Ô0060: 00 00 00 00 9a 00 00 c0 ......À0068: 28 00 00 51 e5 70 00 01 (..Qåp..0070: 00 00 00 00 00 00 00 00 ........0078: 00 00 00 00 00 00 00 00 ........0080: 00 00 00 00 00 00 00 00 ........0088: 00 00 00 00 00 00 00 00 ........ In case it makes a difference, here's the same event with the data indicates as Words: Tipo evento: AvvisoOrigine evento: DiskCategoria evento: NessunoID evento: 51Data: 01/11/2010Ora: 21.49.31Utente: N/DComputer: GMEDIACENTERDescrizione:Errore nella periferica \Device\Harddisk0\D durante un'operazione di paging.Per ulteriori informazioni, consultare la Guida in linea e supporto tecnico all'indirizzo http://go.microsoft.com/fwlink/events.asp.Dati:0000: 00680003 00b60001 00000000 800400330010: 0000012d c000009a 00000000 000000000020: 00000000 00000000 00000447 000000000030: ffffffff 00000001 30000040 000000000040: 120a2000 40200340 00020000 000000140050: 00000000 84d428f0 00000000 84d426c00060: 00000000 c000009a 51000028 010070e50070: 00000000 00000000 00000000 000000000080: 00000000 00000000 00000000 00000000Any clues here?I'll check for firmware updates.Thanks.Johnnino

This is a recent error from my own computer:Event Type: WarningEvent Source: DiskEvent Category: NoneEvent ID: 51Date: 03/10/2010Time: 06:23:46User: N/AComputer: GERALD-51E521F8Description:An error was detected on device \Device\Harddisk1\D during a paging operation.The solution was to upgrade the drive firmware and replace the sata cable. This was a new Seagate hard drive, where I used the cable from the previous drive. My understanding isthe connectors on the cable can be the problem. You can buy replacementsnts for 2/3 Euros. http://seagate.custkb.com/seagate/crm/selfservice/search.jsp?DocId=207931&NewLang=enI would try replacing the cable first. Hope this helps, Gerry Cornell

Firmware query for both drives draws a response of "no action required". I'll try replacing the Sata cable and see if that makes a difference. Thanks.

I've got a cable but for two days now I haven't been getting the disk warning at startup. So I'm going to hold off on replacing the old one until the error returns.

Thanks for keeping us informed.Hope this helps, Gerry Cornell

Well, the error returned -- so I replaced the Sata cable on the affected drive and on the next startup, I'm sorry to say Gerry, there they were again: 21 warnings of a disk error during a paging operation. It's beginning to look like the hardware is innocent. Should I experiment with different pagefile settings?

A 20 mb setting is smaller than one I would have suggested.Should the file be left on Drive C:?The slowest aspect of getting at a file on a hard disk is in head movement (‘seeking’). If you have only one physical drive then the file is best left where the heads are most likely to be, so where most activity is going on — on drive C:. If you have a second physical drive, it is in principle better to put the file there, because it is then less likely that the heads will have moved away from it. If, though, you have a modern large size of RAM, actual traffic on the file is likely to be low, even if programs are rolled out to it, inactive, so the point becomes an academic one. If you do put the file elsewhere, you should leave a small amount on C: — an initial size of 2MB with a Maximum of 50 is suitable — so it can be used in emergency. Without this, the system is inclined to ignore the settings and either have no page file at all (and complain) or make a very large one indeed on C:Quote from:http://aumha.org/win5/a/xpvm.htmWhat is the drive make and model?Hope this helps, Gerry Cornell

Okay, I'll raise you 30. I'll make it 50 and see what happens. I've noticed that for startup Windows pages 19-20 MB to C and 20-22 to F. Maybe it needs to page it all to one place. Dunno. As I said, I thought it was supposed to go automatically for the the larger space. I'll let you know what happens.The drive is a Maxtor 7L250S0

I set the pagefile on C to 50 MB but the error hasn't gone away. After startup I see 13-23 MB paged to C and a similar amount to F. With the passage of time it rises to 49 MB on C -- and of course goes no further. The disk error warnings occur only at startup and not again.I would like to quote from Holy Writ: "How to to configure paging files for optimization and recovery in Windows XP " http://support.microsoft.com/kb/314482The optimal solution is to create one paging file that is stored on the boot partition, and then create one paging file on another partition that is less frequently accessed on a different physical hard disk if a different physical hard disk is available. Additionally, it is optimal to create the second paging file so that it exists on its own partition, with no data or operating-system-specific files.By design, Windows uses the paging file on the less frequently accessed partition over the paging file on the more heavily accessed boot partition. An internal algorithm is used to determine which paging file to use for virtual memory management.My question is, should Windows be paging to the C partition at all? Is that inevitable at startup? And does it normallly take place without errors?

My question is, should Windows be paging to the C partition at all? Is that inevitable at startup? And does it normallly take place without errors?By default during normal operation Windows will use the pagefile on the least busy disk, when Windows is booting the Session Manager can only write to the boot volume so you need a pagefile on the boot volume to capture memory dumps. The pagefile most likely not the real culprit, paging simply means that data is brought into the RAM from the disk or that it is moved out of the RAM to the disk, it doesn't necessarily mean that the pagefile is directly involved. There are many causes for this error that often revolve around bad drivers or bad hardware.John

So we don't even know if the error was in writing or reading?

There are some other lines of enquiry in this link:http://www.eventid.net/display.asp?eventid=51&eventno=793&source=Disk&phase=1Further information here:http://support.microsoft.com/kb/244780Intel Application accelerator?http://www.intel.com/support/detect.htmHope this helps, Gerry Cornell

You have to decode the Data Words in the event message to get more information.You could always reconfigure the pagefile settings to have one large pagefile on the boot volume and see what happens, or configure it to have none and see what happens.John

Great reading list Gerry! It's like having the shop manuals for my jalopy. And although I haven't got Intel Application Accelerator I was glad to make the acquaintance of the Intel Driver Update Utility. I have the tools now to start homing in on this error. Thanks so much.

With the help of the reference works Gerry pointed me to I'll start decoding. Then I'll try the variations you suggest, John. It may be a while before you hear from me. I've got a lot of homework to do. Thanks for the guidance.

Well, it's been a long, tortuous journey over rocky and arid terrain. The toughest part, I guess, was the altitude. Much of the reading material, especially the Microsoft support stuff, was way over my technical level. I thought that in http://support.microsoft.com/kb/244780 "how to decode the data section of an Event ID 51 event message" I was going to get an easy key to the cipher. Instead I found the article itself needed to be decoded. The references to "offsets", "bytes" and "octets" set my head spinning. And helpful pointers only added to my confusion: "Note: When you interpret the hexadecimal data in the Event ID to the status code, remember that the values are represented in the little endian format." Another note was very unsettling, calling into question the reliability of the data in the Event ID: "On Windows XP and Windows Server 2003, the DeviceName may be truncated because of the size limitation of the event log entry. As a result, the displayed harddisk number, or the device object name itself may be incorrect." In fact,after much digging I discovered the the device name in my error message, "Harddisk0\D"was truncated. The "D" does not refer to partition D (in my case, a DVD drive), but turns out to be the shortened form of "DR0" -- or Drive 0. That is to say, we're still talking about the main drive, which holds the Operating System.The only other thing I was able to glean was the type of operation. So the furthest I got with the Microsoft article was that the error was a read operation on the main disk. Then I went to eventid.net. This was a lot more accessible, especially because it contains anecdotal comments from people who had the problem. There I learned that some were able to solve it by simple actions like changing the prefetch parameters or disabling write cache for the disk. I resolved to give them a try.But first I googled "Harddisk0\DR0" and that was the real breakthrough because it led me to http://www.hardwareanalysis.com/content/topic/12946/ This is a 33-page thread on Event ID 51 that has been running for more than seven years! There you can read of people buying new cables, controllers, disks -- whole computers, in a desperate attempt to get to get rid of the warning. Sometime in 2005 someone reported that he tried setting memory usage in the Advanced System properties to "Programs" instead of "System Cache" and it worked. A few days later on the thread someone wrote in to say that he tried it without success. But I tried it -- and it worked for me! Over the years a few people reported success with the maneuver -- but just as many claim to have solved the error by doing just the opposite, i.e., by switching from "Programs" to "System Cache." This endless thread seems to show that Event ID 51 is enveloped in heavy fog. I can only report that I resolved it by changing Memory Usage from "System Cache" to "Programs." I don't think I'm paying a performance price for that and I'm happy to be done with it. Good luck to all.

Congratulations.That's an excellent outcome. Well done.I am marking your reply as the answer which will give you points.I have placed the Article and will try to read iot at the weekend. It is a ____ of a long thread. Regards, Gerry Cornell

Gerry,I think what I found was merelyananswer. In two senses:It does not appear to be the answer to my original question about settings for the pagefile, which was provided by John John. It responds to the issue of Event ID 51, an error on the hard disk. And the reference in the error report to a "paging operation" is misleading. As John John pointed out, that is "Microtalk" for an I/O operation. There is everything to indicate that this is an entirely new question, having nothing to do with pagefile settings.Second, as is clear from the endless thread on Event ID 51 atwww.hardwareanalysis.com, the solution I found is just one possible solution, the one that worked for me. It has a lot to recommend it, I think. It is extremely fast and simple, reversible and without risk. So it's certainly worth a try.Also, the first lead I got in the case came from you, Gerry: the Microsoft article on ID 51, KB244780. It was complete news to me because when I clicked on the link for further information in the error window, this is what I got: ID:51Source:DiskWe're sorryThere is no additional information about this issue in the Error and Event Log Messages or Knowledge Base databases at this time. You can use the links in the Support area to determine whether any additional information might be available elsewhere.I'd like to make Microsoft aware of this but don't know how to do it. Even if the article is esoteric, there should still be a link to it. And maybe something could be done to make such an article more accessible to the non-expert by cross-linking to explanations of difficult terms and concepts.One other loose end. You brought up the issue of Intel Application Accelerator. As I told you, I don't have it. But I do have Intel Matrix Storage Manager -- and on every startup it runs a service called "Intel Matrix Storage Event Monitor". Its working name is "iaantmon.exe", which, if I'm not mistaken, is the same as Application Accelerator (that's the "aa" in the name). So the question is, is Matrix Storage manager just as much a mischief maker and possible suspect as Application Accelerator?I hasten to point out that I am not volunteering to test this hypothesis. I leave that to IT sherpas. I'm still catching my breath from my marathon search, content to have a working solution, with no yearning for further investigations.Johnnino

"the reference in the error report to a "paging operation" is misleading"My view is to the contrary. What tends to mislead is "An error was detected on device \Device\Harddisk1\D".What confuses most users is that they do not link the references to 0 or 1 to the drive label in Disk Management. The meaning of the reference to D is taken to be a drive letter which it is not. You picked up on this earlier. Event Viewer reports are a good starting point for investigation of a problem. They are clues, however, not solutions. You need to colate all clues to determine the actual problem. There can be any number of problems resulting in a particular Event ID.Knowledge Base Articles are often written after a problem has been resolved. Many solutions never result in a Knowledge Base Article. Users frequently read a Knowlege Base Article and think they have found a solution. Often they have not because they cannot see the difference(s) between their error and what is described in the Knowledge Base Article. Articles about specific Stop Error reports are notorious in this respect.Windows XP is the operating system before last. Microsoft are no longer fixing bugs or making improvements. They wiil end support (security updates) as soon as they sense they can live with the consequences. A report from you will fall on very stony ground.I think you may be mistaken regarding the Intel Application Accelerator and the Intel Matrix Storage Manager. The first is a software add-on to speed up the flow of data to and from a drive. It was always uninstallable. The Intel Matrix Storage Manager manages / controls sata / raid drives. Hope this helps, Gerry Cornell

Well, it's nice to hear Matrix Storage Manager has a clean record. I'll sleep easier.