May 12th, 2020

When you start getting in-page errors on your hard drive, it’s time to go shopping for a new hard drive, redux

Raymond Chen

Some time ago, I explained that receiving a STATUS_IN_PAGE_ERROR from your hard drive means that it’s time to go shopping for a new hard drive. It means that the system needed to page in code or data from a storage device, but the storage device was unable to produce the data. This can happen if you are running a program from removable media or over the network, but if it happens for your hard drive, that’s bad news.

Your hard drive is starting to die. Back up everything you can and replace it as soon as possible.

Now, the STATUS_IN_PAGE_ERROR code is used when the system could not page data in for a user-mode application, and the kernel is forced to terminate the program. But the I/O failure could also occur if the system could not page data in for the kernel itself, and that’s a much more dire situation.

There is no process to terminate when the kernel itself suffers an in-page error. The only thing it can do is terminate itself, and that gives you the infamous blue screen of death.

There are two different stop codes for in-page errors, depending on whether the thing that couldn’t be paged in was static data or stack data.

Stop number	Stop code
`0x77`	`KERNEL_DATA_INPAGE_ERROR`
`0x7A`	`KERNEL_STACK_INPAGE_ERROR`

The difference between them is not really important. They both mean, “Back up all your data and get a new hard drive ASAP.”

Author

Raymond Chen

Raymond has been involved in the evolution of Windows for more than 30 years. In 2003, he began a Web site known as The Old New Thing which has grown in popularity far beyond his wildest imagination, a development which still gives him the heebie-jeebies. The Web site spawned a book, coincidentally also titled The Old New Thing (Addison Wesley 2007). He occasionally appears on the Windows Dev Docs Twitter account to tell stories which convey no useful information.

4 comments

Discussion is closed. Login to edit/delete existing comments.

Sort by :

Newest

Newest Popular Oldest

qbprog May 12, 2020 0

Couldn’t be a sort of bug in the drivers or windows itself that may cause that error? 🤔
Martin Raiber May 12, 2020 0

On the other hand if it happens only once it could be "normal". Desktop hard disks are usually rated for an URE rate of one read error every 10e14 bits and Desktop SSDs aren't really designed for anything better. Of couse having a "normal" failure in the relatively small area that is used for the page cache is either really bad luck or indicates that the whole drive is failing. So you a probably right...
Read more
On the other hand if it happens only once it could be “normal”. Desktop hard disks are usually rated for an URE rate of one read error every 10e14 bits and Desktop SSDs aren’t really designed for anything better. Of couse having a “normal” failure in the relatively small area that is used for the page cache is either really bad luck or indicates that the whole drive is failing. So you a probably right in the majority of cases.

Read less
chazste@yahoo.com May 12, 2020 0

There is another possible cause for this. Different but related. On laptop mechanical hard drives there is a shock detection device. It is possible to have that kick in cause this same BSOD. In this case if it clogs up user space everything stops but may not get terminated. Very interesting to debug.
- Ian Yates May 12, 2020 0
  
  Definitely had this happen to me. Disabling that hdd tilt (was supposed to be drop, but any lap use resulted in constant pausing) detection was the best thing I did for perf on a couple of older laptops.