Delayed Return of Paged Pool Causes Error 1450 "Insufficient Resources"ID: Q236964
|
The problem described in this article is difficult to identify. One way to check for the problem if you suspect you are experiencing it is to calculate the number of files open on the Window NT-based computer and look at their size. If you have approximately 180 gigabytes (GB) of open files that are not being closed continuously or are opened and closed rapidly in a looping condition, you may be experiencing this problem. Also, your programs may be receiving an INSUFFICIENT_RESOURCES error message when attempting to open, create, or extend a file.
This problem can be intermittent. If you try the operation again it is possible for it to succeed. This behavior occurs because it is possible to have a program create and close files faster than Windows NT can clean up. When this occurs it is possible for the operation to succeed if retried.
Another method that you can use to identify the problem is to start Task Manager and click the Performance tab. At the bottom of the tab, locate the Kernel Memory section and watch the value for Paged Memory. You may experience the problem when that value reaches approximately 190. If this value goes back down immediately, you should be able to retry the operation that caused the error message and have it succeed (if the system has enough resources to fulfill the program's request).
In Windows NT 4.0 Service Pack 4 (SP4), optimization was added to the cache manager and the NTFS file system that allows NTFS to return a FALSE response when the cache manager tries to acquire a file for "lazy" writing. NTFS does this when it is in the middle of a clean checkpoint, because clean checkpoints can take time to finish. This allows the cache manager to move on to other files on other volumes instead of waiting for the clean checkpoint to finish.
In this case, NTFS needs to free recently deallocated clusters. NTFS does this by forcing a clean checkpoint. In this particular situation, the cache manager should wait for the checkpoint to finish instead of moving on to work on other files, because it is not really a clean checkpoint and should not take very long.
The solution is not to let NTFS return a FALSE response to the cache manager when it is performing a clean checkpoint as a result of needing recently deallocated clusters freed.
A supported fix that corrects this problem is now available from Microsoft, but
it has not been fully regression tested and should be applied only to systems
experiencing this specific problem. If you are not severely affected by this
specific problem, Microsoft recommends that you wait for the next Windows NT 4.0 service pack
that contains this fix.
To resolve this problem immediately, contact Microsoft Product Support Services
to obtain the fix. For a complete list of Microsoft Product Support Services
phone numbers and information on support costs, please go to the following
address on the World Wide Web:
http://www.microsoft.com/support/supportnet/overview/overview.aspThe English version of this fix should have the following file attributes or later:
Date Time Size File name Platform
-------------------------------------------------------------
07/01/99 03:35p 363,152 Ntfs.sys x86
07/01/99 03:34p 559,248 Ntfs.sys Alpha
Q154871 Determining If You Are Eligible for No-Charge Technical Support
Microsoft has confirmed this to be a problem in Windows NT 4.0.
Additional query words: insufficient resources
Keywords : kbbug4.00 kbfix4.00
Version : winnt:4.0 SP4
Platform : winnt
Issue type : kbbug
Last Reviewed: August 3, 1999