This article has been archived. It is offered "as is" and will no longer be updated.
A client program that uses Remote Procedure Call (RPC) and is running on Microsoft Windows XP may randomly experience access violations when the program is communicating with an RPC server that is running on a UNIX server.
This problem is caused by an incorrect buffer in the RPC run-time DLL (Rpcrt4.dll). The buffer overrun occurs only if a client makes an outgoing RPC call to a UNIX server after the client has been idle for between 40 and 60 seconds, and if the RPC call expects a large response. The overrun buffer corrupts the memory block that follows the buffer that is used by RPC, and may cause random access violations in the client program.
Service pack information
To resolve this problem, obtain the latest service pack for Microsoft Windows XP. For additional information, click the following article number to view the article in the Microsoft Knowledge Base:
322389 How to obtain the latest Windows XP service pack
A supported hotfix is available from Microsoft. However, this hotfix is intended to correct only the problem that is described in this article. Apply this hotfix only to systems that are experiencing this specific problem. This hotfix might receive additional testing. Therefore, if you are not severely affected by this problem, we recommend that you wait for the next software update that contains this hotfix.
If the hotfix is available for download, there is a "Hotfix download available" section at the top of this Knowledge Base article. If this section does not appear, contact Microsoft Customer Service and Support to obtain the hotfix.
Note If additional issues occur or if any troubleshooting is required, you might have to create a separate service request. The usual support costs will apply to additional support questions and issues that do not qualify for this specific hotfix. For a complete list of Microsoft Customer Service and Support telephone numbers or to create a separate service request, visit the following Microsoft Web site:
Note The "Hotfix download available" form displays the languages for which the hotfix is available. If you do not see your language, it is because a hotfix is not available for that language. The English version of this fix has the file attributes (or later) that are listed in the following table. The dates and times for these files are listed in coordinated universal time (UTC). When you view the file information, it is converted to local time. To find the difference between UTC and local time, use the Time Zone tab in the Date and Time tool in Control Panel.
Date Time Version Size File name ------------------------------------------------------ 23-Dec-2002 09:31 5.1.2600.1154 505,856 Rpcrt4.dll
Microsoft has confirmed that this is a problem in the Microsoft products that are listed in the "Applies to" section. This problem was first corrected in Microsoft Windows XP Service Pack 2.
Some RPC servers may request that a client program end its idle connections to reduce resource usage. This is done by sending a shutdown Protocol Data Unit (PDU) to the client. On the client, RPC checks its receive buffer only when it is making an outgoing call to the RPC server. Some UNIX servers send shutdown PDUs after every 20 seconds of idle time. If the client is idle long enough for the server to send 2 consecutive shutdown PDUs, RPC on the client coalesces the PDUs in its receive buffer. If the client decides to make an outgoing call at this time, it tries to process the two coalesced PDUs in its receive buffer first. During processing, the coalescing causes the receive buffer to be reallocated. However, buffer-size information is not tracked correctly. This can result in a buffer that is reported to be larger than it really is. The outgoing call is then made. If the response is larger than the real (not the reported) size of the reallocated receive buffer, an access violation occurs.
If the client is idle long enough for the server to send three shutdown PDUs, RPC closes the connection. A new connection is then created to run the outgoing call. In this case, the problem does not occur. Therefore, the problem is likely to occur after the client program is idle for more than 40 seconds, but less than 60 seconds. However, this depends on how long the UNIX server waits before sending a shutdown PDU.