Applies To:
  • CitectSCADA 5.41
  • CitectHMI 5.41
Summary:
I would be nice to have additional checks besides the [ioserver]dcbprotect=1 available while drivers are in test or if unknown crashes occur which seem to be driver related. 

Solution:
A new option [debug]drivercheck=1 is available.

When this is set, the structure used to talk to drivers is monitored for changes.

If an error occurs a kernel and syslog message is put out .

With [debug]drivercheck=1 set, the DCB is checksummed

On return the checksum is checked as well as a flag at the end of the DCB buffer . e.g.

*** WARNING: DriverCheck=1 DCB->buffer overrun ***

or

*** WARNING: DriverCheck=1 DCB corrupted/changed ***

Not all instances of this may be fatal or an issue, if a driver cleared the whole DCB buffer of 554 bytes, then expect this error !

if a driver modifies the POINT information then expect the DCB corruption message.

This option is for driver developers or in special cases as directed by Citect Support.

 
Keywords:
 

Attachments