Before Release 4.0, this parameter determined the number of entries (and therefore the size) of a shared memory cache used by the spool work process of an R/3 spool server. The cache had to be large enough to hold an entry for each printer served by a spool work process.
At customers with very large numbers of printers, this implementation required that large areas (multiple MB; around 750 bytes per entry) of shared memory be allocated for this printer cache. Further, if a cache entry could not be allocated to an R/3 printer, then the spool work process did not send output requests to the printer.
Example: if a spool server sends output to 2,000 devices, then the parameter value had to be set to at least 2,000.
In Release 4.0, the parameter still determines the size of a shared memory cache used by the R/3 spool system. But the use and management of the cache have been changed:
The changes to the printer cache make it possible to reduce the value of the RSPO/GLOBAL_SHM/PRINTER_LIST parameter.
You should check the value of the parameter in each of your R/3 instance profiles using report RSPFPAR (on each of your R/3 instances -- servers that use the same instance profile) or the CCMS profile management tool (transaction RZ10).
If the parameter has been set to a value higher than the current default, then you can lower it to the default value of 300 (unless you expect more than 300 printers to work on R/3 output requests at the same time).
You should not lower the value below the default value of 300.