[comp.unix.xenix] SCO Xenix, fork - "too many files"?

toppin@melpar.UUCP (Doug Toppin) (11/18/89)

We are running SCO Xenix 2.2 on the 286.
I am running an application that initializes itself and then
starts a number of others in the following manner:
    attach to shared memory seg 1
    attach to shared memory seg 2
    for 10 processes
        switch fork
            child
                open window device for stdio
                exec application to run
On the second process started, the fork fails with the error "too many files".
Configure says these are set:
NINODE=200, NFILE=200, NMOUNT=1, NFLOCKS=1, NPROC=60, MAXUPRC=30, NTEXT=60
I can't figure out which parameter is associated with with max files
open per user process (if there is one). If anyone knows what I'm doing
wrong please let me know. This is the only process I'm running and there
are maybe 10 other processes running.
thanks
Doug Toppin
uunet!melpar!toppin

cpcahil@virtech.uucp (Conor P. Cahill) (11/21/89)

In article <248@melpar.UUCP>, toppin@melpar.UUCP (Doug Toppin) writes:
> We are running SCO Xenix 2.2 on the 286.

  [ description of problem with forking deleted]

> On the second process started, the fork fails with the error "too many files".

By "too many files" I presume you mean "too many open files" as in

	#define	EMFILE	24	/* Too many open files			*/

From the fork(s) man page:

	fork will fail....

		The shared memory table overflows [EMFILE].

> Configure says these are set:
> NINODE=200, NFILE=200, NMOUNT=1, NFLOCKS=1, NPROC=60, MAXUPRC=30, NTEXT=60

None of these configuration parameters should have any effect on this problem.
I would suggest reviewing your /usr/sys/conf/master file to see which shared
memory variables you need to change.  I don't know the correct one off the
top of my head, 



-- 
+-----------------------------------------------------------------------+
| Conor P. Cahill     uunet!virtech!cpcahil      	703-430-9247	!
| Virtual Technologies Inc.,    P. O. Box 876,   Sterling, VA 22170     |
+-----------------------------------------------------------------------+