Mac OS X Server v10.6: Understanding process limits

Summary

In versions of Mac OS X Server prior to version 10.6 (Snow Leopard) the maximum number of processes available on a server was limited to 2500.  The 64-bit kernel included in Mac OS X Server v10.6 automatically scales the maximum number of processes, vnodes (open file handles), and threads based on the amount of memory installed in the server.

Products Affected

Mac OS X Server 10.6

For hardware that supports the 64-bit kernel, the following limits are automatically set by the operating system based on the amount of memory present:

  • For each 8GB of installed memory, 2500 processes and 150,000 vnodes are available.  The maximum number of threads is set to five times (5x) the number of maximum processes.
  • A single user ID (uid) can use up to 75% of the maximum number of processes.
  • A single process can allocate up to 20% of the maximum threads value.

The following Terminal commands can be used to read the values on the running system:

sysctl kern.maxproc
sysctl kern.maxvnodes
sysctl kern.maxfiles

Additional Information

This table details the expanded limits as memory is added to a Mac OS X Server v10.6-based server:

8 2500 150000 12500
16 5000 300000 25000
24 7500 450000 37500
32 10000 600000 50000
40 12500 750000 62500
48 15000 900000 75000
56 17500 1050000 87500
64 20000 1200000 100000
72 22500 1350000 112500
80 25000 1500000 125000
88 27500 1650000 137500
96 30000 1800000 150000
Not helpful Somewhat helpful Helpful Very helpful Solved my problem
Ask other users about this article
in Apple Support Communities
See all questions on this article See all questions I have asked