One of my client runs dedicated NS1 and NS2 to host more than 3000+ domains. Recently they upgraded their servers to latest Dual Core Dual AMD server with CentOS 5.0 and BIND server.
By default BIND / named will try to determine the number of CPUs present and create one thread per CPU. If it is unable to determine the number of CPUs, a single worker thread will be created. However due to some reason the BIND server failed to automatically utilize all of the system’s available CPUs. So how do you force DNS Server to take advantage of multiple CPUs under CentOS Linux?
After a little investigation, named man page pointed out me in right direction ~ -n #CPU option, which creates #cpus worker threads to take advantage of multiple CPUs.
Force BIND DNS Server to take advantage of multiple CPUs
In order to enable multiple CPU open /etc/sysconfig/named file under CentOS / RHEL / Fedora Linux:
# vi /etc/sysconfig/named
To force bind to take advantage of 4 CPUs, add / modify as follows:
Save and close the file. Restart named service:
# /etc/init.d/named restart
A note about Debian / Ubuntu Linux user
If you are a Debian / Ubuntu Linux modify /etc/defaults/bind9 file:
$ sudo vi /etc/defaults/bind9
Append config line:
Please note append -n 4 to the end of other options (if any), for example:
OPTIONS="-4 -6 -n 4"
Save and close the file. Restart BIND server:
$ sudo /etc/init.d/bind9 restart
More more information read named man page. Sysadmin because even developers need heroes!!!
Q. I would like to restrict number of CPU activated for some software licensing issues under Linux kernel 2.6.xx.. How can I limit the number of CPUs activated in SMP mode?
A. Pass a special parameter called maxcpus to kernel. It specify maximum number of processors that an SMP Linux kernel should make use of. For example if you have four cpus and would like to use 2 CPU then pass 2 as a number to maxcpus.
This is useful to test different software performances and configurations. Some commercial software such as ERP software or Oracle are licenced per CPU. In such a case maxcpus is a life saver.
Edit your grub.conf file:
# vi grub.conf
Append parameter maxcpus at the end of Kernel line. A the end it should read as follows:
kernel /boot/vmlinuz-2.6.13-Ora10g root=/dev/sda1 ro maxcpus=2
Save and close the file. Reboot system:
Alternatively, you need to enter parameter at grub or lilo boot prompt. For example if you are using Grub as a boot loader, at Grub prompt press ‘e’ to edit command before booting.
- Select second line
- Again, press ‘e’ to edit selected command
- Select kernel line
- Append maxcpus=2 parameters.
- Press b to boot system
UPDATED for accuracy.
When you are using SMP (Symmetric MultiProcessing) you might want to override the kernel’s process scheduling and bind a certain process to a specific CPU(s).
But what is CPU affinity?
CPU affinity is nothing but a scheduler property that “bonds” a process to a given set of CPUs on the SMP system. The Linux scheduler will honor the given CPU affinity and the process will not run on any other CPUs. Note that the Linux scheduler also supports natural CPU affinity:
The scheduler attempts to keep processes on the same CPU as long as practical for performance reasons. Therefore, forcing a specific CPU affinity is useful only in certain applications. For example, application such as Oracle (ERP apps) use # of cpusÃ‚ per instance licensed. You can bound Oracle to specific CPU to avoid license problem. This is a really useful on large server having 4 or 8 CPUS
Setting processor affinity for a certain task or process using taskset command
taskset is used to set or retrieve the CPU affinity of a running process given its PID or to launch a new COMMAND with a given CPU affinity. However taskset is not installed by default. You need to install schedutils (Linux scheduler utilities) package.
# apt-get install schedutils
Red Hat Enterprise Linux:
# up2date schedutils
# rpm -ivh schedutils*
Under latest version of Debian / Ubuntu Linux taskset is installed by default using util-linux package.
The CPU affinity is represented as a bitmask, with the lowest order bit corresponding to the first logical CPU and the highest order bit corresponding to the last logical CPU. For example:
- 0x00000001 is processor #0 (1st processor)
- 0x00000003 is processors #0 and #1
- 0x00000004 is processors #2 (3rd processor)
To set the processor affinity of process 13545 to processor #0 (1st processor) type following command:
# taskset 0x00000001 -p 13545
If you find a bitmask hard to use, then you can specify a numerical list of processors instead of a bitmask using -c flag:
# taskset -c 1 -p 13545
# taskset -c 3,4 -p 13545
- -p : Operate on an existing PID and not launch a new task (default is to launch a new task)
UPDATED for accuracy.
Whenever a Linux system CPU is occupied by a process, it is unavailable for processing other requests. Rest of pending requests must wait till CPU is free. This becomes a bottleneck in the system. Following command will help you to identify CPU utilization, so that you can troubleshoot CPU related performance problems on a Linux based system.
[click to continue…]