对docker container进行内核参数调优(tune)

对于docker container的调优还是和普通的Linux调优有很大的区别

修改最大文件数(open files)

直接修改 container的 /etc/security/limits.conf无效
正确的做法是(以CentOS host为例)

在host上执行

sudo sh -c 'printf "\nulimit -HSn 999999\n" >> /etc/sysconfig/docker'
sudo service docker restart

这样再进入docker container, ulimit -a就可以看到更改了

[root@f258c7598394 /]# ulimit -a
core file size          (blocks, -c) 0
data seg size           (kbytes, -d) unlimited
scheduling priority             (-e) 0
file size               (blocks, -f) unlimited
pending signals                 (-i) 7810
max locked memory       (kbytes, -l) 64
max memory size         (kbytes, -m) unlimited
open files                      (-n) 999999
pipe size            (512 bytes, -p) 8
POSIX message queues     (bytes, -q) 819200
real-time priority              (-r) 0
stack size              (kbytes, -s) 10240
cpu time               (seconds, -t) unlimited
max user processes              (-u) 7810
virtual memory          (kbytes, -v) unlimited
file locks                      (-x) unlimited

对于Ubuntu请参考此文

修改tcp/ip变量(sysctl)

  • 如果docker container是以 –net=”host” 方式启动的,那么container和host共用同一个网络堆栈,只需要修改host的配置
sudo sh -c 'printf "\nnet.ipv4.ip_local_port_range = 1024 65535\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.ipv4.tcp_tw_recycle = 1\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.ipv4.tcp_tw_reuse = 1\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.core.rmem_max = 16777216\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.core.wmem_max = 16777216\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.ipv4.tcp_max_syn_backlog = 4096\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.ipv4.tcp_syncookies = 1\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.core.somaxconn = 1024\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.ipv4.tcp_window_scaling = 1\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.ipv4.tcp_rmem = 4096 87380 16777216\n" >> /etc/sysctl.conf';
sudo sh -c 'printf "net.ipv4.tcp_wmem = 4096 16384 16777216\n" >> /etc/sysctl.conf';
sudo sysctl -p;

然后重启docker container生效。

  • 如果docker container不是以 –net=”host” 方式启动的,那么它将有自己独立的网络堆栈。修改host的配置将会无效。在container中又无法直接修改/proc,因为docker会以只读的方式重新挂载/proc/sys。 对于这个问题,可以在container启动的时候将/proc挂载到另一可读写位置,譬如
docker run -ti -v /proc:/writable-proc ubuntu:14.04 /bin/bash

然后就可以在container内部进行修改了

echo 1024 65535 > /writable-proc/sys/net/ipv4/ip_local_port_range
echo 1 > /writable-proc/sys/net/ipv4/tcp_tw_recycle
echo 1 > /writable-proc/sys/net/ipv4/tcp_tw_reuse
echo 4096 > /writable-proc/sys/net/ipv4/tcp_max_syn_backlog
echo 1 > /writable-proc/sys/net/ipv4/tcp_syncookies
echo 1 > /writable-proc/sys/net/ipv4/tcp_window_scaling
echo 4096 16384 16777216 > /writable-proc/sys/net/ipv4/tcp_wmem
echo 4096 87380 16777216 > /writable-proc/sys/net/ipv4/tcp_rmem
echo 16777216 > /writable-proc/sys/net/core/rmem_max
echo 16777216 > /writable-proc/sys/net/core/wmem_max
echo 1024 > /writable-proc/sys/net/core/somaxconn
echo 999999 > /writable-proc/sys/fs/file-max

然后在container中可以查看到修改生效

cat /proc/sys/net/ipv4/ip_local_port_range
cat /proc/sys/net/ipv4/tcp_tw_recycle
cat /proc/sys/net/ipv4/tcp_tw_reuse
cat /proc/sys/net/ipv4/tcp_max_syn_backlog
cat /proc/sys/net/ipv4/tcp_syncookies
cat /proc/sys/net/ipv4/tcp_window_scaling
cat /proc/sys/net/ipv4/tcp_wmem
cat /proc/sys/net/ipv4/tcp_rmem
cat /proc/sys/net/core/rmem_max
cat /proc/sys/net/core/wmem_max
cat /proc/sys/net/core/somaxconn
cat /proc/sys/fs/file-max

需要注意的是,如果内核版本过低(<3.12), 有些参数是无法修改的

上面提到的Linux内核各调优参数可以参考此文

你可能感兴趣的:(docker,container)