The sysctl preload/configuration file can be created at
/etc/sysctl.d/99-sysctl.conf. For systemd,
/usr/lib/sysctl.d/ are drop-in directories for kernel sysctl parameters. The naming and source directory decide the order of processing, which is important since the last parameter processed may override earlier ones. For example, parameters in a
/usr/lib/sysctl.d/50-default.conf will be overriden by equal parameters in
/etc/sysctl.d/50-default.conf and any configuration file processed later from both directories.
To load all configuration files manually, execute
# sysctl --system
which will also output the applied hierarchy. A single parameter file can also be loaded explicitly with
# sysctl -p filename.conf
See the new configuration files and more specifically for more information.
The parameters available are those listed under
/proc/sys/. For example, the
kernel.sysrq parameter refers to the file
/proc/sys/kernel/sysrq on the file system. The
sysctl -a command can be used to display all currently available values.
Settings can be changed through file manipulation or using the
sysctl utility. For example, to temporarily enable the magic SysRq key:
# sysctl kernel.sysrq=1
# echo "1" > /proc/sys/kernel/sysrq
To preserve changes between reboots, add or modify the appropriate lines in
/etc/sysctl.d/99-sysctl.conf or another applicable parameter file in
# The maximum size of the receive queue. # The received frames will be stored in this queue after taking them from the ring buffer on the NIC. # Use high value for high speed cards to prevent loosing packets. # In real time application like SIP router, long queue must be assigned with high speed CPU otherwise the data in the queue will be out of date (old). net.core.netdev_max_backlog = 65536
# The maximum ancillary buffer size allowed per socket. # Ancillary data is a sequence of struct cmsghdr structures with appended data. net.core.optmem_max = 65536
# The upper limit on the value of the backlog parameter passed to the listen function. # Setting to higher values is only needed on a single highloaded server where new connection rate is high/bursty net.core.somaxconn = 16384
# The default and maximum amount for the receive/send socket memory # By default the Linux network stack is not configured for high speed large file transfer across WAN links. # This is done to save memory resources. # You can easily tune Linux network stack by increasing network buffers size for high-speed networks that connect server systems to handle more network packets. net.core.rmem_default = 1048576 net.core.wmem_default = 1048576 net.core.rmem_max = 16777216 net.core.wmem_max = 16777216 net.ipv4.tcp_rmem = 4096 87380 16777216 net.ipv4.tcp_wmem = 4096 65536 16777216 net.ipv4.udp_rmem_min = 16384 net.ipv4.udp_wmem_min = 16384
# An extension to the transmission control protocol (TCP) that helps reduce network latency by enabling data to be exchanged during the sender’s initial TCP SYN. # If both of your server and client are deployed on Linux 3.7.1 or higher, you can turn on fast_open for lower latency net.ipv4.tcp_fastopen = 3
# The maximum queue length of pending connections 'Waiting Acknowledgment' # In the event of a synflood DOS attack, this queue can fill up pretty quickly, at which point tcp_syncookies will kick in allowing your system to continue to respond to legitimate traffic, and allowing you to gain access to block malicious IPs. # If the server suffers from overloads at peak times, you may want to increase this value a little bit. net.ipv4.tcp_max_syn_backlog = 65536
# The maximum number of sockets in 'TIME_WAIT' state. # After reaching this number the system will start destroying the socket in this state. # Increase this to prevent simple DOS attacks net.ipv4.tcp_max_tw_buckets = 65536
# Whether TCP should start at the default window size only for new connections or also for existing connections that have been idle for too long. # It kills persistent single connection performance and should be turned off. net.ipv4.tcp_slow_start_after_idle = 0
# Whether TCP should reuse an existing connection in the TIME-WAIT state for a new outgoing connection if the new timestamp is strictly bigger than the most recent timestamp recorded for the previous connection. # This helps avoid from running out of available network sockets. net.ipv4.tcp_tw_reuse = 1
# Fast-fail FIN connections which are useless. net.ipv4.tcp_fin_timeout = 15
# TCP keepalive is a mechanism for TCP connections that help to determine whether the other end has stopped responding or not. # TCP will send the keepalive probe contains null data to the network peer several times after a period of idle time. If the peer does not respond, the socket will be closed automatically. # By default, TCP keepalive process waits for two hours (7200 secs) for socket activity before sending the first keepalive probe, and then resend it every 75 seconds. As long as there is TCP/IP socket communications going on and active, no keepalive packets are needed. # With the following settings, your application will detect dead TCP connections after 120 seconds (60s + 10s + 10s + 10s + 10s + 10s + 10s) net.ipv4.tcp_keepalive_time = 60 net.ipv4.tcp_keepalive_intvl = 10 net.ipv4.tcp_keepalive_probes = 6
# The longer the MTU the better for performance, but the worse for reliability. # This is because a lost packet means more data to be retransmitted and because many routers on the Internet can't deliver very long packets. # Enable smart MTU discovery when an ICMP black hole detected. net.ipv4.tcp_mtu_probing = 1
# Turn timestamps off to reduce performance spikes related to timestamp generation. net.ipv4.tcp_timestamps = 0
- You can use http://www.speedtest.net to benchmark internetwork performance before/after the change
TCP/IP stack hardening
The following specifies a parameter set to tighten network security options of the kernel for the IPv4 protocol and related IPv6 parameters where an equivalent exists.
For some usecases, for example using the system as a router, other parameters may be useful or required as well.
#### ipv4 networking and equivalent ipv6 parameters #### ## TCP SYN cookie protection (default) ## helps protect against SYN flood attacks ## only kicks in when net.ipv4.tcp_max_syn_backlog is reached net.ipv4.tcp_syncookies = 1 ## protect against tcp time-wait assassination hazards ## drop RST packets for sockets in the time-wait state ## (not widely supported outside of linux, but conforms to RFC) net.ipv4.tcp_rfc1337 = 1 ## sets the kernels reverse path filtering mechanism to value 1 (on) ## will do source validation of the packet's recieved from all the interfaces on the machine ## protects from attackers that are using ip spoofing methods to do harm net.ipv4.conf.default.rp_filter = 1 net.ipv4.conf.all.rp_filter = 1 ## tcp timestamps ## + protect against wrapping sequence numbers (at gigabit speeds) ## + round trip time calculation implemented in TCP ## - causes extra overhead and allows uptime detection by scanners like nmap ## enable @ gigabit speeds net.ipv4.tcp_timestamps = 0 #net.ipv4.tcp_timestamps = 1 ## log martian packets net.ipv4.conf.default.log_martians = 1 net.ipv4.conf.all.log_martians = 1 ## ignore echo broadcast requests to prevent being part of smurf attacks (default) net.ipv4.icmp_echo_ignore_broadcasts = 1 ## ignore bogus icmp errors (default) net.ipv4.icmp_ignore_bogus_error_responses = 1 ## send redirects (not a router, disable it) net.ipv4.conf.default.send_redirects = 0 net.ipv4.conf.all.send_redirects = 0 ## ICMP routing redirects (only secure) #net.ipv4.conf.default.secure_redirects = 1 (default) #net.ipv4.conf.all.secure_redirects = 1 (default) net.ipv4.conf.default.accept_redirects=0 net.ipv4.conf.all.accept_redirects=0 net.ipv6.conf.default.accept_redirects=0 net.ipv6.conf.all.accept_redirects=0
There are several key parameters to tune the operation of the virtual memory (VM) subsystem of the Linux kernel and the write out of dirty data to disk. See the official Linux kernel documentation for more information. For example:
vm.dirty_ratio = 3
- Contains, as a percentage of total available memory that contains free pages and reclaimable pages, the number of pages at which a process which is generating disk writes will itself start writing out dirty data.
vm.dirty_background_ratio = 2
- Contains, as a percentage of total available memory that contains free pages and reclaimable pages, the number of pages at which the background kernel flusher threads will start writing out dirty data.
As noted in the comments for the parameters, one needs to consider the total amount of RAM when setting these values. For example, simplifying by taking the installed system RAM instead of available memory:
- Consensus is that setting
vm.dirty_ratioto 10% of RAM is a sane value if RAM is say 1 GB (so 10% is 100 MB). But if the machine has much more RAM, say 16 GB (10% is 1.6 GB), the percentage may be out of proportion as it becomes several seconds of writeback on spinning disks. A more sane value in this case is 3 (3% of 16 GB is approximately 491 MB).
- Similarly, setting
vm.dirty_background_ratioto 5 may be just fine for small memory values, but again, consider and adjust accordingly for the amount of RAM on a particular system.
Another parameter is:
vm.vfs_cache_pressure = 60
- The value controls the tendency of the kernel to reclaim the memory which is used for caching of directory and inode objects (VFS cache). Lowering it from the default value of 100 makes the kernel less inclined to reclaim VFS cache (do not set it to 0, this may produce out-of-memory conditions).
When the kernel performs a resync operation of a software raid device it tries not to create a high system load by restricting the speed of the operation. Using sysctl it is possible to change the lower and upper speed limit.
# Set maximum and minimum speed of raid resyncing operations dev.raid.speed_limit_max = 10000 dev.raid.speed_limit_min = 1000
If mdadm is compiled as a module
md_mod, the above settings are available only after the module has been loaded. If the settings shall be loaded on boot via
/etc/sysctl.d, the module
md_mod may be loaded beforehand through
Small periodic system freezes
Set dirty bytes to small enough value (for example 4M):
vm.dirty_background_bytes = 4194304 vm.dirty_bytes = 4194304
Try to change
kernel.io_delay_type (x86 only):
- 0 - IO_DELAY_TYPE_0X80
- 1 - IO_DELAY_TYPE_0XED
- 2 - IO_DELAY_TYPE_UDELAY
- 3 - IO_DELAY_TYPE_NONE