Linux Memory Recovery kernel parameters

Source: Internet
Author: User

ss-atu| awk '/^tcp/{++s[$2]} END {for (a in S) print A,s[a]} '

PS up PID (RSS: Actual memory size, long-standing memory)

PS o pid,comm,minflt,majflt PID (second page interrupt, page break) page break: The process of allocating memory is called page Break home interrupt from swap, (home interrupt too much will affect performance) page break from memory page error: Re-establish physical memory with virtual inside Correspondence between the existing

Vm.swappiness rhel-7 (30%) 100 tendencies with swap 0 tend to buff/cache (memory)

Memory Reclamation Parameters:

Vm.dirty_expire_centisecs = 3000 (Dirty page aging time, default 30 seconds, how long to brush the disk once)

Vm.dirty_writeback_centisecs = How often does Pdflush work (monitoring)

Vm.dirty_background_ratio = 10 If the dirty page reaches the remaining 10%,kernel in tune, not Pdflush

Vm.dirty_ratio = 30 Low memory 30 seconds once (to 20%) small data file, then a number of more, you can appropriately set this value a little bit better performance. (Brush disk slow)

vm.dirty_bytes = 0; vm.dirty_background_bytes = 0 Arrival byte


Oom Killer

Vm.panic_on_oom = 0 0: Open 1: Off

/proc/pid/oom_adj Range: 17 to 15 (-17 smaller less likely to be killed)


Please refer to Kernel-doc for details.

Catalog:/USR/SHARE/DOC/KERNEL-DOC-*/DOCUMENTATION/VM Memory



Internet

sysctl-a | grep Rmem

Net.core.rmem_default = 212992

Net.core.rmem_max = 212992

Net.ipv4.tcp_rmem = 4096873806291456

Net.ipv4.udp_rmem_min = 4096


sysctl-a | grep Tcp_mem

Net.ipv4.tcp_mem = 422165629184432


Overcommit_memory = 0 Default (3 modes)

0: The kernel checks to see if there is enough memory available for the process to use, and if enough memory is available, the memory request is allowed; otherwise, the memory request fails and the error is returned to the application process

1: Always allow excessive use, suitable for some scientific applications, classic example is code using sparse array

The composition relies almost on virtual memory.

2: Do not allow excessive use,

Commit is rejected when the system allocates more memory than swap+50% (the default) physical RAM (n% is determined by Vm.overcommit_ratio).

The Overcommit policy is set through Vm.overcommit_memory.

The percentage of Overcommit is set by Vm.overcommit_ratio.


Vm.overcommit_ratio = 50



Vm.min_free_kbytes




This article is from the "Silence" blog, make sure to keep this source http://silencezone.blog.51cto.com/3613477/1870881

Linux Memory Recovery kernel parameters

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.