当前位置:网站首页>Netdata data data persistence configuration
Netdata data data persistence configuration
2022-06-29 13:14:00 【gsls200808】
By default ,netdata The storage time of historical data is short , Only circular storage 2 To 3 Days of historical data , So we need to change it netdata Configure to store data for longer periods of time .
Default netdata.conf The configuration file has no data , need curl wget Command to import templates
curl -o /etc/netdata/netdata.conf http://localhost:19999/netdata.conf
Modify the following
Earlier versions were in global In the label
vi /etc/netdata/netdata.conf
Edit the configuration file as follows
Legacy configuration global label
[global]
memory mode = dbengine
page cache size = 32
dbengine disk space = 3200
The new version is configured in db In the label
[db]
memory mode = dbengine
page cache size = 32
dbengine disk space =3200
This configuration configures 32M cache ,3.2G disk space , Can support Per second 2000 Data points ,30 God .
memory mode The default configuration is 32M cache ,256M disk space , The required disk space can be calculated according to the actual storage
dbengine The storage path of is
/var/cache/netdata/dbengine
View the database size
cd /var/cache/netdata/
du -sh dbenginemomory mode Meaning of options
1.dbengine,( Default ) The data is in the database file . The default configuration 32M cache ,256M disk space
2.ram, Data is in memory . The data will never be saved on the hard disk . This mode uses mmap() And support KSM.
3.save, netdata Run time memory , Drop disk and read disk during restart . It also uses mmap() And support KSM.
4.map, The data is in a memory mapped file . This will keep writing to your disk . This mode uses mmap() But does not support KSM.
5.none, No database ( Collected metrics can only be streamed to another Netdata).
6.alloc, Follow ram The pattern is the same, but it uses calloc() And does not support KSM. This mode can be removed from none Other modes of mode fallback .
边栏推荐
- Golang image/png 处理图片 旋转 写入
- 【云驻共创】工业智慧“大脑”,老厂焕新的加速秘籍
- 从零搭建Pytorch模型教程(四)编写训练过程--参数解析
- Comment calculer Win / Tai / Loss in paired t - test
- Simple introduction to matlab
- Pygame 精准检测图像碰撞
- Cvpr2022 𞓜 thin domain adaptation
- CVPR 2022 | unknown target detection module study: learning unknown targets in video
- Recurrence of recommended models (III): recall models youtubednn and DSSM
- Cvpr2022 | knowledge distillation through target aware transformer
猜你喜欢

Schiederwerk Power Supply repair smps12 / 50 pfc3800 Analysis

Viewing splitchunks code segmentation from MPX resource construction optimization

CVPR 2022 | 未知目标检测模块STUD:学习视频中的未知目标

Testing -- automated testing: about the unittest framework

Qt中的UI文件介绍

Cvpr2022 𞓜 loss problem in weakly supervised multi label classification

Yolo series combs (IX) first taste of newly baked yolov6

Cvpr2022 | reexamine pooling: your receptive field is not the best

Qt的信号与槽

Matlab简单入门
随机推荐
B+树|MYSQL索引使用原则
NvtBack
QQ group was stolen, a large-scale social death scene caught off guard
Precautions for Beifu controller connecting Panasonic EtherCAT servo
@Table爆红
Cvpr2022 | panopticdepth: a unified framework for depth aware panoramic segmentation
RT thread memory management
Yolo series combs (IX) first taste of newly baked yolov6
ZALSM_ EXCEL_ TO_ INTERNAL_ Solving the big problem of importing data from table
从零搭建Pytorch模型教程(五)编写训练过程--一些基本的配置
asp.net 项目使用aspnet_compiler.exe发布
[cloud native] 2.4 kubernetes core practice (middle)
C#通过线索二叉树进行中序遍历输出
Pygame 对图像进行翻转
Interview shock 61: tell me about MySQL transaction isolation level?
C#二叉树结构定义、添加节点值
PyGame flips the image
UI file introduction in QT
Cvpr2022 | a convnet for the 2020s & how to design neural network Summary
CVPR2022 | 可精简域适应