Difference between revisions of "Linux utils"

From Wasya Wiki
Jump to: navigation, search
(→‎Utils)
(→‎Disks, Storage)
 
(97 intermediate revisions by the same user not shown)
Line 1: Line 1:
 +
= Utils =
 +
* [[ css ]]
  
 +
* [[ Git cheatsheet ]]
 +
* [[ IO Utils ]]
 +
* [[ json ]]
 +
* [[ kubernetes ]]
 +
* [[ Memory Utils ]]
  
−
* [[Pi Roles]]
+
* [[ Network Utils ]]
 +
* [[ Process Utils ]]
 +
* [[ xml ]]
  
−
=== Nodes ===
+
=== Replace all occurrences of a string in a folder ===
 +
ubuntu?
 +
find /home/user/ -type f | xargs sed -i  's/a.example.com/b.example.com/g'
 +
mac:
 +
<pre>
 +
LC_ALL=C find . -type f -name "*.html" -exec sed -i '' "s/UA-47088821-3/UA-47088821-4/" {} +
 +
</pre>
  
−
This one, before I started sticking Node.js in there.
+
=== last 3 accessed files ===
 +
  find . -type f -exec stat -c '%X %n' {} \; | sort -nr | awk 'NR==1,NR==3 {print $2}'
  
−
WARNING! in the codebase node listens on 3000. In production it listens on 3007.
+
=== last 3 modified files ===
 +
  find . -type f -exec stat -c '%Y %n' {} \; | sort -nr | awk 'NR==1,NR==3 {print $2}'
  
−
<pre>
+
=== Find 10 largest files in a folder ===
−
piousbox@piousbox-laptop:~/projects/rails-quick-start$ knife node show app_server_21
+
−
Node Name:  app_server_21
+
−
Environment: _default
+
−
FQDN:       
+
−
IP:          10.0.1.151
+
−
Run List:    role[base], role[ish_static], role[marchesi_static], role[wiki_cac], role[wp_wasya]
+
−
Roles:      base, ish_static, marchesi_static, wiki_cac, wp_wasya
+
−
Recipes:    zsh, users::sysadmins, sudo, apt, nagios::client, git, build-essential, ish::base_apache, ish::static_site, ish_apache::static_site, ish_apache::base_apache, ish::base_php, mysql::client, mediawiki::default, mediawiki::appserver, wordpress::default
+
−
Platform:    ubuntu 12.04
+
−
Tags:     
+
−
</pre>
+
  
−
This is a successful deployment except change node's listening from 3000 to 3007:
+
  du -a /var | sort -n -r | head -n 10
−
<pre>
+
   du -a .    | sort -n -r | head -n 15
−
piousbox@piousbox-laptop:~/projects/rails-quick-start$ knife node show app_server_21
+
−
Node Name:   app_server_21
+
−
Environment: _default
+
−
FQDN:       
+
−
IP:          10.0.1.151
+
−
Run List:   role[base], role[node_exampler]
+
−
Roles:      base, node_exampler
+
−
Recipes:    zsh, users::sysadmins, sudo, apt, nagios::client, git, build-essential, node::base, node::upstream
+
−
Platform:    ubuntu 12.04
+
−
Tags:       
+
−
</pre>
+
  
−
The load balancer:
+
For mac:  
−
<pre>
+
−
ubuntu@ip-10-0-0-52:/etc/apache2/sites-available$ pwd
+
−
/etc/apache2/sites-available
+
−
ubuntu@ip-10-0-0-52:/etc/apache2/sites-available$ cat startups_staging
+
  
−
<VirtualHost *:80>
+
du -a . | sort -n -r | head
−
  ServerAdmin piousbox@gmail.com
+
−
  ServerName annesque.com
+
−
  KeepAlive On
+
−
  ProxyVia On
+
−
  ProxyPreserveHost On
+
−
  ServerAlias staging.annesque.com
+
−
  ServerAlias startups.piousbox.com
+
−
  ServerAlias staging.startups.piousbox.com
+
−
  ProxyPreserveHost On
+
−
  ProxyVia On
+
−
  <proxy>
+
−
    Order deny,allow
+
−
    Allow from all
+
−
  </proxy>
+
−
  ProxyPass /api/ http://10.0.1.151:3007/ connectiontimeout=15 timeout=45
+
−
  ProxyPassReverse /api/ http://10.0.1.151:3007/
+
−
  ProxyPass / http://10.0.1.72:3010/ connectiontimeout=15 timeout=45
+
−
  ProxyPassReverse / http://10.0.1.72:3010/
+
−
</VirtualHost>
+
−
</pre>
+
  
−
=== Utils ===
+
Limit how much logs journalctl keeps:
−
Remember the virtual site for combining a node backend into some other service?
+
  
 +
  journalctl --vacuum-time=2d
 +
  journalctl --vacuum-size=500M
 +
 +
=== Count Lines in a Folder ===
 +
find . -name "*js" -print | grep -v node_modules | xargs cat | grep -v "^//" | grep -v "^$" | wc -l
 +
 +
 +
=== cut after ag for shorter lines ===
 +
 +
ag <your-search-keyword> | cut -c-800
 +
 +
=== increase number of open files, file descriptors ===
 +
 +
From: /etc/security/limits.conf
 
<pre>
 
<pre>
−
<VirtualHost *:3098>
+
* - nofile 131072
−
  ServerAdmin piousbox@gmail.com
+
 
−
  ServerName localhost
+
*    soft nofile 64000
−
  KeepAlive On
+
*    hard nofile 64000
−
  ProxyVia On
+
root soft nofile 64000
−
  ProxyPreserveHost On
+
root hard nofile 64000
−
  <proxy>
+
−
    Order deny,allow
+
−
    Allow from all
+
−
  </proxy>
+
−
  ProxyPass /api/ http://localhost:3000/ connectiontimeout=15 timeout=45
+
−
  ProxyPassReverse /api/ http://localhost:3000/
+
−
  ProxyPass / http://localhost:8000/ connectiontimeout=15 timeout=45
+
−
  ProxyPassReverse / http://localhost:8000/
+
−
</VirtualHost>
+
 
</pre>
 
</pre>
  
−
==== Establish a simple ssh port forward ====
+
Reload services after making changes, logout.
  
−
10.0.0.x is public, 10.0.1.x is private. Do:
+
=== count open files ===
 +
lsof | wc -
 +
Per user:
 +
lsof | grep ' opera ' | awk '{print $NF}' | sort | wc -l
  
−
sudo sysctl net.ipv4.ip_forward=1
+
=== Set date ===
−
  sudo iptables -t nat -A PREROUTING -p tcp -d 10.0.0.229 --dport 2270 -j DNAT --to-destination 10.0.1.224: 22
+
  sudo date --set "25 Sep 2013 15:00:00"
−
sudo iptables -t nat -A POSTROUTING -j MASQUERADE
+
  
−
=== Linking ===
+
=== Get date in milliseconds ===
−
<pre>
+
date -d 2017-05-05 +%s
−
ln -s /home/ubuntu/projects/ish_lib/current/lib          /home/ubuntu/projects/microsites2-cities/current
+
−
ln -s /home/ubuntu/projects/ish_lib/current/app/models    /home/ubuntu/projects/microsites2-cities/current/app
+
−
ln -s /home/ubuntu/projects/ish_lib/current/vendor/assets /home/ubuntu/projects/microsites2-cities/current/vendor
+
  
  
−
mkdir -p /var/lib/tftpboot/images/nexenta3/platform/i86pc/kernel/amd64
 
−
mkdir -p /var/lib/tftpboot/images/nexenta3/platform/i86pc/amd64
 
−
cp /mnt/nexenta3/platform/i86pc/amd64/miniroot /var/lib/tftpboot/images/nexenta3/platform/i86pc/amd64/miniroot
 
−
cp /mnt/nexenta3/platform/i86pc/kernel/amd64/unix /var/lib/tftpboot/images/nexenta3/platform/i86pc/kernel/amd64/unix
 
−
</pre>
 
  
−
=== Chef startup test requirements - Steps [startups.git] ===
+
== Disks, Storage ==
−
   
+
 
−
* Deploy the node.js app
+
=== check disks on ubuntu ===
−
** deploy resource
+
 
−
** run npm, install dependencies
+
  sudo lsblk
−
** create & link config files
+
 
−
* Deploy the python app
+
=== Mount a block device ===
−
** deploy resource
+
 
−
** install dependencies
+
From: https://www.digitalocean.com/community/tutorials/how-to-use-block-storage-on-digitalocean
−
** create & link config files
+
 
−
* Configure the proxy
+
  parted /dev/sda mklabel gpt
−
** create & link config files
+
  parted -a opt /dev/sda mkpart primary ext4 0% 100%
 +
  mkfs.ext4 /dev/sda
 +
  mkdir -p /data
 +
  '/dev/sda /data ext4 defaults,nofail,discard 0 2' | sudo tee -a /etc/fstab
 +
  mount -a
 +
 
 +
=== Find large files ===
 +
  find / -xdev -type f -size +100M
 +
 
 +
=== see timestamps in history ===
 +
echo 'export HISTTIMEFORMAT="%d/%m/%y %T "' >> ~/.bashrc ; source ~/.bashrc
 +
 
 +
=== Generate large random file ===
 +
dd if=/dev/urandom of=/var/log/mongodb/randomFile bs=1M count=200
 +
 
 +
 
 +
=== Create a user ===
 +
 
 +
su - <username>
 +
sudo visudo -f /etc/sudoers
 +
usernameusedforlogin ALL=(ALL) NOPASSWD:ALL
 +
sudo adduser --disabled-password <username>
 +
 
 +
add user to group:
 +
 
 +
usermod -a -G wheel niceguy
 +
useradd -g wheel niceguy ## only is user doesn't exist!
 +
 
 +
or in rhel:
 +
 
 +
passwd -f -u deploy
 +
(and there is deluser)
 +
(I don't think I need this: usermod -aG sudo username ) - for making him a sudoer
 +
 
 +
useradd? adduser?
 +
 
 +
adduser : add user with full profile and info (pass, quota, permission, etc.) adduser is friendlier in that it sets up the account's home folders and other settings (e.g. automatically loading system stats and notifications on login),
 +
 
 +
whereas useradd just creates the user
 +
 
 +
 
 +
 
 +
=== Delete files older than 5 days ===
 +
find /path/to/files* -mtime +5 -exec rm {} \;
 +
find /path/to/files* -type f -mtime +5 -exec rm {} -v \;
 +
 
 +
=== get children of a process ===
 +
pgrep -P $your_process1_pid
  
−
=== Reset Chef UI Password ===
+
=== restrict resources given to a process ===
−
sudo -u opscode-pgsql  /opt/chef-server/embedded/bin/psql opscode_chef
+
−
update osc_users set hashed_password = '$2a$12$y31Wno2MKiGXS3FSgVg5UunKG48gJz0pRV//RMy1osDxVbrb0On4W' , salt ='$2a$12$y31Wno2MKiGXS3FSgVg5Uu' where username ='admin';
+
  
−
and login with user 'admin' and password 'password'.
+
sudo apt-get install cgroup-bin -y
 +
sudo cgcreate -g memory:/rmGroup
 +
sudo
 +
  echo $(( 500 * 1024 * 1024 )) > /sys/fs/cgroup/memory/rmGroup/memory.limit_in_bytes
 +
  # echo $(( 5000 * 1024 * 1024 )) > /sys/fs/cgroup/memory/rmGroup/memory.memsw.limit_in_bytes # swap only
 +
# cgexec -g memory:rmGroup <command>
 +
cgexec -g memory:rmGroup find cache/ -type f -mtime +10 -exec rm {} -v \;

Latest revision as of 19:11, 29 September 2024

Utils

Replace all occurrences of a string in a folder

ubuntu? 

find /home/user/ -type f | xargs sed -i  's/a.example.com/b.example.com/g'

mac: 

 LC_ALL=C find . -type f -name "*.html" -exec sed -i '' "s/UA-47088821-3/UA-47088821-4/" {} +

last 3 accessed files

 find . -type f -exec stat -c '%X %n' {} \; | sort -nr | awk 'NR==1,NR==3 {print $2}'

last 3 modified files

  find . -type f -exec stat -c '%Y %n' {} \; | sort -nr | awk 'NR==1,NR==3 {print $2}'

Find 10 largest files in a folder

 du -a /var | sort -n -r | head -n 10
 du -a .    | sort -n -r | head -n 15

For mac: 

du -a . | sort -n -r | head

Limit how much logs journalctl keeps: 

 journalctl --vacuum-time=2d
 journalctl --vacuum-size=500M

Count Lines in a Folder

find . -name "*js" -print | grep -v node_modules | xargs cat | grep -v "^//" | grep -v "^$" | wc -l


cut after ag for shorter lines

ag <your-search-keyword> | cut -c-800

increase number of open files, file descriptors

From: /etc/security/limits.conf 

* - nofile 131072

*    soft nofile 64000
*    hard nofile 64000
root soft nofile 64000
root hard nofile 64000

Reload services after making changes, logout.

count open files

lsof | wc -

Per user: 

lsof | grep ' opera ' | awk '{print $NF}' | sort | wc -l

Set date

sudo date --set "25 Sep 2013 15:00:00"

Get date in milliseconds

date -d 2017-05-05 +%s


Disks, Storage

check disks on ubuntu

 sudo lsblk

Mount a block device

From: https://www.digitalocean.com/community/tutorials/how-to-use-block-storage-on-digitalocean 

 parted /dev/sda mklabel gpt
 parted -a opt /dev/sda mkpart primary ext4 0% 100% 
 mkfs.ext4 /dev/sda
 mkdir -p /data
 '/dev/sda /data ext4 defaults,nofail,discard 0 2' | sudo tee -a /etc/fstab
 mount -a

Find large files

find / -xdev -type f -size +100M

see timestamps in history

echo 'export HISTTIMEFORMAT="%d/%m/%y %T "' >> ~/.bashrc ; source ~/.bashrc

Generate large random file

dd if=/dev/urandom of=/var/log/mongodb/randomFile bs=1M count=200


Create a user

su - <username>
sudo visudo -f /etc/sudoers
usernameusedforlogin ALL=(ALL) NOPASSWD:ALL
sudo adduser --disabled-password <username>

add user to group: 

usermod -a -G wheel niceguy
useradd -g wheel niceguy ## only is user doesn't exist!

or in rhel: 

passwd -f -u deploy
(and there is deluser)
(I don't think I need this: usermod -aG sudo username ) - for making him a sudoer

useradd? adduser?

adduser : add user with full profile and info (pass, quota, permission, etc.) adduser is friendlier in that it sets up the account's home folders and other settings (e.g. automatically loading system stats and notifications on login),

whereas useradd just creates the user


Delete files older than 5 days

find /path/to/files* -mtime +5 -exec rm {} \;
find /path/to/files* -type f -mtime +5 -exec rm {} -v \;

get children of a process

pgrep -P $your_process1_pid

restrict resources given to a process

sudo apt-get install cgroup-bin -y
sudo cgcreate -g memory:/rmGroup
sudo
 echo $(( 500 * 1024 * 1024 )) > /sys/fs/cgroup/memory/rmGroup/memory.limit_in_bytes
 # echo $(( 5000 * 1024 * 1024 )) > /sys/fs/cgroup/memory/rmGroup/memory.memsw.limit_in_bytes # swap only
# cgexec -g memory:rmGroup <command>
cgexec -g memory:rmGroup find cache/ -type f -mtime +10 -exec rm {} -v \;
Retrieved from "https://wiki.wasya.co/index.php?title=Linux_utils&oldid=3152"