Skip to content

Latest commit

 

History

History
250 lines (185 loc) · 6.53 KB

linux.md

File metadata and controls

250 lines (185 loc) · 6.53 KB
Raw

LINUX

Tips:

Pipe magic

cat data_access.log 2>&1 | grep key == cat data_access.log |& grep key cat data_access.log | grep key == cat data_access.log > >(grep key)

Feature comparison of ack, ag, git-grep, GNU grep and ripgrep

https://beyondgrep.com/feature-comparison/

BI with linux tools:

grep XXXX data_access.log | sort | uniq -c | sort -h
cut -d " " -f 2 data_access.log | grep XXXX | sort | uniq -c | sort -h | sed 's/\s\+/ /g'
grep -o "start.*end" data_access.log | sort | uniq

https://www.thegeekstuff.com/2013/06/cut-command-examples https://github.com/dbohdan/structured-text-tools https://www.datascienceatthecommandline.com

Execute xargs in parallel

find src -type f -name "*.php" -print0 | xargs -0 -n1 -P8 php -l
printf %s\\n {0..99} | xargs -n 1 -P 8 script-to-run.sh input/ output/
cat commands.sh | xargs -P4  -I "{}" sh -c "{}"

Note: See also parallel. https://stackoverflow.com/questions/28357997/running-programs-in-parallel-using-xargs https://mike42.me/blog/how-to-use-parallel-to-speed-up-your-work

Trace the Execution

#Trace the Execution of an Executable
strace ls
#Trace a Specific System Calls in an Executable Using Option -e
strace -e trace=open,read ls /home
#Save the Trace Execution to a File Using Option -o
strace -o output.txt ls
#Execute Strace on a Running Linux Process Using Option -p
strace -p 1725 -o firefox_trace.txt
#Print Timestamp for Each Trace Output Line Using Option -t
strace -t -e open ls /home
#Print Relative Time for System Calls Using Option -r
strace -r ls
#Generate Statistics Report of System Calls Using Option -c
strace -c ls /home
#Resolve every fd into its name, and every socket fd into its address line
strace -yy wget https://cdn0.bodas.net/usuarios/fotos/7/1/9/0/sfxb_376425.jpg

http://www.thegeekstuff.com/2011/11/strace-examples/

Show info about a exec

#Shared library dependencies
ldd /bin/ls
objdump -p /bin/ls | grep NEEDED # or grep DLL for Windows libs

#Disassemble
objdump -d /bin/ls or nm /bin/ls or readelf -a /bin/ls

#Debug (man ld.so)
LD_DEBUG=libs /bin/ls # or ltrace

#Print the strings of printable characters in files
strings /bin/ls

https://www.cs.swarthmore.edu/~newhall/unixhelp/compilecycle.html

Convert addresses into file names and line numbers with addr2line

php-fpm[6048]: segfault at 10 ip 00007f46db77a8fb sp 00007fffa155e2d0 error 4 in xcache.so[7f46db763000+23000]

# 00007f46db77a8fb - 7f46db763000 = 178FB
addr2line -e /usr/lib64/20131226/xcache.so 178FB
#out /root/source/xcache-3.2.0/mod_cacher/xc_cacher.c:778

https://ablagoev.github.io/linux/adventures/commands/2017/02/19/adventures-in-usr-bin.html#addr2line

Exec as root without sudo

pkexec emacs /etc/sudoers

http://manpages.ubuntu.com/manpages/precise/en/man1/pkexec.1.html

Flock for Cron jobs (and trap)

flock -n /tmp/path.to.lockfile -c command with args

http://www.elevatedcode.com/2013/05/07/flock-for-cron-jobs.html http://www.computerhope.com/unix/utrap.htm Other in GO croncape jobber

Xephyr: Pseudo-X server

Xephyr -ac -screen 1280x1024 -br -reset -terminate 2> /dev/null :3 &
DISPLAY=:3.0 gnome-wm &
DISPLAY=:3.0 ssh -X jennie@desktop xterm

http://jeffskinnerbox.me/posts/2014/Apr/29/howto-using-xephyr-to-create-a-new-display-in-a-window/

VNC server

sudo apt-get -y install x11vnc
x11vnc -display :0

Ubuntu: Recompile package from source

apt-get install devscripts

apt-get source XXXXX
cd XXXXXX
sudo apt-get build-dep XXXXXX
dpkg-buildpackage -b -uc

Debian: generate deb package.

Simple: use Effing package management (fpm)

fpm -s dir -t deb \
  --deb-no-default-config-files \
  --config-files "etc/galicasterpro/conf.ini" \
  -d python \
  -d python-pip \
  -d "package > 1.0" \
  --after-install $OLD_PWD"/postinst" \
  --before-install $OLD_PWD"/preinst" \
  --url "http://rubenrua.es" \
  --vendor "My Company" \
  -m "rubenrua <[email protected]>" \
  --description "My descriptn" \
  -n name -v $VERSION_NUMBER .

SSH: Avoid disconnection timeout

cat ~/.ssh/config
Host *
    ServerAliveInterval 120
    ServerAliveCountMax 3

Intercept stdout and stderr of a process

Since I'm not allowed to edit Jauco's answer, I'll give the full answer that worked for me (Russell's page relies on unguaranteed behaviour that, if you close fd 1 for stdout, the next creat call will open fd 1.

So, run a simple endless script like this:

import time while True: print 'test' time.sleep(1)

Save it to test.py, run with

python test.py
#Get the pid:

ps auxw | grep test.py
#Now, attach gdb:

gdb -p (pid)
#and do the fd magic:

(gdb) call creat("/tmp/stdout", 0600)
$1 = 3
(gdb) call dup2(3, 1)
$2 = 1

Now you can tail /tmp/stdout and see the output that used to go to stdout.

http://stackoverflow.com/questions/249703/how-can-a-process-intercept-stdout-and-stderr-of-another-process-on-linux

HTTP Server easy in python

$ python3 -m http.server.

https://gist.github.com/willurd/5720255

Share or access to local ports

socat or ssh -L MY_PORT:HOST_IN_LOCAL_NET:LOCAL_PORT [email protected]

Example:

socat  -v TCP-LISTEN:80,fork,reuseaddr TCP:172.0.0.1:8080
ssh -L 2222:192.168.1.38:22 -L 8080:192.168.1.38:80 [email protected]
ssh -p 2222 localhost && wget localhost:8080

~C (to open the ssh command prompt) and type a command (e.g.) “-L 80:localhost:8000”

See: http://www.vicente-navarro.com/blog/2009/05/24/creando-tuneles-tcpip-port-forwarding-con-ssh-los-8-escenarios-posibles-usando-openssh/

See also: https://ngrok.com, https://github.com/agrinman/wormhole, https://tmate.io/ and socks

supervisord with systemd unit

[Unit]
Description=Laravel queues

[Service]
User=www-user
Group=www-user
Type=simple
Restart=on-failure
RestartSec=30
Nice=10
WorkingDirectory=/www/laravel/
ExecStart=/usr/bin/php artisan queue:work --queue=default --sleep=30 --timeout=1800 --tries=1 --memory=256
StandardOutput=null
TimeoutStartSec=30

[Install]
RequiredBy=multi-user.target

LINKS: