Table des matières
- 2026:
- 2025:
13 billet(s) pour novembre 2026
| Réseaux - protocole QUIC mieux que TCP | 2026/11/17 21:26 | Jean-Baptiste |
| Notes conversion JSON YAML | 2026/11/13 11:16 | Jean-Baptiste |
| Git commit perdu | 2026/11/12 18:16 | Jean-Baptiste |
| Notes mojo lang | 2026/11/11 20:34 | Jean-Baptiste |
| Notes rust lang compilation | 2026/11/11 18:35 | Jean-Baptiste |
| Mémoire VSIZE VMSIZE SEGV lang C | 2026/11/11 15:59 | Jean-Baptiste |
| Ansible-navigator utilisation de ansible-vault | 2026/11/06 11:33 | Jean-Baptiste |
| Err Ansible - Module result deserialization failed No start of json char found | 2026/11/06 09:09 | Jean-Baptiste |
| Script Python conversion JSON vers CSV | 2026/11/06 08:14 | Jean-Baptiste |
| Filter Incoming IPv4 Packets Containing IP Options - cve-2026-98432 mitigation - Kernel TCP Use-After-Free | 2026/11/04 21:04 | Jean-Baptiste |
| Notes ZFS | 2026/11/03 18:04 | Jean-Baptiste |
| Diff image container | 2026/11/03 16:30 | Jean-Baptiste |
| Notes processus multi-thread | 2026/11/03 10:59 | Jean-Baptiste |
Proxy et apt-get
Pour définir une conf spécifique à apt-get (Adaptez et dé-commenter les lignes voulu)
/etc/apt/apt.conf.d/02proxy
#Acquire::http::Proxy "false"; #Acquire::ftp::Proxy "false"; #Acquire::http::Proxy "http://10.9.3.100:3128"; #Acquire::http::Proxy "http://utilisateur:motdepasse@example.com:port";
Mais nous préférerons à la place définir la conf avec les variables d’environnement
/etc/environment
# http_proxy=http://10.9.3.100:3128 # https_proxy=http://10.9.3.100:3128 ALL_PROXY=http://10.9.3.100:3128 NO_PROXY="localhost,127.0.0.1,.lan"
Notes : /etc/environment s'applique à tous les utilisateurs, à la place on peut utiliser le bashrc.
Où
http_proxy=http://10.9.3.100:3128 apt-get update && apt-get upgrade -y
Protéger l’accès à vos applications web grâce à un mot de passe à usage unique OTP
Pour se faire nous utiliserons :
- Nginx en mode reverse proxy
Un logiciel pour vous identifier en générant un mot de passe jetable. Sous Android :
- andOTP
- flauth
- mOTP
- SecureOTP
apt-get install python3-pip adduser --group --system --disabled-password --home /opt/auth-otp app-otp su - app-otp -s /bin/bash python3 -m pip install --user pyotp git clone https://github.com/newhouseb/simpleotp cd simpleotp
main.py
-PORT = 8000 +PORT = 8001 -httpd = socketserver.TCPServer(("", PORT), AuthHandler) +httpd = socketserver.TCPServer(("127.0.0.1", PORT), AuthHandler)
Après il faut lancer le script Python dans un tmux ou avec Systemd ou encore supervisord.
python3 main.py
python3 -c "import pyotp; print(pyotp.random_base32())" > .totp_secret chmod 600 .totp_secret
Cong Nginx
server { listen 443 ssl; listen [::]:443 ssl; server_name cockpit.acme.fr; location /auth { proxy_pass http://127.0.0.1:8001; # This is the TOTP Server proxy_set_header X-Original-URI $request_uri; } # This ensures that if the TOTP server returns 401 we redirect to login error_page 401 = @error401; location @error401 { return 302 /auth/login; } location / { auth_request /auth/check; # Required to proxy the connection to Cockpit proxy_pass https://127.0.0.1:9090; proxy_set_header Host $host; proxy_set_header X-Forwarded-Proto $scheme; # Required for web sockets to function proxy_http_version 1.1; proxy_buffering off; proxy_set_header Upgrade $http_upgrade; proxy_set_header Connection "upgrade"; # Pass ETag header from Cockpit to clients. # See: https://github.com/cockpit-project/cockpit/issues/5239 gzip off; } ssl_certificate /etc/letsencrypt/live/cockpit.acme.fr/fullchain.pem; # managed by Certbot ssl_certificate_key /etc/letsencrypt/live/cockpit.acme.fr/privkey.pem; # managed by Certbot } server { if ($host = cockpit.acme.fr) { return 301 https://$host$request_uri; } # managed by Certbot listen 80; server_name cockpit.acme.fr; return 404; # managed by Certbot }
Conf andOTP
Créer un “details” de type “TOTP”, dans “secret”, saisissez le contenu du fichier .totp_secret
Service SystemD
/etc/systemd/system/simple-otp.service
[Unit] Description=Simple OTP Wants=network.target After=network.target nginx.service [Service] WorkingDirectory=/opt/auth-otp/simpleotp ExecStart=/usr/bin/python3 /opt/auth-otp/simpleotp/main.py User=app-otp Restart=always [Install] WantedBy=multi-user.target
systemctl daemon-reload
systemctl start simple-otp
systemctl status simple-otp
systemctl enable simple-otp
Programmation
Voir :
Lisp / Programmation Fonctionnelle :
La fin de la programmation ? https://www.lemonde.fr/blog/internetactu/2016/05/21/demain-la-fin-du-code/ (début de l'article intéressant)
Les Processus sous Linux
Voir aussi :
/proc/sys/kernel/ns_last_pid
Tuer tous les processus d'un utilisateur
pkill -u plop
Appels système au noyaux et signaux
kill -l man 7 signal
strace -p $PID strace -f -e trace=file $CHEMIN_EXE 2>&1 | grep '/home/'
La table des processus
ps -ef | grep $PID # ps -Aww -o pid,ppid,user,start,state,pri,pcpu,time:12,pmem,rsz:10,vsz:10,cmd
2em colone ⇒ PID 3em colone ⇒ PPID
Qui a lancé ce processus ? PPID ? Service ?
journalctl _PID=2636 -o json-pretty
Threads
# Classer par nombre de threads ps -eL --no-headers -o nlwp,pid,lwp,comm,command | sort -u # Avoir des information sur un thread cat /proc/5990/task/5990/comm
Que fait TTY1 ?
peekfd -cn8 $(ps -t tty1 | grep -v login | sort -nr | head -1 | awk '{print $1}')
ProcFS
Voir egalement /proc/$PID/ Voir man proc ou man procfs
Cacher /proc et la table des processus aux utilusateurs :
Avec Grsecurity Ou en en remontant /proc linux-hide-processes-from-other-users hide-process-information-for-other-users hidepid
Modifier la ligne /proc dans /etc/fstab. Si aucune ligne ne correspond, ajouter la fstab
proc /proc proc defaults,hidepid=2,gid=4 0 0
Pour appliquer la modif sans redemarrer :
mount -o remount /proc/
Audit process
Voir execsnoop et autres : using-selected-bcc-tools-for-performance-analyses_analyzing-system-performance-with-bpf-compiler_collection paquet bcc-tools
Arrière plan
Using the Job Control of bash to send the process into the background:
- Ctrl+Z to stop (pause) the program and get back to the shell.
- bg to run it in the background.
- disown -h [job-spec] where [job-spec] is the job number (like %1 for the first running job; find about your number with the jobs command) so that the job isn't killed when the terminal closes.
Source : how-do-i-put-an-already-running-process-under-nohup
info coreutils 'nohup invocation'
Voir :
- nohup $EXE
- $EXE &
- ctrl + z
- bg
- fg
- pstree
- ps
- setsid
- runit / shpst
- trap
# kill -l 18) SIGCONT 19) SIGSTOP
Exemple
Exemple
# jobs [1]+ Stoppé rsync -azS --partial /data/1/ $FIC root@192.168.205.2:/data/files/1/ [2]- En cours d'exécution rsync -azS --partial /data/2/ $FIC root@192.168.205.2:/data/files/2/ & jobs %% jobs %1
Stopper le job1
kill -19 %1
Lancer le job2 une fois le job1 terminé
wait %2; kill -18 %1
http://linuxfr.org/users/srb/journaux/waitend-executer-une-commande-apres-une-autre-deja-lancee
sleep $(( 60 * 3 )) ; echo "Les œufs sont prêt"
sleep $(( 60 * 3 )) && echo "Les œufs sont prêt"
sleep $(( 60 * 3 )) & wait $! echo "Les œufs sont prêt"
sleep $(( 60 * 3 ))
[Ctrl] + [Z]
fg && echo "Les œufs sont prêt"
Autres
Voir atop, htop Voir cgroup
kthreadd
Voir https://www.it-connect.fr/les-processus-sous-linux
# ps -ef | head -40 UID PID PPID C STIME TTY TIME CMD root 1 0 0 Mar15 ? 00:02:12 /sbin/init root 2 0 0 Mar15 ? 00:00:00 [kthreadd] root 3 2 0 Mar15 ? 19:02:47 [migration/0] root 4 2 0 Mar15 ? 00:07:26 [ksoftirqd/0] root 5 2 0 Mar15 ? 00:00:00 [stopper/0] root 6 2 0 Mar15 ? 00:07:33 [watchdog/0] root 7 2 0 Mar15 ? 18:46:13 [migration/1] root 8 2 0 Mar15 ? 00:00:00 [stopper/1] root 9 2 0 Mar15 ? 00:07:05 [ksoftirqd/1] root 10 2 0 Mar15 ? 00:06:16 [watchdog/1] root 11 2 0 Mar15 ? 18:55:58 [migration/2] root 12 2 0 Mar15 ? 00:00:00 [stopper/2] root 13 2 0 Mar15 ? 00:06:45 [ksoftirqd/2] root 14 2 0 Mar15 ? 00:06:04 [watchdog/2] root 15 2 0 Mar15 ? 19:15:39 [migration/3] root 16 2 0 Mar15 ? 00:00:00 [stopper/3] root 17 2 0 Mar15 ? 00:06:43 [ksoftirqd/3] root 18 2 0 Mar15 ? 00:05:41 [watchdog/3] root 19 2 0 Mar15 ? 18:54:50 [migration/4] root 20 2 0 Mar15 ? 00:00:00 [stopper/4] root 21 2 0 Mar15 ? 00:06:48 [ksoftirqd/4] root 22 2 0 Mar15 ? 00:06:09 [watchdog/4] root 23 2 0 Mar15 ? 18:51:16 [migration/5] root 24 2 0 Mar15 ? 00:00:00 [stopper/5] root 25 2 0 Mar15 ? 00:06:06 [ksoftirqd/5] root 26 2 0 Mar15 ? 00:05:27 [watchdog/5] root 27 2 0 Mar15 ? 18:59:33 [migration/6] root 28 2 0 Mar15 ? 00:00:00 [stopper/6] root 29 2 0 Mar15 ? 00:06:27 [ksoftirqd/6] root 30 2 0 Mar15 ? 00:05:13 [watchdog/6] root 31 2 0 Mar15 ? 19:15:16 [migration/7] root 32 2 0 Mar15 ? 00:00:00 [stopper/7] root 33 2 0 Mar15 ? 00:06:16 [ksoftirqd/7] root 34 2 0 Mar15 ? 00:06:33 [watchdog/7]
Nous pouvons constater immédiatement quatre choses :
- Les processus '[migration/n]', contrairement aux processus classiques, on leurs noms entre des crochets []
- Le process père est [kthreadd] (PPID 2)
- Nous avons 8 processus : migration/0..migration/7. Soit un par core.
- Ces processus sont démarrés depuis le démarrage du système d'exploitation soit le 15 mars.
$ LANG=C man ps "Sometimes the process args will be unavailable; when this happens, ps will instead print the executable name in brackets"
Les zombies
Tuer les zombies !
Source : https://www.linuxjournal.com/content/how-kill-zombie-processes-linux
Also known as “defunct” or “dead” process – In simple words, a Zombie process is one that is dead but is present in the system’s process table. Ideally, it should have been cleaned from the process table once it completed its job/execution but for some reason, its parent process didn’t clean it up properly after the execution.
In a just (Linux) world, a process notifies its parent process once it has completed its execution and has exited. Then the parent process would remove the process from process table. At this step, if the parent process is unable to read the process status from its child (the completed process), it won’t be able to remove the process from memory and thus the process being dead still continues to exist in the process table – hence, called a Zombie!
In order to kill a Zombie process, we need to identify it first. The following command can be used to find zombie processes:
ps aux | egrep "Z|defunct"
Z in the STAT column and/or [defunct] in the last (COMMAND) column of the output would identify a Zombie process.
Now practically you can’t kill a Zombie because it is already dead! What can be done is to notify its parent process explicitly so that it can retry to read the child (dead) process’s status and eventually clean them from the process table. This can be done by sending a SIGCHLD signal to the parent process. The following command can be used to find the parent process ID (PID):
ps -o ppid= <Child PID>
Once you have the Zombie’s parent process ID, you can use the following command to send a SIGCHLD signal to the parent process:
kill -s SIGCHLD <Parent PID>
However, if this does not help clearing out the Zombie process, you will have to kill or restart its parent process OR in case of a huge surge in Zombie processes causing or heading towards system outage, you will have no choice but to go for a system reboot. The following command can be used to kill its parent process:
kill -9 <Parent PID>
Note that killing a parent process will affect all of its child processes, so a quick double check will be helpful to be safe. Alternatively, if few lying zombie processes are not consuming much CPU/Memory, it’s better to kill the parent process or reboot the system in the next scheduled system maintenance.
Strace
Voir : https://aurelien-riv.github.io/php/2019/12/28/what-process-waiting-for.html
Exemple :
strace -e trace=network -f iceweasel http://acme.com 2>&1 | grep '8.8.8.8' strace -ttfv -s128 -o /var/tmp/ping-$(hostname)-$(date +%Y.%m.%d-%H.%M).out ping -c1 acme.com strace -f -p $(pidof mariadbd) --trace=sendto,recvfrom -s 2048 2>&1 | grep -i update vimdiff <(cat plop.log | sed -e 's/^\[pid [0-9]\+\] //') <(cat plop.err | sed -e 's/^\[pid [0-9]\+\] //')
Process audit quand strace n'est pas là
Voir :
- /proc/[PID]/
- procmap
lsof -p 33279 ps -efL | grep plop |wc -l ss -tnp | grep plop tcpdump -A -s0 host 192.168.100.5 and port 21 dmesg pmap 33279 strace -r -o trace.log sudo echo hi
Pour voir les E/S sur les FD
peekfd -n -8 -d -c 32167 0 1 2
Notes strace
A debugging utility for Linux, monitors system calls used by a program and all received signals
Voir aussi :
- eBPF – Linux kernel tracing backend providing a set of features similar to DTrace[28] since kernel version 4.9
- ftrace – a tracing framework for the Linux kernel, capable of tracing scheduling events, interrupts, memory-mapped I/O, CPU power state transitions, etc.
- ltrace – a Linux debugging utility, displays the calls a userland application makes to shared libraries
- SystemTap – a scripting language and utility used for instrumenting Linux installations
-s strsize Specifies the maximum string size to print (the default is 32). Note that filenames are not considered strings and are always printed in full.
Activer / interdire strace
kernel.yama.ptrace_scope = 0: all processes can be debugged, as long as they have same uid. This is the classical way of how ptracing worked.
kernel.yama.ptrace_scope = 1: only a parent process can be debugged.
kernel.yama.ptrace_scope = 2: Only admin can use ptrace, as it required CAP_SYS_PTRACE capability.
kernel.yama.ptrace_scope = 3: No processes may be traced with ptrace. Once set, a reboot is needed to enable ptracing again.
# LANG=C strace -p 1171 strace: attach: ptrace(PTRACE_ATTACH, 1171): Operation not permitted # echo 0 > /proc/sys/kernel/yama/ptrace_scope # LANG=C strace -p 1171
strace -T -tt -f -v -q -s 4096 -o /tmp/nonworking_cp-strace.out <cp command> &> /tmp/nonworking_cp.out strace -fvttTyyx -s 1024 -o /tmp/strace.out <command> &> /tmp/df.out
Autre
Auditer les signaux / les kills
Comparaison entre différents environnements
Voir procenv
