Do the daily work on Linux servers

$ run as a normal user# run as root (or with sudo)

Skills in this chapter

Daily server work — 6

systemctl — services

systemctl starts, stops, enables and checks services (units) of systemd.

#systemctl status nginx
#systemctl enable --now node_exporter
#systemctl restart shop && systemctl is-active shop
$systemctl --failed

11.2

journalctl — logs of the system

journalctl reads the logs of all services in one place.

#journalctl -u shop -f
#journalctl -u shop --since "10 min ago" -p err
#journalctl -b -1 -p warning

11.3

ssh — log in and run commands

ssh is how you reach every server — with keys, a jump host and an ssh config file.

$ssh -J bastion.example.com web1.internal
$ssh web1 'sudo systemctl is-active nginx'
$ssh-keygen -t ed25519 -C "hosam@laptop"

Good practice:

Keys only, no root login, no passwords on internet-facing servers.

11.4

rsync — copy and sync files

rsync copies only what changed — for deploys of static files, backups and migrations.

$rsync -avz --delete dist/ web1:/var/www/shop/
$rsync -aHAX --info=progress2 /srv/data/ backup1:/backups/data/
$rsync -avn --delete dist/ web1:/var/www/shop/

11.5

crontab — scheduled jobs

crontab runs scripts on a schedule (backups, clean-ups, reports). systemd timers are the modern choice.

$crontab -l
$crontab -e
$systemctl list-timers

11.6

logrotate — keep log files small

logrotate rotates, compresses and deletes old log files so disks do not fill up.

#logrotate -d /etc/logrotate.d/shop
#logrotate -f /etc/logrotate.d/shop

11.7

Write a runbook for every server role

A short page per server role saves hours at 3 in the night.

What runs here, which ports, which config files.
How to check health (systemctl status, a curl URL, a dashboard link).
Where the logs are (journalctl -u …, Loki labels).
How to restart safely, and who must know.
How backups work and when the last restore test was.

Remember

Every manual fix you do twice belongs in Ansible, a script or the runbook.

All commands in this chapter