Day-to-day admin

The jobs that come up while DRSM is running: hosts in and out, thresholds, notices for users, upgrades and finding out why something is red. Everything is plain text files and a few scripts, so each job is also easy to hand to an AI assistant.

Commands below use ~/drsm and ~/.drsm.sh. If your install uses DRSM_CONFIG, put DRSM_CONFIG=<your config> in front of each command. Scripts that change files show a dry run first and need --apply.

Every day

Open the Overview: the banner and Needs attention list what is wrong, worst first.
Check that reports are current: a host shown as Stale (purple) means its checks did not run. Look at cron and $LOGdir/drsm_run*.log.
Alerts go to the addresses in alert.email.list or the site's ALERT_EMAIL, see Thresholds and alerts.

Add a host

Linux host from a host file

$ cp ~/drsm/etc/hosts/example.conf_example ~/drsm/etc/hosts/web2.conf
$ vi ~/drsm/etc/hosts/web2.conf          # HOST, DESCRIPTION, IMPACT, GROUP
$ ~/drsm/bin/drsm_run.sh -n -H web2 -v   # one run, no alerts

The monitor account's SSH key must be in the host's authorized_keys.

Linux host from Ansible

With ANSIBLE_INVENTORY (or Ansible Automation Platform, ANSIBLE_AAP_URL) set, add the host to a group under one of the ANSIBLE_TIERS groups in Ansible. DRSM checks it from the next run. utils/drsm_ansible_hosts.sh lists what DRSM takes from Ansible. See Ansible inventory.

Windows host

Install the reporting agent. When its first report arrives, utils/wra_hosts.sh (dry run, then --apply) makes its host file. See Windows hosts.

Network device or address

A host file with COLLECTOR=none, PLUGINS="" and PROBES="network/ping" (add network/https for a web service).

Remove a host

$ ~/drsm/utils/drsm_remove_host.sh web2              # dry run
$ ~/drsm/utils/drsm_remove_host.sh --apply web2
$ ~/drsm/utils/drsm_remove_host.sh --apply -f retired.txt   # one host per line

This removes the host file (a copy goes to etc/hosts/removed) and its reports. A host that comes from Ansible is removed from Ansible; with ANSIBLE_EXCLUSIVE=yes its checks stop at once, and drsm_remove_host.sh clears its old reports.

Change a threshold

Every plugin setting has a default in plugins/<category>/<name>/plugin.conf. Override it for all hosts in etc/hosts/_defaults.conf, for a site in its site file, or for one host in its host file, as <category>_<name>_<SETTING>=value:

storage_disk_WARN=85
storage_disk_SKIP="/mnt/scratch"
core_cpu_TIMEOUT=60

Status messages

Tell everyone about an outage or maintenance: the message scrolls across the top of every page. See Status messages.

$ ~/drsm/utils/drsm_message.sh --apply add warning "Mail relay maintenance 18:00 to 19:00" --until "2026-11-03 19:00"

Your own scripts

A new check on a host: write a plugin in plugins/local, see Writing a plugin.
A script that already writes a report file: list it in $EXTreports to show it on the dashboard with a state, see Reports from your own scripts. Start status lines with ERROR, WARNING or INFO and write a REPORT_ETIME: line with the run time.
Dashboard settings and menus are in $CONFIGdir/drsm.php, colors in $WWWdir/site/custom.css, see Customizing the dashboard.

Upgrade

Unpack the new release beside the old source.
Run utils/install.sh from it as the DRSM user. Host files, site files, plugins/local, _defaults.conf, config/drsm.php (the new one is written as drsm.php_install_new) and $WWWdir/site are kept.
Compare drsm.php_install_new with drsm.php for new settings, read What's new.
drsm_run.sh -l, then one run with -n.

Troubleshooting

Symptom	Look at
Host is Stale	Cron, `$LOGdir/drsm_run*.log`, `drsm_run.sh -n -H <host> -v`
Host is red	Its report page: the first ERROR line under its check
A check is Unknown	A command in the plugin's `REQUIRES` is missing on the host
Host waits for its first report	It is new, or its report folder is left from a removed host (`drsm_remove_host.sh`)
Page blank or wrong	The Web server error log, `php -l` on the page
No alert email	`-n` on the cron line, the address lists, mail relay