Severity by source
AV:L/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:H
Attacker must already control the local nagios account (AV:L, PR:L) and injection is trivially reliable (AC:L); escalation to root yields full C/I/A impact.
Primary rating from Vendor (https://github.com/Linuxfabrik/monitoring-plugins).
CVSS VectorVendor: https://github.com/Linuxfabrik/monitoring-plugins
Lifecycle Timeline
2DescriptionCVE.org
Summary
When a check plugin places user provided input inside a command which is passed to shell_exec, an attacker can abuse this to run arbitrary commands. This is mainly dangerous for plugins which are listed in the sudoers file, because this allows an attacker controlling the nagios user to get root privileges.
Details
An example for this is the restic-check plugin, where the --repo argument is placed inside the command argument of shell_exec. As an example, an attacker could use the --repo argument |touch /root/nagios-was-here|. The full restic command is assembled to the string restic --json --repo=|touch /root/nagios-was-here| --password-file= check before it is passed to shell_exec. shell_exec then splits the command up in three parts at the | boundaries and executes the parts separately, which also executes the embedded command touch /root/nagios-was-here.
PoC
This PoC shows how the nagios user can use this to create a file inside /root.
nagios@test-vm:/$ sudo /usr/lib64/nagios/plugins/restic-check --repo '|touch /root/nagios-was-here|'Impact
The vulnerability is a local privilege escalation.
Fix
Switch from | to an array
Remove the | split functionality. Instead, modify shell_exec to accept either a string or an array of strings. If an array is provided, the commands are chained together like they currently are when using |. If a string is provided, no split should be performed. You could also introduce a separate function like shell_exec_with_user_input() which implements this such that the current shell_exec function can stay like it is.
This leaves the problem that an attacker can still specify arbitrary arguments inside a command. An example for this would be to use the --repo argument sftp://example.com --cache-dir /tmp, which would lead to the execution of: restic --json --repo=sftp://example.com --cache-dir /tmp --password-file=None check. Please note that this example should mainly highlight the problem in general. To prevent the problem, there is either escaping or again array-syntax. Escaping would use shlex.quote to place the user provided argument inside quotes and which also escapes everything which needs to be escaped. Using array syntax would mean providing the full command as an array like ['restic', '--json', '--repo', 'sftp://example.com']. The array can then be given as-is to Popen. With this method, the proposed shell_exec_with_user_input would accept an array of array of strings.
Patches
The fix follows the array-syntax approach proposed above:
linuxfabrik-lib5.0.0:lib.shell.shell_exec()requires the command as a list of
arguments (argv) and always runs with shell=False. The | split functionality, command strings and the shell= parameter have been removed, so user-provided input can no longer break out of a command. lib.shell.safe_cli_value() additionally guards positional arguments (such as an ssh destination or a ping target) against option injection, and lib.ssh builds argument lists as well.
- Linuxfabrik Monitoring Plugins: all plugins assemble their external commands as argv lists
(commit 23bb570f4). Contained in every release after v5.2.0.
AnalysisAI
Local privilege escalation in Linuxfabrik Monitoring Plugins (and the underlying linuxfabrik-lib Python library) lets an attacker who already controls the low-privileged nagios account execute arbitrary OS commands, typically as root. The flaw stems from the library's shell_exec() helper splitting assembled command strings on the pipe (|) character, so user-controlled plugin arguments such as restic-check's --repo can inject additional commands. Publicly available exploit code exists (a PoC is included in the vendor advisory), but the issue is not listed in CISA KEV and no EPSS score was provided.
Technical ContextAI
The affected component is linuxfabrik-lib (PyPI package linuxfabrik-lib, pkg:pip/linuxfabrik-lib), a shared Python library used by the Linuxfabrik Monitoring Plugins suite for Nagios/Icinga-style checks. Its lib.shell.shell_exec() historically built a single shell command string and ran it via a shell, splitting the string on | to chain sub-commands. Because plugins interpolate user-supplied arguments directly into that string before execution, this is a classic CWE-78 (OS Command Injection) issue: the shell metacharacter | is interpreted rather than treated as literal data, letting an embedded command run as its own process. The restic-check plugin is the demonstrated example, where --repo=|touch /root/nagios-was-here| assembles to 'restic --json --repo=|touch /root/nagios-was-here| --password-file= check' and the pipe-delimited segment executes independently.
RemediationAI
Vendor-released patch: upgrade linuxfabrik-lib to 5.0.0 or later, where shell_exec() requires an argv list and always runs with shell=False, the | split logic and shell= parameter are removed, safe_cli_value() guards positional arguments against option injection, and lib.ssh builds argument lists as well. Correspondingly, update the Linuxfabrik Monitoring Plugins to a release after v5.2.0 (the argv-list conversion landed in commit 23bb570f4). As a compensating control until patched, remove or tightly scope any Linuxfabrik plugin entries in the sudoers file so the nagios user cannot invoke them as root - trade-off: checks that legitimately need elevated privileges (e.g., restic-check reading a protected repo) will stop working - and audit plugins for user-controlled arguments reaching shell_exec(). Refer to advisory GHSA-798h-hpph-m24j (https://github.com/Linuxfabrik/monitoring-plugins/security/advisories/GHSA-798h-hpph-m24j) for the authoritative fix guidance.
Same weakness CWE-78 – OS Command Injection
View allSame technique Privilege Escalation
View allShare
External POC / Exploit Code
Leaving vuln.today
EUVD-2026-61493
GHSA-798h-hpph-m24j