Skip to content
Blog

rConfig · rConfig Core · rConfig Pro

rConfig CVE-2026-124079: Unauthenticated Code Execution Through the Laravel Horizon Redis Queue

CVE-2026-124079 (9.8): a serialized job pushed into Redis gives code execution on rConfig Core before 8.2.19 and Pro before 8.4.1. Risks, detection, fixes.

4 min readCVE-2026-124079

CVE-2026-124079 was published in NVD on November 9, 2026 by VulnCheck (acting as CNA), rated 9.8. It affects rConfig, the network device configuration management tool, in both its Core and Pro editions.

What makes this one serious is what the tool does. rConfig logs into your routers, switches and firewalls to back up their configuration, which means it stores their credentials. Code execution on the rConfig server can open the door to every network device it manages.


The vulnerability

FieldValue
CVECVE-2026-124079
CVSS 3.1 (VulnCheck)9.8 (CRITICAL) — AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H
CVSS 4.0 (VulnCheck)9.3 (CRITICAL)
TypeDeserialization of untrusted data (CWE-502)
AuthenticationNone if Redis is reachable
NVD publication2026-11-09
NVD statusDeferred (no NIST analysis, no CPE configuration)
CISA KEVNot listed at the time of writing

How the attack works

rConfig relies on Laravel Horizon for background jobs, which live in a Redis queue. According to the NVD description, in affected deployments that Redis queue runs without authentication. The chain:

  1. The attacker pushes a crafted serialized job onto the Redis queue.
  2. Laravel's CallQueuedHandler unserializes it with no class restrictions.
  3. The job instantiates an existing application class, App\Jobs\TaskDownloadRun, which runs an Artisan command the attacker picks.
  4. Through the laravel/tinker package bundled with rConfig, that command runs arbitrary PHP and operating system commands as the queue worker's user.

The consequences the NVD description lists:

  • a persistent web shell written to storage/app/public, which the web server serves;
  • database credentials read;
  • stored device credentials read;
  • API tokens read.

What if Redis isn't exposed?

The description covers that case too. An authenticated rConfig user can deliver the payload through the application itself by configuring a device target pointing at the rConfig server's loopback address. rConfig then connects to its own local Redis. Closing the Redis port narrows the risk to logged-in users, but doesn't remove it.


Affected versions

EditionAffectedFirst version outside the range
rConfig Core8.0.0 up to, not including, 8.2.198.2.19
rConfig Probefore 8.4.18.4.1

These bounds come from the VulnCheck description. NVD references the core-8.2.19 release on rConfig's GitHub repository and the vendor's security advisories page. Check there for the recommended version for your edition, ideally the latest release.


Detection

  • Reachable Redis: from another host, test whether the rConfig server's Redis port answers, and whether it accepts commands without a password.
  • Unexpected files in storage/app/public, especially .php files: that's the web shell location the description names.
  • Unusual Horizon jobs: TaskDownloadRun executions that don't match any scheduled task.
  • Device entries whose address is 127.0.0.1, localhost or the rConfig server's own address.
  • Child processes of the queue worker (shells, curl, wget) and outbound connections from the server.

Mitigation

1. Upgrade

rConfig Core to 8.2.19 or later, rConfig Pro to 8.4.1 or later.

2. Lock down Redis

Redis should listen on the local interface only and require a password. There's no reason for it to be reachable from the network.

3. Restrict access to rConfig

Only the network team should reach the interface, from a management network. That also limits the authenticated loopback route.

4. If exploitation was possible

An upgrade won't remove a web shell that's already in place. If the server was exposed:

  1. find and remove unknown files in storage/app/public;
  2. rotate every device credential stored in rConfig. That's the priority, since those credentials open the network itself;
  3. change the database password and revoke API tokens;
  4. check your network devices for unexpected logins from service accounts.

Why continuous monitoring matters for network management tools

Tools that manage the network hold the secrets to every other device. A single flaw on the rConfig server can be worth access to every router and firewall it backs up.

With cveo.tech, add those tools to your inventory with their exact version and get an automatic alert as soon as a CVE that affects them is published.

Know which CVEs hit the versions you run.

Add the product and its version: cveo.tech lists the CVEs that affect it right away, then alerts you by email when a new one affects it — including before NVD has finished its analysis. Free, an account is all it takes.

Every Monday

The week's critical CVEs, in your inbox

One email a week: the CVSS ≥ 9 vulnerabilities published in the last seven days, plus our latest analyses. Nothing else.

Double opt-in by email. Unsubscribe in one click, any time.