Bazarr 1.6.1 Security Response: Patch or Isolate, Rotate Keys, and Check for RCE
Quick Answer
Isolate affected Bazarr installations from untrusted networks, preserve evidence, and move to a reviewed current fixed release. The advisory lists versions through 1.6.1-beta.15 as affected and 1.6.1-beta.16 as patched; do not read that as a recommendation to install an old beta. Stable 1.6.1 includes the session fix, and 1.6.2 is the current stable release checked October 2, 2026. Verify the actual running build, assess exposed secrets and investigate the vulnerable period before reopening access.
Expose
Include reverse proxies, direct ports, VPN routes, container networks, and guest VLAN reachability.
Isolate
Block the service from the internet and untrusted LANs before investigating logs or changing credentials.
Patch
Use the vendor release notes and verify the running container or package after restart.
Rotate
Rotate Bazarr, Sonarr, Radarr, proxy, and automation secrets reachable from logs or backups.
Decision Matrix
| Area | What to Check | Operational Standard |
|---|---|---|
| Version | Running Bazarr build, image tag, image digest, and release channel | Upgrade to the fixed build available for your channel and verify it after restart. |
| Exposure | Public reverse proxy, direct port, VPN-only access, LAN-only access, guest VLAN reachability | Untrusted networks cannot reach Bazarr, even before authentication. |
| Secrets | Bazarr API key, Arr API keys, proxy credentials, webhook tokens | Rotate any secret exposed in logs, backups, or proxied requests. |
| Evidence | Access logs, app logs, config backups, container filesystem, reverse proxy history | Treat evidence gaps as uncertainty. Rebuild from known-good software when integrity or prior access cannot be bounded sufficiently for the risk. |
Define Exposure Before You Touch The Upgrade Button
Start by identifying every place Bazarr is reachable. The most important split is not public versus private in a vague sense; it is who can open the service, which reverse proxies can reach it, whether container bridge networks expose it to adjacent services, and whether backup or diagnostic paths contain secrets.
Capture the current version, image digest or package source, bind address, proxy route, authentication mode, and the identities that can administer it. That snapshot gives you evidence for the incident notes and a rollback reference if the remediation fails.
- Record the exact running version and compare it to the fixed floor: the patched release identified by the Bazarr advisory.
- List inbound paths: VPN, reverse proxy, direct LAN, container network, and any published port.
- Save a configuration backup, but treat it as sensitive because it may contain tokens, password hashes, API keys, or internal addresses.
- Stop public or guest-network access while the exposure decision is unresolved.
Patch, Isolate, Or Rebuild: Choose The Right Response
If there is no credible sign of access and the service was never reachable from untrusted networks, patching may be enough. If Bazarr was exposed to the internet, a guest network, a shared LAN, or a compromised host, assume the problem is wider than the package version.
The clean response is staged: isolate first, patch second, rotate secrets third, then validate from a network that should fail. A successful dashboard login after an upgrade is useful, but it is not proof that the old path is closed or that previous access did not happen.
- Patch immediately when a fixed build exists for your channel.
- Keep untrusted access isolated while patching or investigating; do not wait for the end of the day to contain an affected reachable service.
- Investigate prior reachability, downloadable backups, suspicious activity and exposed keys. Rebuild when tampering is found or trust cannot be established to the required level; exposure or debug logging alone is not proof of exploitation.
- Do not assume form authentication helped; the advisory specifically affects the authentication path.
Rotate What The Vulnerable Path Could Reach
Secret rotation should follow reachability, not wishful thinking. Rotate the application key itself, any upstream application keys it could proxy or read, reverse-proxy credentials, webhook tokens, automation tokens, and any password that appeared in clear text or recoverable form inside backups or logs.
Do not rotate only the account you personally use. In homelab stacks, service accounts are often more powerful than user accounts because they can read media libraries, update configuration, call APIs, or write post-processing commands.
- Invalidate active sessions where the application supports it.
- Revoke known-abused credentials promptly after containment, accepting an integration outage if needed. Issue replacements from a clean administration path and update dependent applications in a recorded order.
- Search proxy logs, app logs, diagnostic bundles, and backup archives for old tokens.
- Keep a dated rotation record so stale keys are not accidentally reintroduced from backups.
Rebuild When The Evidence Is Dirty
A clean rebuild can replace untrusted software and state, but it does not resolve missing history or revoke stolen secrets by itself. Rebuild from a known-clean image, restore only reviewed configuration, and reconnect dependencies after credentials have been replaced.
For containers, rebuilding means more than pulling the latest image. Recreate the container, verify bind mounts, compare persistent configuration against a known-good baseline, and keep unreviewed temporary files, scripts and backups out of the recovered environment. Preserve originals and custody records before any authorized cleanup; do not erase incident evidence.
Validation Checklist
- From an untrusted network, Bazarr is unreachable at the network layer or blocked by the reverse proxy before application authentication.
- From a trusted admin path, the dashboard reports the expected fixed version after restart.
- Old Bazarr and Arr API keys no longer work against the affected services.
- New logs and shared support bundles no longer expose active tokens. Keep historical evidence and backups restricted under the retention plan rather than rewriting them; verify that any credentials they contain have been revoked and will not be restored into service.
- A test restore of the reviewed configuration starts cleanly without unknown post-processing commands.
What This Does Not Prove
This article is documentation-backed and response-oriented. It does not prove that your individual host was exploited, that your logs are complete, or that a patched container removed every malicious file if an attacker already had command execution.
Security, Privacy, Legal, And Recovery Boundaries
- Do not publish exploit commands, stolen backup names, real API keys, private hostnames, or internal addresses.
- Keep Bazarr and the Arr applications behind private access unless there is a documented reason to expose them.
- Use lawful media workflows only; this article is about securing subtitle automation for libraries you are authorized to manage.
- Before reconnecting a rebuilt service, rotate credentials and verify dependent integrations with least privilege.
Advisory Scope
The advisory describes form-authentication bypass with log disclosure and server-side request access. The backup-to-command-execution chain additionally requires an existing backup and identification of its name to obtain the API key. No CVE was assigned in the advisory metadata checked October 2, 2026. Preserve its revision and verify the actual package or container build against the current release notes.
Related TechGeeks Resources
- Reverse Proxy Setup Without Security Mistakes
- Container Security for Normal Homelabs
- How to Audit Your Exposed Services


