IP Camera and NVR Cybersecurity Checklist

IP camera and NVR cybersecurity checklist: passwords, firmware, services, HTTPS, segmentation, external exposure testing, permissions, logs and order of implementation.
Technical summary
Technology
IP cameras, NVR, VMS, TLS, VLAN
Topic
Cybersecurity
Level
Intermediate
In this article7 sections

A single NVR running with a default password, combined with a port forward forgotten on the router, turns the camera system into an open door for an attacker. In most organizations, IP cameras, NVRs and VMS servers are the largest group of devices; once installed, they run untouched for years. Vulnerabilities affecting cameras and recorders regularly appear in lists of known exploited vulnerabilities. The list below follows the order we use on site to check the security hygiene of a video surveillance system across four layers. Besides device settings, decide which networks can access NVRs and cameras in a corporate network security and segmentation project.

1. Device level

A common picture: the installer gave all cameras the same password, the password sits in a spreadsheet, and the firmware is still the version from installation day. Do the checks device by device. An answer of “usually, yes” counts as a finding.

  • No default passwords: Have factory passwords been changed on all cameras, NVRs and switches? Is the same password used on every device?
  • Password management: Are admin passwords stored per device in a password vault?
  • Firmware: Is each device’s version in the inventory? Are manufacturer security bulletins checked periodically?
  • Unnecessary services: Are Telnet, FTP, UPnP, SSH (if unused) and old RTSP authentication modes disabled?
  • Encrypted management: Is the management interface over HTTPS? Can corporate certificates be used instead of self-signed ones?
  • Authentication: On the ONVIF side, is digest or TLS used instead of the old username token? ONVIF cites the weakness of username token as one of the reasons for deprecating Profile S (ONVIF guide).
  • Time: Are all devices synchronized to a reliable NTP source? A wrong clock makes logs and incident review meaningless.
  • Support lifetime: Are devices past manufacturer support (EOL) listed and isolated?

Be careful when turning off services: some VMS products need certain services for camera discovery, receiving events or PTZ control. Make the change on a test camera first; verify in the VMS that video, motion events and PTZ work, then roll it out as a template. A bulk change made without testing ends with a morning list of cameras that have stopped recording.

A common mistake when moving to HTTPS is loading the certificate on the camera but leaving the VMS connection on HTTP. Management traffic between camera and VMS must also be encrypted. If the VMS validates certificates, the corporate CA’s root certificate must be added to the server; otherwise cameras drop with a “connection error”.

2. Network level

If the camera network is flat, device hardening alone isn’t enough. If a compromised laptop on the user network can reach the cameras’ management interface directly, the problem is in the network design.

  • Segmentation: Are cameras in a segment separate from the corporate user network and the OT network? Is traffic between segments limited by rules? (camera VLAN)
  • Internet exposure: Is there port forwarding for the NVR or cameras? Are there ports opened automatically via UPnP? Has this been verified with a scan from outside the organization?
  • Cloud/P2P services: Are the manufacturer’s P2P connection services deliberately enabled?
  • Cameras’ outbound access: Is cameras’ access to the internet closed?
  • Remote access: Is a VPN or an audited intermediary service used? Is there multi-factor authentication?
  • Switch security: Are unused ports shut down? Has port security or 802.1X been considered?

Check internet exposure by measuring from outside, not by looking at the rule table. Run an authorized port scan of your own public IP addresses from a connection outside the organization. If the NVR’s web port, RTSP (554) or manufacturer-specific service ports respond, there’s a forward you can’t see in the firewall. A router with UPnP left on does this silently.

For cameras’ outbound traffic, write a default “deny” rule from the camera segment to the internet; open exceptions only for NTP, the VMS and, if needed, the update server. This rule also stops P2P services from opening a tunnel to the manufacturer cloud without your knowledge. Limiting DNS in the camera segment to the internal DNS server only makes outbound connections visible in the logs.

3. Access and permissions

A shared “admin” account leaves the post-incident question “who exported the recording?” unanswered. This topic is less technical than procedural, but it’s one of the topics with the most findings in audits.

  • Personal user accounts: Does every user have their own account instead of a shared “admin” account?
  • Role-based permissions: Are operator, reviewer and system administrator permissions separated? Is export permission limited?
  • Camera scope: Do branch or store users access only their own cameras?
  • Leavers: Is access closed for departing staff and service companies?
  • Review: Are permissions reviewed every three to six months? (access and audit trail)

Giving service companies permanent admin accounts is a common mistake. A maintenance account should be opened for a limited time and closed when the job is done. When closing the account, the company’s remote access path should also be closed; deleting only the VMS account can leave an old VPN user on the router active. If the VMS supports directory (AD/LDAP) integration, manage users from there. When a departing person is disabled in the directory, camera access is closed too.

4. Operations and monitoring

If hardening is done once and left, six months later newly added cameras are running to the old standard. Operational processes make it stick.

  • Inventory: Are model, serial number, IP, MAC, switch port, firmware and location up to date?
  • Logs: Are login attempts, configuration changes and exports logged? Are logs sent to a central system?
  • Change management: Are configuration changes recorded?
  • Backups: Are NVR/VMS configuration backups taken, and has a restore been tried?
  • Update process: Are updates tried on a test group and applied in a planned window? (firmware management)
  • Physical security: Are NVRs and recording servers in a locked cabinet or room? Are cameras’ accessible cables protected?

A firmware update carries its own risk. A new version can change ONVIF behavior, stream profiles or authentication mode. Apply it first to a test group of two or three cameras; verify recording, events and export in the VMS, then roll it out in waves. Keep the old firmware file and a device configuration backup for rollback. A device without a backup isn’t updated.

Order for applying the checklist on site

  1. Build the inventory: compare the switch MAC table, DHCP leases and the VMS camera list. A device that doesn’t appear in any list is the riskiest device.
  2. Verify external exposure; if there are open ports, close them before anything else.
  3. Change default and shared passwords and store them per device in a password vault.
  4. Apply segmentation and outbound rules.
  5. Verify services, HTTPS and the authentication mode on a test camera and roll them out with a template.
  6. Set up the user and role structure and route logs to a central system.
  7. Start firmware updates in a planned window, with a test group.

The order matters. Working on certificates while an NVR is exposed to the internet is a waste of time.

Prioritization

Reporting findings with equal weight tires decision makers and hides the real risk. The classification below makes clear what gets closed in the first week.

Priority Example finding Action
Emergency Internet-exposed NVR, default password Close immediately and change passwords
High Flat network, shared admin account, EOL devices Segmentation, personal user accounts, isolation
Intermediate HTTP management, old firmware HTTPS, planned updates
Process Missing inventory and logs Inventory, central logging, review schedule

Frequently asked questions

Our cameras are on a closed network; do they still need updates?

Yes. A closed network reduces risk but doesn’t eliminate it; a device or person with network access can exploit known vulnerabilities. Updates should be tested and applied in a planned way.

Should we turn off the manufacturer cloud (P2P) service?

If you don’t use it, turn it off. If you do, assess encryption, authentication and the data path; in a corporate environment, a VPN or an audited access method is preferred.

Does this checklist replace a penetration test?

No. It’s a configuration and hygiene check. An authorized penetration test can reveal additional findings from an attacker’s perspective.

To apply this list to your own system and fix findings in a controlled way, see our IP camera and NVR cybersecurity hardening service. For organization-wide penetration testing and security audits, we work together with Eretek Cybersecurity.

Support

Contact us for support or a discovery session on this topic.

Contact form
[email protected]

Eretek Video Surveillance

IP cameras, recording and central management infrastructure. We plan blind spots, retention periods and the camera network around the risks of each site.

Knowledge Center

Other articles