Explore our new Cybersecurity Terms Glossary, explained by our experts.
Cybersecurity Terms Glossary
TACACS+ authenticates administrators to network devices and can authorise individual commands, which is what separates it from RADIUS in a network operations context.

What Is TACACS+?
TACACS+ exists to answer a question RADIUS answers badly: when an engineer logs into a core switch, which commands may they run? Cisco developed the protocol in the early 1990s, and although it remained proprietary for decades it was finally published as an informational specification, RFC 8907, in 2020. It is supported by Cisco, Juniper, Arista, HPE, and most other serious network vendors.
The AAA acronym—authentication, authorisation, accounting—applies to both TACACS+ and RADIUS, but only TACACS+ separates them into independent exchanges.
The practical consequence is a model of network administration that is otherwise very hard to build. A first-line engineer can be permitted show commands and interface descriptions but refused configure terminal. A contractor can be allowed to manage access ports and refused anything touching routing. A change window can enable a wider command set for named people and close again afterwards.
This is the principle of least privilege applied to a command line, and it is why TACACS+ is a common finding in network security reviews where RADIUS has been used for device administration.
| TACACS+ | RADIUS | |
|---|---|---|
| Transport | TCP, port 49 | UDP, ports 1812 and 1813 |
| AAA separation | Independent exchanges | Authentication and authorisation combined |
| Payload protection | Entire body obfuscated | Only the password attribute |
| Per-command authorisation | Yes | No |
| Typical use | Device administration | Network access for users |
| Interoperability | Good, long proprietary history | Open standard since 1997 |
The division of labour most organisations settle on is TACACS+ for administrative access to network equipment, and RADIUS with EAP for user network access over 802.1X and VPN. Cisco ISE and Aruba ClearPass both speak each protocol for its respective purpose.
TACACS+ encrypts more than RADIUS, which is a low bar. The body of each packet is obfuscated using an MD5-based stream derived from a shared secret, the session identifier, and the sequence number. This is not authenticated encryption: there is no integrity protection, the construction is homegrown rather than standard, and RFC 8907 says plainly that the mechanism should be considered insecure and that the protocol ought to be carried over a protected transport.
What follows from that is straightforward. Carry TACACS+ on a dedicated, isolated out-of-band management network. Use a unique, long, randomly generated shared secret for every device rather than one value pasted into every configuration. Restrict which source addresses the server will accept. And where your platform supports TACACS+ over TLS, use it.
A few decisions account for most of the value, and most of the incidents.
Decide the fallback. If the TACACS+ server is unreachable, the device either falls back to a local account or refuses everyone. Both are defensible; only one is right for a given site. Whichever you choose, keep exactly one local emergency account per device with a unique vaulted password, and make sure its use raises an alert.
Make use of the accounting. Command accounting produces a complete record of every command typed on every device, attributed to a named person. That record is the fastest way to answer “who changed the access list on Tuesday evening”, and it is frequently the only evidence available after a network outage. Forward it to a SIEM, not just to a local file, and alert on command sets associated with configuration tampering or log clearing.
Add a second factor. TACACS+ can chain to a one-time password or push system, and administrative access to network infrastructure is exactly the category where a password alone is insufficient. Pair this with privileged access management so that sessions are brokered and recorded as well as authorised.
Build redundancy. Two servers in separate locations, because an authentication outage during a network incident is the worst possible time to discover you have one.
At Privia Security, we review network device administration and AAA design in our network security services.
Use TACACS+ for administrative access to network devices, because it can authorise individual commands and produces a full command accounting trail. Use RADIUS for user network access over 802.1X and VPN, where its attribute model for VLAN and access list assignment is what you need. Most organisations run both, often on the same platform.
The packet body is obfuscated with an MD5-based construction, which is better than RADIUS protecting only the password field but is not modern encryption. There is no integrity protection, and RFC 8907 recommends treating the mechanism as insecure and carrying the protocol over a protected transport. Use an isolated management network, unique per-device secrets, and TACACS+ over TLS where available.
The device asks the TACACS+ server whether to permit each command before executing it, rather than granting a privilege level at login and allowing everything within it. That lets you give an engineer read-only diagnostic commands, or allow interface configuration while refusing routing changes, without creating custom privilege levels on every device.
Decide deliberately rather than accepting the default. Falling back to a local account keeps devices manageable during an outage but creates a credential that must be unique per device, vaulted, and alerted on. Refusing all access is safer but can leave you locked out during exactly the incident you need to fix. Redundant servers in separate locations reduce the dilemma.
You May Be Interested In These
We use a few cookies that are needed to make this website work, and they are always on. We would also like to set analytics cookies, to see which pages are useful; functional cookies, so that our live chat works; and advertising cookies, to measure our campaigns. We will not set any of those until you say yes, and you can change your mind at any time from the Cookie Settings link in the footer.
Choose the cookies you are happy for us to use. Strictly necessary cookies are always on; everything else is off until you turn it on.
These cookies keep the website working: they hold your security checks, your form submissions and the choice you make here. The site cannot run without them, so they are always on.
These let our live chat open and remember your conversation, and let the newsletter sign-up work. Switch them off and the chat window will not appear.
These tell us which pages are read and how people find us, so that we can improve the site. The reports are about groups of visitors, never about you personally.
These let us measure our advertising and show you our adverts on other websites, such as LinkedIn, Google and Meta. Switch them off and you will still see adverts, but they will be less relevant.