Self-signed certificate
What is Self-signed certificate?
A self-signed certificate is an X.509 certificate whose issuer and subject are the same entity, signed with its own key rather than a certificate authority, so no trust chain backs it. AdminKit's SSL checker tool reports whether a live certificate is self-signed or chains to a publicly trusted root.
By AdminKit team, Netbase JSC · Updated
A self-signed certificate is an X.509 certificate — the certificate format RFC 5280 (current document status at its IETF Datatracker entry) defines — where the issuer and subject are the same entity — it’s signed with its own private key rather than by a separate certificate authority, so no chain of trust connects it back to a root a browser or OS already trusts. This makes it cryptographically valid for encrypting a connection but useless for proving identity to a stranger: anyone, including an attacker performing a man-in-the-middle attack, can generate a self-signed certificate for any hostname, so a client has no way to tell a legitimate one from a forged one without some other, out-of-band way of confirming it. Browsers show a hard trust warning for exactly this reason, following the publicly-trusted-root rules the CA/Browser Forum’s baseline requirements set for certificate authorities. Creating one takes a single command: the OpenSSL req documentation notes that the command can create self-signed certificates, for use as root CAs for example, and gives openssl req -x509 -newkey rsa:2048 -keyout key.pem -out req.pem as its example for generating a self-signed root certificate. Self-signed certificates remain legitimate for internal services, local development, and device management interfaces where the client explicitly imports the certificate ahead of time, rather than relying on the public CA trust model at all.
AdminKit groups self-signed certificate under its machine protocols cluster — browse the AdminKit glossary index for other related terms.
Frequently asked questions
What makes a certificate self-signed?
Its issuer and subject are the same entity, signed with its own private key rather than a separate certificate authority, so no chain of trust connects it back to a trusted root.
Why do browsers show a warning for self-signed certificates?
Anyone, including an attacker, can generate a self-signed certificate for any hostname, so a client has no way to distinguish a legitimate one from a forged one without out-of-band verification.
Are self-signed certificates ever appropriate to use?
Yes -- for internal services, local development and device management interfaces where the client explicitly imports the certificate ahead of time, rather than relying on the public CA trust model.
Does a self-signed certificate still encrypt the connection?
Yes -- it is cryptographically valid for encrypting traffic, just not for proving identity to a stranger.