JFrog Artifactory Certified DevOps Engineer key terms

A repository fully hosted and managed by your own Artifactory instance, holding artifacts you deploy yourself. It is the normal destination for build outputs and never proxies an external source.

A repository that proxies and caches an external package source such as Maven Central or npmjs. Requested artifacts are fetched over the network on first request and stored locally for later use.

A logical aggregation of several local and remote repositories exposed as one URL. Clients resolve and deploy through it, so repository topology can change without reconfiguring builds.

Configuration that defines, for a package type, how artifact paths map to module, version and folder conventions. It also determines where generated package metadata is written and read.

The artifact format a repository serves, such as Maven, npm, Docker, PyPI, NuGet or Helm. It determines the client protocol, the metadata Artifactory extracts, and the layout options available.

Artifactory stores each unique binary once, addressed by its checksum, and links every artifact path to that single copy. Re-deploying identical content therefore consumes no additional disk space.

The front door of a JFrog Platform Deployment. It receives all client traffic on ports 8080 and 8082 and forwards each request to the correct microservice, such as Artifactory, Access or Distribution.

A complete installation of Artifactory together with the platform services it depends on - router, access, metadata, event - plus optional products like Distribution or Xray, sharing one database and binary store.

The platform microservice responsible for authentication, authorization and token issuance. It stores users, groups and permission targets, and issues the access tokens every other service validates.

A platform microservice that provides shared metadata storage for other JFrog services, keeping their records separate from the binary repository content that Artifactory manages.

A platform microservice that records and publishes platform events, such as an artifact being deployed or a release bundle being created, so webhooks and other services can react asynchronously.

The main configuration file of a JFrog Platform Deployment, covering router ports, database, binary store, filestore and shared node settings. Changes are applied by restarting the affected services with the provided scripts.

The file that declares which storage backend holds artifact binaries - filesystem, S3, Azure, GCS - and how they are sharded, cached or made redundant. It is referenced by system.yaml.

Splitting binary storage across several independent providers according to checksum ranges. This distributes read and write load and lets capacity grow beyond a single volume or bucket.

A maintenance job that permanently deletes binary blobs no longer referenced by any artifact path, reclaiming storage after deletions. It runs on a schedule and can also be triggered manually.

Running several Artifactory nodes against one shared database and binary store behind a load balancer, giving failover and more throughput. A license and a shared filestore are required.

A web server such as Nginx or Apache placed in front of the platform to terminate TLS, rewrite URLs, enforce upload size limits and forward traffic to the router.

Artifactory keeps configuration and metadata in a relational database; production deployments require PostgreSQL, Oracle or MSSQL, while the bundled Derby database is only for evaluation.

A mechanism that writes repositories and configuration to a folder for backup or migration, and restores them on another instance. Useful for moving between binary store backends.

A named rule that grants or denies actions such as read, deploy, annotate, delete and manage on selected repositories or build names, to specific users and groups.

A set of users managed in the Access service and used to assign permissions collectively. Groups can be created locally or mapped from LDAP or SAML identity provider claims.

A global setting that lets unauthenticated clients read from repositories, with actual access still controlled per repository by permission targets. Commonly used for public read-only caches.

Configuring the platform to verify credentials against an external directory server and to resolve that directory's groups, which are then mapped to Artifactory groups for authorization.

Browser single sign-on in which an external identity provider authenticates the user and sends an assertion, after which the platform creates a session and derives group membership from the assertion.

A bearer credential issued by the Access service with an expiry and optional scope, used by CI jobs and scripts instead of passwords. Tokens can be revoked without changing the user's password.

A feature that synchronizes entities such as users, groups, permission targets and tokens between multiple JFrog platforms, keeping identity and authorization consistent across sites and regions.

A structured record of a build run, including modules, dependencies, deployed artifacts, checksums and environment details. Published to Artifactory, it enables traceability and reproducible builds.

Using a native build tool extension or the JFrog CLI so that dependency resolution and deployment go through Artifactory while build info is captured automatically.

A command-line client that wraps platform REST APIs for upload, download, search and build workflows, and can run Maven, Gradle, npm and similar tools while recording build info.

Key-value pairs attached to artifacts or folders, such as build.number or approval status. They can be searched, used in queries, and drive cleanup or promotion rules.

Artifactory Query Language, a domain-specific syntax for finding artifacts by properties, build info and repository attributes. Queries run over REST or the CLI and return structured JSON.

Moving a build's artifacts to another repository, for example from staging to release, copying or moving files and optionally updating their properties and status.

The HTTP interface used to automate the platform: creating repositories, uploading and downloading artifacts, running searches, managing permissions and configuring replication, authenticated with tokens or credentials.

A repository setting that sends newly deployed or changed artifacts to a remote Artifactory instance on a cron schedule, optionally filtered by path patterns.

The local instance periodically fetches artifacts from a repository on a remote Artifactory, useful when the source cannot initiate connections or when sync timing must be controlled locally.

An immutable, versioned and signed collection of artifacts created by Distribution and promoted through stages, so remote edge nodes can serve exactly the same content.

A platform product that creates release bundles, signs them and distributes them to edge nodes at remote sites, giving controlled, verifiable delivery of released artifacts.

A Distribution component at a remote site that receives release bundles and serves their artifacts locally, cutting cross-site latency and central bandwidth use.

An HTTP callback registered in the platform that fires on events such as artifact deployed or release bundle created, posting a JSON payload to an external endpoint.