Canonical: https://zela-web.nexusrun.app/docs/glossary/ Source: https://github.com/nexusrun/zela/blob/main/docs/glossary.md Learn the words you will see in Zela, without the jargon. # Plain-language glossary [All guides](https://zela-web.nexusrun.app/docs/) ยท [Browser management guide](https://zela-web.nexusrun.app/docs/console-guide/) You do not need to memorize these terms. Look them up when a setting or message uses one. ## Applications and servers | Term | What it means in Zela | | ------------------------------- | ----------------------------------------------------------------------------------------------------------- | | **Cluster** | Your group of servers, managed together | | **Manager / control plane** | The computer and Zela program that provide the dashboard and coordinate work | | **Worker / host / node** | A Linux server where applications run | | **Agent** | The small `zela-worker` program that lets the manager communicate with Docker and collect host measurements | | **Docker** | Software on a worker that runs packaged applications in containers | | **Container image** | The packaged application used to create a container. `nginx:alpine` is an example | | **Image tag** | The version label after `:` in an image name. Changing it can change the software you deploy | | **Container** | One instance created from an image, with its own running process and local writable files | | **Deployment** | The saved application settings Zela manages: image, copies, resources, and placement | | **Replica** | One copy of a deployment. Two replicas mean two containers | | **Stateless application** | An app whose important data is stored elsewhere, so its containers can be replaced without losing that data | | **Volume / persistent storage** | Storage intended to keep data when a container is replaced. Zela does not copy it between workers | | **Environment variable** | A named application setting, such as `APP_ENV=production`; it may contain a secret | | **CPU / memory / RAM** | Processing power and working memory available to an application | | **Reservation** | Resources Zela accounts for when deciding whether a worker has room. It is different from measured usage | ## Status and actions | Term | Meaning | | -------------------------- | ------------------------------------------------------------------------------------------------------------------------- | | **Online** | The manager can reach the worker. Check freshness and any reported errors too | | **Active** | A worker may receive new application placements | | **Paused** | No new placements; existing containers continue running | | **Drain** | Try to move eligible managed applications away before server maintenance. Check the result before rebooting | | **Scale** | Change the number of application copies | | **Rollout / update image** | Replace application copies using another image | | **Degraded** | The deployment is not meeting its intended running state; inspect its error and replicas | | **Operation** | A tracked change, such as deploying or draining. Starting it is not the same as completing it | | **Failover / recovery** | Recreate lost managed application copies on another suitable worker when possible | | **High availability / HA** | Extra managers waiting to take over management if the active manager fails; this does not replicate your application data | | **Acknowledge** | Mark an alert as reviewed. It does not fix the underlying problem | ## Networking and access | Term | Meaning | | ---------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------ | | **IP address / hostname** | An address identifying a computer, such as `10.0.0.21` or `worker.example.com` | | **Port** | A numbered entry point on a computer. One computer can serve different programs on different ports | | **Published port** | A mapping from a worker's port to an application's container port. `8081:80` means worker port 8081 leads to container port 80 | | **localhost / 127.0.0.1** | The computer where the command or browser is running; inside a container it refers to that container | | **Cluster mesh / WireGuard** | The optional encrypted private network connecting Linux workers | | **Subnet / CIDR** | An address range. `10.200.0.0/16` describes the size and location of one range; ask your network administrator to choose a non-overlapping range | | **Peer endpoint** | The worker hostname or IP address the other workers use for their encrypted network connection | | **Bridge network** | A Docker network local to one worker. Creating one does not connect different workers | | **VIP / virtual IP** | A stable private service address that can send traffic to the deployment's healthy copies | | **DNS** | The system that turns a name into an address. Internal names such as `web.zela` are different from public website DNS | | **Edge proxy / route** | A worker service and rule that forward a website hostname/path to an application | | **HTTPS / TLS certificate** | Encrypted browser communication and the certificate that identifies the server | | **SSH** | A secure way to access a server. Zela can use it to install and manage a worker | | **SSH login key** | Your private key used to sign in to a server; store it in the Secrets vault | | **SSH host public key** | The server's identity. Verify it through a trusted source before giving it to Zela | | **Administrator token** | The secret used to sign in and fully control the cluster | | **Vault key** | The master encryption key used to unlock saved SSH secrets. Keep it with your recovery materials | | **MCP** | The connection that lets an AI assistant call Zela's supported tools | | **Read / management access** | Permission to inspect information / permission to make changes as well | | **PostgreSQL** | The database used to save Zela's configuration and history; separate from application databases |