Picture this: you just rolled out a new switch lineup, your CI pipeline is humming, and then a developer asks for access to an Arista control plane repo tied to a legacy Mercurial setup. Everyone freezes. Someone mutters, “Wait, we still have that?” This is where things get interesting.
Arista Mercurial refers to the layered interaction between Arista infrastructure automation and Mercurial’s distributed version control. Arista’s EOS and CloudVision already make network provisioning repeatable. Mercurial brings versioned configuration management that’s fast, simple, and locally resilient. Put the two together and you get auditable, versioned network automation that fits into modern DevOps flow, yet avoids the sprawl of traditional Git-based playbooks.
The pairing works through configuration repositories. Engineers push intent—VLAN mappings, ACL updates, or interface policy—into Mercurial branches. Arista’s automation tools then consume that repo, enforcing configuration through CloudVision or direct EOS APIs. Access control stems from Mercurial’s built‑in ACL system, so you can match commit permissions to your identity provider, often via SAML or OIDC integration. The result is traceable change management with one version trail per environment.
When wiring it all up, keep these patterns in mind. First, map identities early. Whether your source of truth is Okta, AWS IAM, or on-prem LDAP, consistency across version control and Arista workflows prevents ghost users from pushing unreviewed configs. Second, automate deployments through CI triggers that validate syntax and configuration diff before applying. Third, treat every repo branch as a potential production candidate. It keeps rollback simple and prevents drift between test and live paths.
Done right, Arista Mercurial gives you: