Config will not load after an upgrade
Renovate renamed or removed an option in a newer release, and a config pulled from an older example now fails validation on the version Mend is running.
Config valid on the current release
Running Renovate?
Renovate is powerful and opinionated, and most teams that bring us in have either turned every knob or none of them. The symptoms are familiar: a giant dependency dashboard issue that no human reads, a schedule that fires in the middle of a release window, or a prBodyNotes block that was copied from an example and now points at a changelog that does not exist. We rewrite the config until it opens PRs the team actually merges.
What it is
Renovate is an automated dependency update tool originally written by Rhys Arkins and now developed under Mend (formerly WhiteSource). It runs as a hosted GitHub App, a free Mend Renovate service, or a self-hosted CLI against GitHub, GitLab, Bitbucket, Azure DevOps and Gitea.
The configuration model is a single renovate.json or renovate.json5 in the repository, with a rich packageRules system for grouping, scheduling, labelling and automerging. It supports more than ninety package managers and ecosystems, including language ones, Docker, Helm, Terraform, Kubernetes manifests, GitHub Actions, GitLab CI includes, and ad hoc regex matchers for anything else that pins a version in a text file.
The honest version
No technology is all upside. This is where Renovate earns its place and where it tends to cause trouble.
Sound familiar?
The Renovate problems that bring people to us, and what you have once each one is fixed.
Renovate renamed or removed an option in a newer release, and a config pulled from an older example now fails validation on the version Mend is running.
Config valid on the current release
A concurrent PR limit that is too low for a repo with many dependencies, leaving everything queued on the Dependency Dashboard without anyone noticing.
Throughput matches the review capacity
A manager that was enabled against a path convention the repo does not follow, or a file that was deleted during a reshuffle without disabling the matching manager.
Managers scoped to what the repo really has
A notes template that uses a variable name that was renamed between Renovate versions, or a Mustache typo where the triple braces do not match.
PR bodies render with useful context
A public registry outage, or a private registry behind an intermittent proxy, that returns transient errors Renovate treats as hard failures on the first attempt.
Retries tuned so transient errors recover
A custom regex manager with an invalid pattern, or a lookahead that is valid in one regex engine but not Renovate's, which fails the entire run until it is fixed.
Custom managers compile and match
A self-hosted run using a user PAT instead of a GitHub App, or a very large organisation sharing a single PAT across tools that collectively run out of API budget.
Rate budget respected, runs complete
A monorepo with too many dependencies to resolve in a single run, leaving half the packages deferred and the dashboard permanently behind.
Lookup budget sized for the repo
What we do
Pick up the phone for a single stuck repo, or hand us the whole update estate.
How it works
Tell us what's going wrong. We look at it, then tell you plainly what's wrong and what it will cost.
We fix the cause, not the symptom. If something is down, getting it back up comes first.
Backups, monitoring, updates and access, so the same problem can't quietly come back.
You get a plain-language write-up. Or we stay on for hosting, upkeep and new work.
Often found together
We work on the whole stack, so the problem can be wherever it actually is.
By default it opens a PR for every update it finds, which is more than most teams can review. Group patches by ecosystem, bundle minors on development dependencies, and schedule the runs to working hours only. Within a sprint the PR count usually drops to something the team can actually get through.
The bot is hitting the concurrent PR limit, the per-run lookup budget, or a schedule that is more restrictive than the config implies. The dashboard itself labels each item with the reason; the fix is to raise the specific limit, loosen the schedule, or group updates so the backlog clears instead of sitting.
Add a hostRules entry matching the registry hostname and reference an encrypted credential or a secret depending on whether the bot is hosted or self-hosted. For JFrog, Nexus and Artifactory the matchHost pattern must match the exact proxy path the package manager uses, which is the place most first configs go wrong.
Only if you need something Dependabot cannot do: multi-platform support, more complex grouping, custom regex managers, or ecosystems Dependabot does not cover well. For a straightforward GitHub repo, Dependabot is fine and switching for its own sake just adds a config file to maintain.
Yes. The schedule key accepts cron-like windows and human expressions such as before 5am on monday or every weekend. Combining a weekly schedule on main dependencies with daily runs on dev-only packages gives most teams the balance they want.
The estimate is free. Describe the repository layout and the pain points, and a written quote comes back to you before any time is booked.
Free estimate
Tell us what you're running and what's going wrong. A person reads it and comes back with an estimate.
Your message is with a real person, and we'll get back to you soon.
Rather talk now? Call us: