mrkeyoor.com_
Tue 01 Sept 17:41 UTC
Self-Hostedevaluationupdated 30 Aug 2026

wekan review

WeKan is a self-hosted kanban application for organizing cards into boards, lists, and collaborative workflows. It gives individuals and teams a visual way to track work while keeping project data on infrastructure they control.

+2 / 2dstars / 7d
Verdict

Our install finished in 50 seconds, but tests stopped after 7 seconds because Meteor was missing. WeKan is a credible choice for organizations that want mature, actively shipped kanban software and accept the operational responsibility of MongoDB, backups, and frequent upgrades. Choose it for control and deployment flexibility, not for a zero-maintenance experience.

We ran it

Install✓ · 50s1042 packages · 493 MB
Buildn/ano build script
Tests✗ · 7sran, no count parsed
Known vulns00 critical · 0 high · 0 moderate · 0 low (npm audit)
Repo3324 files~340,906 lines of source · 114.2 MB · 12 CI workflows · Dockerfile · tests dir

Answers from our run

Does wekan build from source?

Dependencies installed in 50 seconds (1042 packages), and the project has no separate build step. We cloned commit 6bfc759 into a clean Debian container with 3 CPUs and no project-specific setup.

Do wekan's tests pass?

The test command failed in our container, and its output did not report a pass or fail count.

Does wekan have known vulnerabilities in its dependencies?

npm audit found none in the dependency tree at the time of our run.

Who should not use wekan?

Teams wanting a maintenance-free hosted service, because WeKan expects active administration

What are the alternatives to wekan?

Kanboard, Planka. Our install finished in 50 seconds, but tests stopped after 7 seconds because Meteor was missing.

Setup3/5Install worked, but tests need a separate Meteor toolchain
Docs4/5Clear prerequisites, deployment links, FAQ, and developer guidance
Community4/521k stars, same-day release and push, but 281 open issues
Maturity4/5Long-running project with production guidance and many platforms

Discussed on

  1. hnWekan: An open-source Trello-like kanban306 points
  2. hnWekan: Open-source, trello-like kanban278 points
  3. hnWeKan: Open-Source Meteor Kanban194 points
  4. hn2 years of Wekan: what we archieved5 points
  5. hnWekan – a self-hosted alternative to Trello4 points

Who it’s for

Teams that want an open-source, self-hosted kanban board
Organizations that need control over project data and deployment
Multilingual groups that benefit from WeKan's broad translation coverage
Administrators comfortable maintaining MongoDB, backups, and frequent updates

Who it’s NOT for

Teams wanting a maintenance-free hosted service, because WeKan expects active administration
Very small servers, because production guidance calls for at least 4 GB total RAM
Operators who cannot monitor storage and back up MongoDB daily
Contributors seeking a plain npm-only workflow, because tests require Meteor

Setup reality

Our install succeeded in 50 seconds, placing 1,042 packages and 493 MB on disk, but the test command failed after 7 seconds with exit 127 because meteor was not found. There was no build script or target for our runner, so build was skipped. The README does document Meteor as a prerequisite and offers Docker, Compose, and an interactive build.sh, but this is not a plain clone-and-npm test project; expect to install the required runtime or choose a packaged deployment route.

It is an established kanban system, not a lightweight board widget

WeKan is a full collaborative kanban application built for people who want visual task management without handing their data to a proprietary hosted service. The project dates to 2014, has about 21,067 GitHub stars in the supplied repository snapshot, and is licensed under MIT. Its central promise is straightforward: run shared boards, lists, and cards on infrastructure you control. The README positions it for everything from personal planning to large organizations, and says its largest user has 30,000 users. That range makes WeKan more like an operational application than a library you casually add to another product.

The project also takes international use seriously. Its Transifex program covers 234 languages, with 197 described as essentially complete, while the interface updates in real time for collaboration. Installation choices include Docker images on GitHub, Quay, and Docker Hub, plus a published Compose file and other platform-specific routes through the project site. This breadth is valuable for organizations whose infrastructure rules differ by region or department, although each deployment option still leaves someone responsible for upgrades, storage, and recovery.

What happened when we ran it

On our box, dependency installation succeeded in 50 seconds. npm installed 1,042 packages and used 493 MB on disk. The checked-out repository contained 3,324 files, about 340,906 lines of source, and occupied 114.2 MB before installation. Our runner found no build script or target, so it skipped the build rather than pretending that compilation passed. That distinction matters: a successful dependency install proves the package graph resolved in this environment, not that the full application was ready to serve traffic.

Tests failed after 7 seconds with exit code 127. The command attempted meteor test --once --driver-package meteortesting:mocha, then the shell reported meteor: not found. That is the complete evidence from the failure log, so we cannot claim whether the suite itself would pass once Meteor is installed. The README does list Meteor as a prerequisite and says the main branch uses Meteor 3.5 with Node.js 24.x, whereas our sandbox image supplied Node.js 22. The repository therefore expects more than a routine npm workflow, even though package installation itself was quick.

The security scan was the cleanest result in our run: npm audit reported 0 known vulnerabilities across critical, high, moderate, and low severity. We also observed 12 CI workflow files, a Dockerfile, a Compose file, and a tests directory. Those are good engineering signals, but they are not substitutes for a passing test run. In practical terms, reproduce the documented Node.js 24.x and Meteor setup, then run the project’s own test path before approving a production image.

Its operational honesty is a real strength

The README is unusually direct about production responsibilities. It calls for at least 1 GB of free RAM and recommends a minimum of 4 GB total RAM for a production server. For larger deployments, it sketches multiple frontend servers and a separate MongoDB backend. More importantly, it warns that exhausting disk space can corrupt the database, recommends low-disk alerts, and calls for database backups at least once per day. It also admits there is no undo for some destructive user actions and that database content may occasionally require manual repair.

That candor helps an administrator make a realistic decision. WeKan advises updating to the newest release frequently because only the latest version is supported, and its supplied activity data backs up the claim of active delivery: v11.29 was released on 2026-08-30, with the latest push occurring minutes earlier the same day. The repository has 281 open issues, which is substantial, but issue count needs context. A same-day push and release indicate an actively maintained project with a sizable support surface, not an abandoned backlog.

The rough edges are mostly about ownership and complexity

Self-hosting WeKan transfers control to you, but it also transfers chores. MongoDB health, capacity alerts, daily backups, upgrade testing, and recovery procedures need named owners. The README’s warning that fixes and features can arrive many times per day suggests a fast-moving release stream, which may be reassuring for security but demanding for change control. Its Standard for Public Code assessment from 2023 reported meeting 8 of 16 criteria out of the box. That does not invalidate the software, but it is a useful reminder to evaluate governance and operational fit alongside feature coverage.

The contributor experience is also specialized. Development uses an interactive build.sh menu with separate setup, build, and dev-server stages, plus Meteor 3.5 and Node.js 24.x prerequisites. That can be perfectly workable for regular contributors, but it is less familiar than a single package-manager command. The failed test in our run shows the practical consequence: npm dependencies alone were insufficient. Teams should document a reproducible toolchain, pin the required versions, and validate the exact container or host image they plan to maintain.

It belongs as a managed internal service

In a real stack, WeKan fits behind your existing TLS gateway, identity controls, monitoring, and backup system. Treat the application and MongoDB as a stateful service, not as an expendable frontend container. Start with the official Compose example or another documented package, set storage alerts before inviting users, and prove restore procedures from a 1-day backup cadence. For thousands of users, the README’s example separates several frontend instances from a larger database machine, so capacity planning should cover both application concurrency and persistent storage.

WeKan earns consideration because it combines an MIT license, wide deployment support, 197 essentially complete translations, and very recent maintenance. Kanboard is a better fit when simplicity and a traditional PHP stack matter most; Planka is attractive when a modern Trello-like presentation is the priority. Choose WeKan when data control, multilingual collaboration, and flexible hosting outweigh the need for a one-command developer experience. Its strongest quality is not effortless setup, but clear evidence that the project understands the operational cost of serious self-hosting.

Alternatives

ProjectWhat it isPick it when
KanboardA deliberately simple PHP kanban application with a plugin ecosystem.pick this instead when you value a smaller, more traditional server stack and simpler boards.
Planka gh↗A modern open-source kanban board with a polished visual interface.pick this instead when interface polish and a Trello-like experience matter more than WeKan's deployment breadth.

What people are saying

  1. [velocity-scout] wekan/wekan

Sources

  1. WeKan GitHub repository
  2. WeKan official website

More self-hosted reviews

v2 · OpenShell · wigolo · Mindwtr · club-3090 · reclip · the whole board →