First Login & Setup

The guided installer checks PostgreSQL, applies the Prisma schema, seeds the initial data, and creates the administrator account before it prints the browser URL. This page covers the first sign-in, what the recovery screen is telling you if seeding did not finish, and how to add the next users.

First sign-in

The bootstrap accounts, and when SeqDesk refuses to create them

SeqDesk ships two built-in default accounts as a recovery convenience:

They are not created unconditionally. The seeder distinguishes two situations:

• A brand-new, completely empty database. If the install named no account of its own, the built-in defaults are created so that there is always a way in. An unattended install that omits bootstrap account settings is exactly this case, and is where you will meet admin@example.com in practice.
• A database that was already configured — site settings exist but no users do. Here the shipped password is refused outright. The log says Not creating bootstrap admin account …: This database was already configured, so no account is created with the built-in default password, and it tells you to set bootstrap.users.admin.password (or passwordHash) in settings.json and restart.

The same refusal applies when the install names an account but supplies no password for it: This install names the account but configured no password for it.

Resetting a password

There is no “forgot password” link and no password-reset email — SeqDesk has no outbound mail path it can rely on. Recovery is operator-driven, from the shell on the host:

cd "$HOME/seqdesk"
seqdesk reset-password admin@example.com

Pass --password <value> to supply one instead of having it generated, and --dir /path/to/seqdesk if you are not already in the install directory. This is also how you unlock a researcher who has lost their password.

What else the seed created

Alongside the accounts, the first seed writes:

• The SiteSettings singleton — site name SeqDesk, brand colours, contact email, and the default post-submission instructions researchers see after they submit an order.
• The default sequencing order form: a Sequencing Order Details group (Sequencing Order Name, Number of Samples), a Sequencing Information group (Sequencing Technology, Library Strategy, Library Source), per-sample fields (Organism, Sample Title, Sample Alias, Sample Volume, Concentration), and four admin-only facility fields.
• The default study form: Principal Investigator, Study Abstract, a MIxS metadata field, and collection date.

Everything there is editable from the admin area — see the Admin Quickstart.

For a populated evaluation, configure a sequencing-data path and then open Admin → Settings → Demo data, or run seqdesk demo-data install on the server. The Example Data & Provenance page lists exactly what it creates and distinguishes the synthetic local fixture from the pre-populated hosted demo and real public-read CI fixtures.

Logging in day to day

Sign in at /login. After authentication every user lands on the Sequencing Orders page.

Sessions are JWT-based NextAuth sessions carried in a cookie. There is no separate API token: scripts that need to call the API log in and reuse the same session cookie.

Built-in Help page

Every signed-in user can open /help, an in-app reference rendered from static content with no configuration needed. It covers the five-step workflow (create a sequencing order → submit samples → track sequencing → build studies → facility processing), the four main areas (Sequencing Orders, Studies, Analysis, Publishing), who usually does what, a task-to-location table, a facility-admin setup table, and status basics. It is the fastest way to orient a new user without leaving the app.

The two roles

User.role is a plain string with two values. There is no third role, no read-only auditor, and no per-study permission layer.

RESEARCHER — the default for self-registered accounts.

• Create, edit and submit their own sequencing orders
• Manage samples while the order is DRAFT
• Create studies and assign samples from orders they own
• See sequencing files once the facility publishes the delivery
• See a pipeline run once the facility marks it Final
• Sidebar: Sequencing Orders, Studies, Analysis, Support, Help

FACILITY_ADMIN — everything.

• Sees every sequencing order and every study in the instance
• The only role that can attach FASTQ files to samples
• The only role that can start, stop or delete a pipeline run
• The only role that can publish a delivery or mark a run final
• The only role that reaches /admin/** and the ENA Submissions surface
• Creates invite codes for further admins

The full capability matrix is in How SeqDesk Works.

Inviting users

Researchers self-register

Researchers register themselves at /register. The form collects first and last name, email (unique across the instance), researcher role (Principal Investigator, Postdoctoral Researcher, PhD Student, Master Student, Lab Technician, Other), department, optional institution, and a password of at least 8 characters.

Two things to configure first:

• Departments. The department dropdown is populated from /admin/departments. Create them before sharing the registration URL.
• Account Validation. Turning this module on lets you restrict registration to listed email domains. With enforcement on, an outside address is rejected with Registration is restricted to email addresses from: …; with enforcement off, the user is warned but allowed through.

Admins need an invite code

Admins cannot self-register — the registration API rejects a FACILITY_ADMIN signup without a code.

Manual setup, if auto-seeding fails

Rarely, you may need to run what the installer runs, from the active release directory:

cd "$HOME/seqdesk/current"
npm run db:migrate:deploy    # Apply pending Prisma migrations
npm run db:seed              # Seed default data

These are the same commands the /setup screen suggests. Restart the app afterwards — the bootstrap configuration is read from disk at start-up.

Next steps