conf 1.2.0 adds CONF_SECRETS, an env var to point at the secrets file
directly -- use it in the Docker entrypoint instead of symlinking the
mounted file into /app/conf/secrets.js, so the app no longer needs
write access to its own conf/ directory to pick up mounted secrets.
jq-repeat 2.2.0 is a compatible feature release (sort(), replace(),
faster leading-edge update() timing); no call-site changes needed.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
A repeated Edit-tool bump had overwritten the previous top version
heading instead of inserting a new one above it, silently merging
v1.1.10's release notes into v1.1.11's section with the v1.1.10
heading missing entirely. The underlying content was still present,
just missing its own "## [1.1.10]" header -- restored.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The new concept docs (and their "See also" reciprocal links) reference
each other by real filename -- "concepts-accounts.html" -- which is the
correct, working URL on the Jekyll/GitHub Pages build (a page's URL there
IS its filename stem), but doesn't match this viewer's own short slugs
(DOCS keys, e.g. "accounts" -> /docs/accounts), so fixDocLinks() left
those links unrewritten and 404ing in-app.
Rather than rewrite the docs to two different link forms depending on
target, resolve by filename as a fallback when the slug lookup misses --
one link written in a doc now works correctly on both targets.
Bumps to v1.1.13.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KDEx8ghuZR61pqPXc6da9C
- New docs/concepts-{accounts,oauth-apps,api-tokens}.md -- plain-language
guides aimed at less technical readers, each linking onward to the
existing schema/protocol-level doc for anyone who wants that detail.
Card help links (Users, Groups, OAuth cards, My groups, Members of
<uid>'s group) now point here instead of straight at the technical
docs; the LDAP-protocol-wiring cards (raw connection details for
connecting a 3rd-party app) stay pointed at the technical ldap.md,
since that's genuinely the right depth for that task.
- The "New API Token" card had no help link at all -- added, pointing to
the new API Tokens doc.
- Fixed the in-app docs viewer rendering every docs/*.md page with a
garbled heading + stray <hr> at the top: Jekyll front matter (meant
only for the GitHub Pages build) was never stripped before being
handed to the markdown renderer. Also fixed: cross-doc links
(ldap.html, index.html, etc.) never resolved in-app, since this
viewer serves docs at /docs/<slug> with no .html suffix -- rewritten
to the correct in-app URL, same idea as the existing image-path fix.
Bumps to v1.1.12.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KDEx8ghuZR61pqPXc6da9C
The single header-wide help icon (added last release) pointed at a
per-page doc guess, but a page can have several cards covering different
topics (e.g. Integrations has both OAuth and LDAP cards). Removed it and
added a small help icon directly to each card that has real corresponding
doc content, linking straight to that doc -- Invite User/Add new
user/User List/Service Accounts (users.ejs), group cards (groups.ejs),
OAuth Apps + LDAP connection cards (integrations.ejs), My groups/Members
of <uid>'s group/New API Token (profile.ejs).
Bumps to v1.1.11.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KDEx8ghuZR61pqPXc6da9C
- A ? icon in the top-right header deep-links to the doc most relevant to
the current page (client-side path mapping, same pattern already used
for top-nav active-link highlighting -- no server-side "current section"
local exists to key off of instead). Falls back to the docs index.
- GET /docs/search does a plain line-substring search over the existing
allowlisted doc set. No new dependency, stays usable with no internet
access.
Bumps to v1.1.10.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KDEx8ghuZR61pqPXc6da9C
Every account gets a personal posixGroup at creation (its primary GID
holder) but there was no way to manage its memberUid list -- add
add/remove endpoints and a profile-page UI (admin-only), reusing the
userSelect widget already built for the manager field.
Bumps to v1.1.9.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KDEx8ghuZR61pqPXc6da9C
- Edit form's Mobile Phone field was effectively required (stray validate
attribute) -- removed.
- Service account profiles always showed the literal filler name "Service
Account" -- hidden now, since it's not meaningful. Required computing
isServiceAccount in User.get(), not just listDetail().
- Fresh service accounts could look uncategorized (missing from the
Service Accounts tab, wrong isServiceAccount) for up to 5 minutes after
creation, due to a cache-staleness race in the create route -- the user
gets cached via User.get() before the route marks it as a service
account. Cleared and re-fetched after marking.
- memberOf came back as a bare string instead of a one-element array for
users in exactly one group, causing client-side permission checks to
iterate character-by-character and incorrectly deny access -- normalized
alongside the existing manager normalization.
- Added editable group membership on the profile page ("My groups"),
admin-only, using the existing per-group member endpoints.
Bumps to v1.1.8.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KDEx8ghuZR61pqPXc6da9C
Removes the LDAP bind-only service account type in favor of a single
Unix/POSIX account model, surfaced in a new Users > Service Accounts tab.
Adds a multi-valued `manager` field to every account (defaults to the
creator, editable, and grants edit rights on the accounts a person manages
without needing app_sso_admin). homeDirectory and loginShell are now
editable from the profile edit form.
Bumps to v1.1.7.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01KDEx8ghuZR61pqPXc6da9C
The GitHub Pages site used the generic jekyll-theme-cayman theme --
purple gradient hero, no site nav, no per-page SEO. Replaced with a
custom layout that mirrors the actual app UI: dark fixed navbar with
the theta42 logo, Bootstrap 5 + Font Awesome (same stack the app
uses), content in a card, dark footer matching bottom.ejs
(copyright, MIT license, GitHub, Changelog links).
- New cross-page nav (Home/Deployment/Configuration/OAuth/LDAP/
Changelog) -- there was previously no way to get from one docs
page to another except a single "Back to Home" link per page.
- SEO: jekyll-seo-tag + jekyll-sitemap (both GitHub-Pages-supported
plugins, no custom build needed) -- real per-page meta description,
Open Graph/Twitter card tags, canonical URLs, JSON-LD, sitemap.xml,
and a robots.txt referencing it. Added a real description to every
page's front matter (none existed before).
- Mobile: Bootstrap's responsive grid + collapsible navbar; the
screenshot pairs in index.md (inline width="49%" for a two-up
desktop layout) now stack to full-width below 576px instead of
squeezing illegibly small.
Verified with a real Jekyll build (jekyll/jekyll Docker image, no
Ruby available locally) + Playwright: desktop and mobile (375px)
screenshots of the home and deployment pages, mobile nav toggle
open/close, active-link highlighting per page, zero console/page
errors, and confirmed real SEO output (meta description, OG/Twitter
tags, canonical, JSON-LD, sitemap.xml, robots.txt) via curl against
the served site.
jq-repeat 2.1.0 (release notes: https://github.com/wmantly/jq-repeat/releases/tag/v2.1.0)
brings real fixes (throttled-update race conditions, sorted-list
reverse() leaking elements, nested-scope isolation) and a few
behavior changes. Audited every usage in this repo against the
changelog before upgrading:
- push()/unshift() now return the new array length -- every call
site in this repo is a bare statement, none consume the return
value. No risk.
- __setPut/__setTake, jr-order-reverse, nested jq-repeat templates:
not used anywhere in this repo (unlike proxy's companion PR, which
needed the __setPut/__setTake fix).
Real risk found and fixed: update() is now trailing-edge throttled
(~50ms) even on the first call, not just rapid subsequent ones.
profile.ejs's editUser()/editUserSeccess() call $.scope.editProfile
.update()/renderProfile() (which itself calls update()) and
immediately slideDown() the same element -- with the old synchronous
behavior the form was already populated by then; with throttling it
could briefly show stale/empty data. Deferred both slideUp/slideDown
pairs by 60ms (past the throttle window), per the library's own
migration guidance. Verified live (real bundled image + Playwright,
logged in as admin): the edit form's fields show real data, not
empty/stale, when checked right as the slide-open completes.
routes/oauth.js has its own pageLocals object (distinct from
routes/index.js's values and routes/docs.js's own copy) used by
oauth_authorize.ejs/oauth_logout.ejs -- missed in the white-label
change since a grep alias in this environment silently treats this
particular file as binary and skips it. Caught by CI (oauth.test.js),
not local testing. Added logo: conf.logo to match the other two
copies of this locals object.
conf.name was already plumbed into routes/index.js's values object,
but never actually rendered anywhere -- <title>, the navbar brand,
and the favicon were all still hardcoded "SSO - Theta 42"/"SSO
Manager". Now render <%- name %>/<%- logo %> in top.ejs; new
conf.logo key (default: the existing theta42.svg) drives the navbar
image and favicon.
Also fixes a pre-existing broken favicon: top.ejs referenced
/static/favicon.svg, which was never actually served from public/ --
only public/img/theta42.svg existed. The favicon now uses that same
file via conf.logo instead of a nonexistent path.
Footer copyright/logo/GitHub links are left as-is (open-source
attribution, not deployment branding).
The bundled Dockerfile.openldap image's own redis-server binds to
loopback only inside its container (no --bind override), so
Docker's -p 6379:6379 forward from the runner could never actually
reach it -- confirmed by the first real CI run failing with
"Socket closed unexpectedly" the instant the test process tried to
connect. Worked when tested locally only by accident: my override
env vars didn't actually take effect (model-redis's setUpTable only
reads a nested redisConf key, not flat host/port), so the app fell
back to createClient({})'s localhost:6379 default and happened to
hit my own pre-existing local Redis instead of the container's.
Fix: a dedicated redis:7-alpine GHA service container, which binds
correctly and is reachable at localhost:6379 -- matching that same
default, no env override needed.
- New GitHub Actions workflow: builds the real Dockerfile.openldap
image, starts it, seeds the LDAP fixtures the test suite expects
(uid 'test' + 'wmantly', matching the existing "wmantly is always
present in the test LDAP" assumption in several test files), then
runs the full Jest suite against it on Node 18/20/22. This repo
previously had unit tests but no automated workflow running them.
- Found while building this: the bundled default ppolicy entry
(docker-entrypoint.sh + ops/ldap-setup.sh) sets pwdLockout: FALSE,
which is backwards -- it silently makes the admin "deactivate user"
action a no-op for auto-lockout-after-failed-attempts (a related
but distinct ppolicy feature from pwdAccountLockedTime). Fixed to
TRUE in both places; ldap-setup.sh also gets a drift-correction
path so an existing deployment can pick up the fix by re-running it.
- Separately, deactivating a user still doesn't block their LDAP bind
in the bundled image even with this fix -- filed as #68, since it's
a deeper OpenLDAP ppolicy overlay question unrelated to the CI/test
setup here. tests/user_admin.test.js now soft-skips that specific
assertion (with a console warning pointing at #68) instead of
failing, so this known environment gap doesn't block CI.
GitHub Releases already carried real changelog notes per tag, but
those require internet access to view -- exactly what the /docs
route exists to avoid. CHANGELOG.md is a committed, Keep-a-Changelog
style file (backfilled from the v1.1.0/v1.1.1/v1.1.2 release notes),
linked from README and served at /docs/changelog alongside the rest
of the project's docs.
- Removed a dead IE<9-only html5shim script tag pointing at a domain
that no longer resolves.
- New GET /docs (index) and /docs/:slug routes render this project's
own README, DEPLOYMENT, API.md, docs/*.md, and directory_spec.md
server-side via marked -- so the documentation is readable from the
running app with no route to GitHub Pages, where it otherwise only
lives. Public, no auth, rate-limited (middleware/rate_limit.js) like
the other public routes.
- .dockerignore/Dockerfile.openldap updated to copy DEPLOYMENT.md,
API.md, directory_spec.md, and docs/ into the image, mirroring the
existing tos.md -> /tos.md convention.
tos.md was baked into the repo and read once at startup, so changing
the terms required a code change and deploy. It's now a Redis-backed
singleton (models/tos.js), editable from a new "Terms of Service" card
on the admin Dashboard, with the bundled tos.md used only as a
one-time seed for new deployments.
- routes/tos.js: GET (any authenticated user) / PUT (app_sso_admin
only) via /api/tos. Saving can optionally reset every user's
tos_accepted flag so they're asked to re-accept -- off by default,
since a wording fix shouldn't re-prompt everyone.
- routes/index.js: /tos and /onboarding now render the live content
instead of a module-level constant computed once at process start.
ops/backup.sh snapshots LDAP (slapcat), Redis (BGSAVE, dynamic RDB path
lookup), and ./config for standalone deployments, with retention. A
background service polls GitHub releases every 24h and surfaces an
admin-only banner in the UI when a newer version is published.
The Integrations page's Service Accounts (bind-only, organizationalRole)
don't cover the other real use case: an account something actually runs
as on a Linux host -- a media manager, a torrent client, Emby -- with a
real uidNumber/gidNumber that owns files, and a group other accounts
join for write access (e.g. a `stuff_manager` group granting write
rights to a media library). That needs a real posixAccount, which the
bind-only model can't be.
- New well-known group `app_sso_service_account`, seeded the same way as
app_sso_admin/app_sso_invite/app_sso_oauth_admin (docker-entrypoint.sh,
ops/ldap-setup.sh). Not a permission gate -- a marker.
- "Add new user" form gets a "This is a service account" checkbox: swaps
the person-shaped fields (first/last name, birthday, ToS agreement)
for a single account-name field, since none of those make sense for a
non-person account. On create, the route adds the user to
app_sso_service_account.
- User.listDetail() annotates each user with isServiceAccount (checked
against the marker group's member list once per call, not the memberof
overlay's reverse attribute -- not reliably returned by every LDAP
server this app might point at, confirmed against a real external
directory during testing). Users page shows a "service" badge.
- Notification broadcasts (filter_type=all/all_active) exclude service
accounts by default -- nobody reads mail as `stuff_manager`.
- Fixed a real, previously-unrelated bug this surfaced: addPosixAccount
unconditionally set `mail: data.mail` in the LDAP entry even when
undefined, and ldapts/slapd reject an attribute given an explicit
undefined value ("no values for attribute type") rather than treating
it as absent. This meant creating ANY user without an email already
failed outright -- not something a service account (which commonly has
no real mailbox) could route around. Made mail conditional, matching
how mobile/sshPublicKey/dob already work.
- docs/ldap.md now explains both kinds of service account side by side
and when to use which.
Verified end-to-end against a real external LDAP server (not a local
sandbox): created a service account with no email, confirmed it's
correctly flagged and excluded from broadcast recipient resolution,
confirmed a normal user is unaffected, confirmed the code degrades
gracefully if the marker group doesn't exist yet (pre-upgrade
deployments).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
They'd already drifted apart in wording after a recent edit only landed
in one of them, and docs/deployment.md wasn't linked from anywhere
(index.md's marketing rewrite dropped the old doc-links list) --
orphaned and duplicate. Replace it with a short pointer to
DEPLOYMENT.md on GitHub as the single source of truth, matching the
index.md "full docs live in the repo" pattern.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- OAuth Apps and LDAP Info are both "how do other apps/hosts plug into
this SSO" concerns -- merged into a single /integrations page with
tabs, replacing the two separate nav items with one. /oauth-clients
and /ldap-info 301-redirect there for compat.
- Add a Service Accounts section under the LDAP tab: bind-only LDAP
identities (organizationalRole + simpleSecurityObject, no
posixAccount) for apps/hosts, as opposed to real people. Create,
rotate password, and delete, all from the UI -- previously the only
such account (theta-env's bootstrap-created cn=ldapclient) was
invisible to the Users page entirely (filtered out by
conf.ldap.userFilter) and had no GUI way to see or rotate it; the new
ServiceAccount model uses the exact same objectClasses bootstrap.js
already creates cn=ldapclient with, so it recognizes and manages that
account too, not just ones created through this UI.
- The ldap-client bash snippet now points at "create one under Service
Accounts above" instead of a bare textual example.
Verified against a real LDAP server (not just the dev sandbox's usual
unreachable one): created a service account, confirmed it binds
successfully with the generated password, rotated its password, and
deleted it.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
It's mapped to the host by default for LAN/VPN direct-LDAP clients
(ldap-client hosts, apps with their own LDAP auth settings), not for
exposure through a router/firewall -- LDAP simple-bind has no rate
limiting the way the HTTP login endpoints do. Recommend a VPN for
remote direct-LDAP clients instead.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
New admin-only page (nav: "LDAP Info") that answers "what do I put in my
app's LDAP settings" without reading a doc: LDAPS URL, base DN, user/group
search bases, user filter, username attribute, and an example bind DN, all
derived from the running conf.ldap + request host rather than hardcoded --
so it's always correct for the actual deployment, copy-button on every
field.
Also generates a copy-pasteable bash snippet that clones
theta42/ldap-client and writes its ldap.vars file with the real host/base
DN/sso_url already filled in (bind password and SSO API token left as
placeholders with inline instructions, since those need to be created,
not derived).
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Adds a concrete "Connecting a 3rd-party app or container" section:
a bind-parameter reference table, a worked Gitea example, a generic
Docker LDAP_* env var pattern, and a pointer to theta42/ldap-client for
full host-level (SSH/sudo/PAM) integration as opposed to a single app.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The Compose form's "Send to" radio group had "All active users" checked
by default with no confirmation before Send -- anyone opening the
Dashboard to see how the feature works, typing a test subject/message,
and clicking Send would broadcast to every active user. Remove the
default (a target must now be explicitly chosen) and require a confirm
step before actually sending to "all" or "all_active".
Also add a hard safety net in models/email.js: Mail.send is a no-op
under NODE_ENV=test, so the automated test suite (which exercises the
real notification/password-reset/invite/OTP-by-email routes with
NODE_ENV=test) can never deliver real mail regardless of what recipient
list a test resolves.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- Rewrite docs/index.md as a short landing page (what it is, screenshots,
why this over the alternatives, features, a minimal "get it" snippet)
instead of a full documentation dump — full docs live in the repo
(README, docs/*.md) and are linked from here.
- Cross-link to Proxy and theta-env's own Pages sites.
- Screenshots are now clickable (open full size) on both the Pages site
and the README.
- Disable show_downloads in docs/_config.yml — the Cayman theme's
"Download .zip/.tar.gz" buttons are gone; "View on GitHub" (which links
back to the repo) is the only header link now.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Captured from a fresh theta-env install with demo data, via headless
Chrome + Playwright (scripted login, no manual UI interaction needed to
reproduce). Also adds a top-of-README Documentation link pointing at
GitHub Pages, matching theta42/proxy's README.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
It linked to fixed personal infrastructure (718it.biz, vm42.us,
git.theta42.com) with an admin's name in one label — not appropriate for
what should be a generic, publicly-releasable template. Flagged in
theta42/sso-manager-node#45, tracked as #46; resolves that issue.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The admin UI is a traditional multi-page app that loads ~13 separate
vendor/app JS+CSS files on every full navigation; none were compressed and
Cache-Control was max-age=0 (Express's default), forcing a revalidation
round-trip for every asset on every page view. Add gzip (compression
middleware) and sane Cache-Control (7d for vendor libs under
/static-modules, 1h for the app's own /static JS/CSS, which isn't
cache-busted). Matches the equivalent fix in theta42/proxy.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
theta42/proxy fronts an arbitrary number of hosts behind SSO, each with its
own callback URL (https://<host>/__proxy_auth/callback) — proxy's own code
comment already assumed "a wildcard redirect URI covers all", but no
wildcard matching existed here, so every proxied host's callback had to be
registered on the shared OAuth client individually or /oauth/authorize
would reject it with InvalidRedirectURI.
Add `*` (one hostname label) / `**` (any number of labels) wildcard support
to redirect_uri matching, e.g. `https://**.example.com/__proxy_auth/callback`
now covers every host proxy fronts under example.com. Exact matches still
work exactly as before.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
- Replace the separate Profile/API Tokens nav items with a single link
showing the logged-in user's name, pointing at their own profile.
- Merge admin.ejs + notifications.ejs into a new dashboard.ejs page.
/admin and /notifications now 301-redirect to /dashboard.
- Fold the API Tokens page into profile.ejs as a self-service-only
section, gated on isOwnProfile so it never appears when an admin
views another user's profile via /users/:uid. /api-tokens 301s to /.
- Fix: the section must not carry class="row" — app-base.js runs a
page-wide $('div.row').fadeIn() on every page load that would reveal
it regardless of the isOwnProfile check, since it fires before this
page's own gating logic.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Reported: creating any user via the API failed with
{"name":"InvalidSyntaxError","message":"gidNumber: value #0 invalid per syntax Code: 0x15"}
Root cause: addPosixGroup() computes the next gidNumber as
`Math.max(...groups.map(i => i.gidNumber)) + 1`. theta-env's
bootstrap.js creates the first admin via raw ldapadd with a hardcoded
uidNumber/gidNumber (10000) directly on the user entry, but never
creates a matching posixGroup entry -- so on a theta-env-bootstrapped
directory there are zero posixGroup entries, `Math.max()` on an empty
array is `-Infinity` in JS (not 0), and `-Infinity + 1` stringifies to
"-Infinity" -- an invalid LDAP integer, rejected by the directory. This
broke every single user creation, not just this one.
Separately: the reporter's intended scheme is for organically-created
users to start at uidNumber/gidNumber 1500, distinct from the
bootstrap admin's reserved 10000. Fixing the crash with a bare "floor
of 1500" alone wouldn't achieve that, since addPosixAccount's own
Math.max() would still find the admin's posixAccount entry (uidNumber
10000, found via a different, correctly-indexed search) and allocate
10001 for the next user.
Added a shared nextPosixId(entries, key) helper: takes the highest
existing value strictly below conf.ldap.uidGidReservedFloor (default
9000) plus one, or conf.ldap.uidGidMin (default 1500) if there are no
such entries. Ids at/above the reserved floor -- like the bootstrap
admin's 10000 -- are ignored entirely when computing the next
available number, so real users always start at 1500 and grow upward
regardless of the admin's reserved id.
Verified against a real theta-env deployment end to end:
- Reproduced the exact reported crash on a fresh bootstrap
- After the fix: first real user gets uidNumber/gidNumber "1500",
second gets "1501" -- admin's 10000 never enters the calculation
- New unit tests (nodejs/tests/posix_id.test.js, no LDAP required):
6/6 pass, covering the empty-array case, the reserved-floor
exclusion, and the NaN-from-missing-value case
- npm test: 18/18 passing tests still pass (unchanged); the other 155
failures are pre-existing/environmental (no LDAP server in this
sandbox) -- confirmed via git stash before starting this fix
* Fix commit hash not showing in Docker builds
build_info.js computed buildHash via `git rev-parse --short HEAD` at
runtime, but the final image intentionally has no git binary and no
.git directory (kept lean, per .dockerignore) — so this always failed
silently and the footer's version line showed "unknown" for every
Docker deployment. Working correctly only for bare-metal/dev, where
git + .git are actually present.
Added a throwaway gitinfo build stage that reuses the main base image
(no extra pull) with git installed just for this stage, reads .git
from the build context (now no longer excluded — see .dockerignore),
and bakes the resolved short hash into a small file that IS copied
into the final image. build_info.js reads that file first, falling
back to the old git-rev-parse behavior (still needed for bare-metal).
Verified against a real build: `docker exec sso-manager cat
/app/.build_commit` matches `git rev-parse --short HEAD` on the host,
and the footer now shows the real hash instead of "unknown". Same fix
already applied to proxy (theta42/proxy#133).
* Support GIT_COMMIT build-arg override for submodule builds
The gitinfo stage from the previous commit works for a standalone
clone (.git is a real directory) but not when this repo is built as a
git submodule (e.g. from theta-env): a submodule's .git is a pointer
FILE, not a directory — the real object database lives in the
superproject's .git/modules/, outside this repo's own directory and
therefore outside Docker's build context entirely. `git rev-parse`
can never resolve it from in here no matter what, so builds via
theta-env still baked in "unknown" despite the earlier fix.
Add an optional GIT_COMMIT build-arg that, when set, wins over the
in-context git resolution. theta-env's setup.sh now computes it on the
host (where the submodule DOES resolve correctly) and passes it via
docker-compose.yml's build.args. Same fix in proxy: theta42/proxy#133.
Verified via theta-env's actual setup.sh end to end: rebuilding with
this change, `docker exec sso-manager cat /app/.build_commit` now
matches `git -C sso-manager-node rev-parse --short HEAD` on the host
(previously: "unknown").
Auditing mobile/responsive views turned up two classes of bug:
1. Missing .table-responsive wrapper (token.ejs and profile.ejs's other
table already had it): admin.ejs (2 tables), invites.ejs (6 columns),
notifications.ejs (5 columns), users.ejs (7 columns — the worst case).
Without it these overflow the page horizontally on narrow viewports
instead of scrolling within the table.
2. groups.ejs's search/sort/count toolbar was a plain `d-flex` (no
flex-wrap) containing a search input, a <select> with a hard
min-width:175px, and a nowrap count span — on a narrow viewport
there's nowhere for that content to go but off-screen. Added
flex-wrap and gave the search input a flex-basis so it wraps onto
its own line first when space is tight.
Verified: EJS compiles for all five templates, npm test 192/192 pass,
and fetched each route from a running instance to confirm the fixes
are present in the served HTML.
* README: rewrite with feature overview, comparison, and tiered quick start
Expands the README with a fuller feature description, a "why this over
the alternatives" comparison against Keycloak/Authentik/Authelia/Zitadel,
an architecture diagram, and a three-tier quick start (unified theta-env
stack, standalone Docker, bare metal) instead of the old OpenLDAP-setup-
first structure.
* docs/index.md: fix stale setup.env quickstart snippet (.env -> setup.env, CFG_DOMAIN)
* docs: cleanup for public release (fix stale/wrong API docs, LICENSE, versions)
Documentation cleanup ahead of the public release announcement. Fixes a set
of confirmed issues from a prior audit:
- LICENSE: fill in MIT template placeholders (theta42, 2026).
- README.md: fix broken API docs link (api.md -> API.md), correct required
Node.js version (13.x -> 20.x), add the missing app_sso_invite group to
the LDAP groups table, scrub hardcoded dc=theta42,dc=com to the generic
dc=example,dc=com used elsewhere, add a "Recommended: Docker or
install.sh" section pointing to DEPLOYMENT.md/docs before the manual
OpenLDAP walkthrough, and drop an emoji from a warning callout.
- nodejs/api.md: deleted — it was a stale/legacy doc with wrong routes,
wrong request bodies, and endpoints that are dead/commented-out code.
The root API.md is the accurate, current reference; README now links
there directly.
- API.md: add the missing app_sso_invite permission group, fix the
documented invite response to match the real {token, link, mail_sent}
payload, document the previously-undocumented GET/PUT/DELETE
/api/user/invite endpoints, add the real allowed_groups field to the
OAuth client management examples, and document POST /api/oauth/authorize
(the endpoint that actually issues the code after consent).
- nodejs/routes/auth.js + API.md: fix "emaill address" typo in the
password-reset response message (source and docs kept in sync).
- DEPLOYMENT.md: fix the top-level summary to mention Redis, matching
docs/deployment.md and the entrypoint behavior it already documents.
Flagged, not changed: tos.md reads like a personal home-lab acceptable-use
policy (Emby/Gitea/Proxmox/Discord/Signal, first-person "the admin") rather
than generic OSS docs. Left in place pending a manual decision to
genericize, relocate, or remove it.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
* docs: genericize tos.md template, track runtime-editable terms in #39
Removes operator-specific references (Emby, Gitea, Proxmox, Discord,
Signal, first-person "the admin") so the shipped tos.md reads as a
neutral starting template rather than one operator's internal policy.
Actual runtime editability (admin/legal editing terms without a code
change) is tracked in issue #39, not implemented here.
---------
Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com>
The base DN (stack.ldapBaseDn) is the single place the domain is set; the
LDAP DNs (bindDN/userBase/groupBase) and oauth.issuer all derive from it and
must stay consistent. Drifting them apart (leaving bindDN at dc=example,dc=com
while ldapBaseDn is the real domain) makes the SSO bind against a non-existent
root DN and every login fails with Invalid Credentials.
- secrets.js.example: clarifying comment at ldapBaseDn
- DEPLOYMENT.md: "domain entered once as the base DN" note + theta-env
setup.env cross-link (merged the two duplicate theta-env blockquotes)
- README.md: cross-link to DEPLOYMENT.md from "Server set up"
Docs only; no app/secrets-structure change.
Co-authored-by: Claude <noreply@anthropic.com>
- api_tokens.ejs: created_on/last_used_on come back from Redis as strings
(model-redis only coerces fields with an explicit `type`), so `new Date(ms)`
yielded "Invalid date". Use `moment(ms, "x")` (the hosts.ejs/dns.ejs
precedent) which parses a numeric string-or-number as a Unix-ms timestamp.
- api_tokens.ejs: `isExpired` is a class getter not serialized to the client
JSON, so the "expired" badge never showed — compute expiry in the view via
`Date.now() > Number(expires_at)`. Also guard the `last_used_on: 0` / falsy
case (string "0" is truthy) so unset timestamps render "—" not "1970".
- middleware/auth.js: authIO did `checkToken(socket.handshake.auth.token || 0)`,
so any socket connect without a token (login page, pre-login) did an
`AuthToken.get(0)` lookup and logged a noisy `EntryNotFound` trace. Guard:
reject the socket with a generic 401 when there's no token (behavior-
preserving — unauth sockets were already rejected; just no Redis lookup / 404).
Co-authored-by: Claude <noreply@anthropic.com>
Personal access tokens so scripts/CI can call the management API without a
browser session. Each logged-in user mints their own token; it authenticates as
the creator (carries their LDAP group permissions, re-resolved live), so the
existing permission.byGroup checks apply unchanged.
- models/api_token.js: new ApiToken model (sso_<id>_<secret> format; id is the
lookup key, secret bcrypt-hashed + isPrivate, shown once). add()/rotate()/
authenticate(); optional expires_at; best-effort last_used_on. No _ttl
(persists; lifetime via expires_at).
- routes/api_token.js: self-service CRUD (list/get/update/delete/rotate),
owner-scoped (created_by === req.user.uid, 403 otherwise).
- middleware/auth.js + models/auth.js: accept `Authorization: Bearer sso_...`
(precedence over the auth-token session header); checkApiToken collapses
every failure to one generic 401 (no existence/secret/expiry leak).
- views/api_tokens.ejs + routes/index.js (GET /api-tokens): self-service page
(forceLogin, no group gate) — create (token shown once), edit, rotate, revoke.
- views/top.ejs: "API Tokens" nav entry visible to all logged-in users.
- public/js/app.js: app.apiToken client module.
- DEPLOYMENT.md + docs/deployment.md: API tokens section.
Co-authored-by: Claude <noreply@anthropic.com>
Lossless upgrades + config story for the all-in-one image.
Redis persistence (Part A):
- Replace in-memory `--save "" --appendonly no` with AOF + RDB persisted to /data
(--appendonly yes, periodic saves, --dbfilename dump.rdb). OAuth clients,
tokens, and other model-redis state now survive container recreation.
- Add the `sso-data` named volume -> /data in docker-compose.yml.
Config from ./config/sso-secrets.js (Part B):
- docker-entrypoint.sh: when /config/sso-secrets.js is mounted, symlink it to
/app/conf/secrets.js and read the server-side LDAP vars (base DN, admin pass,
org, domain, cert CN, JWT) from the file via one `node` call (base64-decoded,
no eval/quoting hazards). No app_* env is exported in this mode, so the file
is authoritative (@simpleworkjs/conf precedence: base < env < secrets.js <
app_* env). Falls back to the existing LDAP_* env-var mode when the file is
absent (standalone/bare-metal still works).
- docker-compose.yml: trim `environment:` to NODE_ENV/NODE_PORT only and add
`./config:/config:ro`. Removing the app_* env is required — any leftover
app_* would silently override secrets.js.
- secrets.js.example: add orchestrator-only `stack`, `bootstrap`, and
`serviceAccountPass` keys (ignored by the app; read by the entrypoint, the
theta-env bootstrap, and setup.sh).
Backup/restore docs:
- Full "Backups and restore" runbook in DEPLOYMENT.md (what lives where,
manual backup, full / Redis-only / LDAP-only restore, AOF-vs-RDB note,
upgrades). Restore uses slapadd -f (static slapd.conf), and RDB restore
requires deleting the AOF first (AOF wins on startup).
- Pointers in docs/deployment.md and docs/ldap.md; update the Docker Setup
section for the new ./config/ approach (env vars now advanced/optional).
Co-authored-by: Claude <noreply@anthropic.com>
The SSO app (nodejs/models/user_ldap.js addPosixAccount) tags every new user
with objectClasses [inetOrgPerson, sudoRole, ldapPublicKey, posixAccount, top,
theta42Person] and writes sudoHost/sudoCommand/sudoUser + sshPublicKey. The
all-in-one image's slapd.conf only included core/cosine/inetorgperson/nis +
theta42, so creating a user failed: sudoRole and ldapPublicKey were unknown
objectClasses (LDAP objectClassViolation 65) and sudoHost/sudoCommand/sudoUser
and sshPublicKey were unknown attributes.
Ship the two missing schemas and include them in slapd.conf:
- ops/schema/sudo.schema (sudoRole + sudo* attributes)
- ops/schema/openssh-lpk.schema (sshPublicKey + ldapPublicKey)
sudoRole is AUXILIARY here, not STRUCTURAL as in upstream sudo. The app
attaches sudoRole directly onto the user entry, which is already inetOrgPerson
(STRUCTURAL); two unrelated structural classes violate RFC 4512 and OpenLDAP
rejects with 65. AUXILIARY lets it coexist with inetOrgPerson — the app's
per-user-sudoers model. sudo's LDAP backend still finds entries via
(objectClass=sudoRole) regardless. ldapPublicKey is AUXILIARY as in upstream
openssh-lpk.
Also fix the build error from the previous theta42 schema PR: .dockerignore
excluded all of ops/, so 'COPY ops/schema/theta42.schema' failed at build
time ('not found' — file is git-tracked but stripped from the context). Re-
include ops/schema/*.schema with !exceptions, matching the existing
README.md/tos.md pattern.
Co-Authored-By: Claude <noreply@anthropic.com>
The app (nodejs/models/user_ldap.js) adds objectClass theta42Person whenever a
user has a dateOfBirth, and lists it among every new user's objectClasses. But
the all-in-one image's static slapd.conf only included core/cosine/inetorgperson/
nis — never the project's own theta42 schema — so slapd rejects
objectClass: theta42Person with LDAP 0x15 (objectClass: value #0 invalid per
syntax). Symptom: PUT /api/user/<uid> (and user creation) failing.
ops/ldap-setup.sh loads this schema for bare-metal (as a cn=config LDIF); the
Docker image uses slapd.conf, so ship it as a .schema file and include it.
- ops/schema/theta42.schema: dateOfBirth attribute + theta42Person auxiliary
objectClass (same OIDs/definition as ldap-setup.sh section 5).
- Dockerfile.openldap: COPY it to /etc/openldap/schema/theta42.schema.
- docker-entrypoint.sh: include it in the generated slapd.conf (after nis).
Co-Authored-By: Claude <noreply@anthropic.com>
Document how to get logs when running the all-in-one image: docker compose
logs for the app + slapd (both stdout/stderr, slapd runs -d 0), plus a direct
ldapsearch health check. Added to README.md, DEPLOYMENT.md (Method 1), and the
GitHub Pages docs/deployment.md.
Co-Authored-By: Claude <noreply@anthropic.com>
wmantly
