Commit Graph
154 Commits
Author SHA1 Message Date
Lauri Ojansivu 32c24ed84e Import page uploads stream to the server's disk, checked against the free space and an Admin Panel limit.
A file chosen on the Import page for the generalized importer's sources
is POSTed to /api/import/upload/:source and written to
WRITABLE_PATH/import-uploads a part at a time, instead of being read in
the browser and sent whole, as text or base64, through importBoard. The
size the browser announces is checked before the first byte against the
largest import set in Admin Panel / Settings / Security (none: no limit)
and against the free space of that disk less a reserve, and the space is
checked again every 64 MB while it is written; a refused or interrupted
upload leaves no file behind. importBoard then reads the file by its
id: a .zip export (Slack, Google Takeout, Vikunja, Notion, Plane) is
opened from the disk, its entries inflating up to the limit or the free
space, and a mailbox is read a message at a time, also inside a Takeout
zip. The upload is removed once read, or after an hour.

The HTML-only mail text is read through the parsed document rather than
stripped with a pattern. TODO.md says what is left of large files, and
which locale files hold another language.

Thanks to xet7 !
2026-11-20 05:42:27 +02:00
Lauri Ojansivu a8fd080bc8 Docs mirror the Admin Panel and the menus: a page for every menu entry.
The docs tree is laid out like the menus - a menu path is a docs path.
Admin Panel / Settings follows the new structure: Visibility.md is
Appearance.md, with new Features.md, Feature-Profiles.md, Security.md,
Activities-and-Notifications.md and Delete.md; the section indexes,
Admin-Panel.md, Admin-Panel-URLs.md (with the moved-pane redirects) and
Validation.md follow the live menus, and Who-Sees-What.md explains in one
place which layer decides what is shown and what is allowed.

Pages are added for the menu entries that had none, each written from
the code: every Admin Panel / Problems pane, the member menu, the board
header, Board Settings, the All Boards left menu, and the list,
swimlane and card menus. Older links to the moved panes are updated
across docs/.

TODO.md records what documenting the menus found: Edit Profile does not
save a username for non-admins, the Support page's signed-out case is
unreachable, Create board from this card is hidden from members who can
edit the card, All Boards Settings is unreachable, and the Share template
targets live in browser storage.

Thanks to xet7 !
2026-11-19 01:22:22 +02:00
Lauri Ojansivu e1a053d059 Admin Panel: one question per tab - Settings gains Appearance, Features, Feature profiles, Security, Activities and Delete.
Who sees and may use what was spread over three tabs: the feature
switches and feature profiles sat in Settings / Visibility beside the
logo and colours, and three settings (Security, Delete, Notifications)
sat in Problems among the reports, where they had landed when the
Features tab was removed.

- Settings / Visibility is Appearance (how this WeKan looks); its
  Features and Feature profiles groups are panes of their own. One
  template renders all three by `part`, so every field and Save is
  unchanged.
- Problems / Security, Delete and Notifications are Settings /
  Security, Delete and "Activities and notifications". Problems is
  Summary, Backup and the Reports.
- Every old address redirects (MOVED_PANES / movedPanePath in
  models/lib/adminUrls.js, applied by config/router.js).
- People, Teams and Organizations show the feature profile that
  applies, so a profile can be found from either side.
- Two new labels, Appearance and Activities and notifications, are
  translated into every language.

tests/adminPanelStructure.test.cjs and admin-panel-structure.e2e.js
cover the order, the redirects and that no feature switch is left in
Appearance or Problems; older guards and specs follow the new panes.

Thanks to xet7 !
2026-11-19 01:22:13 +02:00
Lauri Ojansivu cfd9a62466 Swap alone is not high RAM usage, and resource monitors are not named like vulnerabilities.
A developer laptop with many programs open showed in Admin Panel /
Problems / Memory usage: "RamBleed, high, high RAM usage started
(>= 90%): RAM 16.1/30.9 GB (52%), swap 18.7/30.9 GB (60%)". Swap at or
above 25% started a high-memory period by itself, made it high severity,
and the text still said ">= 90%". Idle pages in swap while RAM is free
are not pressure.

- Swap now counts only while RAM is also at or above the low threshold
  (80%), and a period ends when RAM is back under it.
- The start row says what started it: (RAM >= 90%) or (swap >= 25% with
  RAM >= 80%). Only RAM at 95% or more is high severity.
- The RAM, CPU, disk and speed monitors named their rows like Hall of
  Fame entries (RamBleed, CpuBleed, DiskBleed, SlowBleed) with CWE-400.
  They are HighRamUsage, HighCpuUsage, HighDiskUsage and SlowRequest now,
  without a CWE, and stored rows are renamed once at startup.

tests/ramHighTracker.test.cjs replays the laptop's numbers as a
negative case; tests/resourceStreamNames.test.cjs pins the names.

Thanks to xet7 !
2026-11-19 00:21:33 +02:00
Lauri Ojansivu fc21b0b177 Feature profiles: an administrator hides features from chosen users, teams and organizations.
Decided by the maintainer on 2026-11-18 (TODO.md): administrator-defined
profiles. Admin Panel / Settings / Visibility / Features gains Feature
profiles: a name, a tick for each optional feature the profile hides,
and the users, teams and organizations it applies to. A user's own
profile wins, then the first profile naming one of their teams, then one
naming one of their organizations.

Strictly opt-in, because hiding features by default was tried and
removed as a support nightmare: a user with no profile sees everything,
a profile only hides (no data or permission changes), and the member
menu tells the user which profile hides how many features. The decision
is models/lib/featureProfiles.js, applied in isFeatureAvailable, so
every existing place that hides a disabled feature follows it.

Profiles are stored in Settings (featureProfiles) and a direct
assignment on the user (featureProfileId, published to that user only
and not writable by them). tests/featureProfiles.test.cjs covers the
decision, cleaning, the admin-only methods and the wiring;
feature-profiles.e2e.js drives the page, the hidden board view and the
notice.

The 11 new strings are translated into every language. Low confidence:
 qu, rup, rn, rw, sah, se, sm, sn, ss, st (low); sd, so, si, scn (medium); sr uses standard Тимови/Организације; ff, fj, gn, gv, haw (low); ha, hsb, fur, gd, ga, fo (medium); ace, ak, ay, bm, bi, bo, bua (low); ba, bho, ar-DZ/ar-EG/ary dialect spelling (medium); iu, kl, ks, kw, lld, lg, ig, la (low); kok (medium); tig, ti, tlh, ve-PP, ve-CC, uz-AR, ve, ts, tn, to, tpi (low); szl (medium); wal, vo, zgh, wa, wo, wa-RR, yi, wuu-Hans (low); chr, dz, ee, csb, cv, co, ckb (low); cmn uses 配置 for profile; nah, nap, nd, nso, om, mai (low); mg, ny, pap, or_IN, ps (medium).

Thanks to mimZD and xet7 !

Fixes #6837.
2026-11-19 00:01:42 +02:00
Lauri Ojansivu 579de7d6af The Cloud Panel is Weddy's, not WeKan's: WeKan stays Meteor 3.
Admin Panel / People / Cloud and the Cloud Panel (73db2a28c3) are removed from
WeKan, with their methods, route, menu item, docs and tests. Many domains from
one Weddy is a Weddy feature: there every domain's WeKan is in one process,
where Node.js WeKan containers would have to work together, which gets
complicated. It is written in Go in caddy-wekan-plugin.

The translations of its strings stay: WeKan, Weddy and the rest use the same
WeKan translations.

Thanks to xet7 !
2026-11-08 00:38:40 +02:00
Lauri Ojansivu 73db2a28c3 Cloud Panel: many WeKan domains from one Weddy, for the SuperAdmin at the Cloud Panel Domain.
Weddy (wekan/caddy-wekan-plugin) runs a WeKan for every domain, each in its own
directory - files/domains/<domain>/files/db/ for its database - and starts
the first domain's WeKan with WEDDY_CLOUD_DIR, every customer's with
WEDDY_TENANT=true.

- Admin Panel / People / Cloud, only in that first WeKan and only for a site
  admin (a per-tenant admin's whitelisted menu never has it): Cloud Panel
  Domain, e.g. cloud.wekan.team, and SuperAdmin Username, a username or
  e-mail address.
- /cloud, the Cloud Panel, where the SuperAdmin is taken at the Cloud Panel
  Domain: Settings with the left menus Domains and CloudFlare, in the Admin
  Panel's layout. Domains is a table page - Domain, Customer name, Customer
  email address, Billing email address, Start Date, End Date and the TLS
  Weddy got - with a form to add and edit. CloudFlare is a write-only API
  key field.
- server/methods/cloudPanel.js checks in every method again that the caller is
  the SuperAdmin and the request's Host is the Cloud Panel Domain - not
  X-Forwarded-Host, which a client could send itself. The collections are not
  published. Every change writes WEDDY_CLOUD_DIR/domains.json for Weddy; the
  key goes to cloudflare.json, mode 0600, and is never sent back.
- models/lib/cloudPanel.js has the rules, shared by client and server; its
  domain rule is the same as Weddy's, since the name becomes a directory.
- Customers' WeKans show none of it.

The 15 new strings are added as pending Transifex, written in Finnish, and
translated into every other locale. Low-confidence translations: ace ak an ay
ba bho bi bm bo bua chr co csb cv dz ee ff fj fo fur gn gv haw hsb ht ig iu
kl kok ks kw lg lld mai mi nah nap nd nso oc om pap qu rm rn rup rw sah sc scn
se sm sn so ss st szl tig tlh tn to tpi ts uz-AR ve ve-CC ve-PP vo wa wa-RR
wal wo wuu-Hans yi zgh. Transifex translations replace them when they arrive.

tests/cloudPanel.test.cjs: the rules with their negatives, and the security
properties the source keeps. The node suites pass, 2160 of 2160. The
templates are not compiled here (no Jade compiler in this sandbox); a Meteor
build checks them. docs/Platforms/FOSS/Weddy/Cloud-Panel.md is the design.

Thanks to xet7 !
2026-11-07 23:32:43 +02:00
Lauri Ojansivu c99aa7f845 Admin Panel / Problems / Password length, and a changed password signs every session out.
PASSWORD_MIN_LENGTH (default 8) is now the length below which a sign-in
is REPORTED, in the new Admin Panel / Problems / Password length, and
PASSWORD_MIN_LENGTH_ENFORCE (off by default) is what makes it a rule: a
shorter new password is refused, and an account that signs in with an
older shorter one sees only "Change your password" until it has a long
enough one. Signing in is never refused, enforced or not, so nobody is
locked out by a minimum set after their password. LDAP, OAuth2, OIDC,
SAML, CAS and Sandstorm accounts are neither reported nor asked: their
password is their provider's. An administrator's Set password no longer
applies WeKan's policy to such an account.

Meteor sends a password hashed, so only the browser knows its length: it
notes it on the sign-in form and, after a sign-in that worked, says "it
was short" - never the password or its length.

The page has both settings and a table, one row per account: username,
first and last short-password login, the latest IPv4 and IPv6 address
with the country's flag and city when a CDN says where it is, the count,
and when it was fixed, with the Too short (default) / Fixed / All filter.
Fixed is a changed password not seen short since; a short one seen
again reopens the row.

Changing one's own password - Member Settings, the forced change, a
reset link - now signs EVERY session of the account out, the one that
changed it too, and only the new password signs in again. The browser
goes to the sign-in page as soon as it is signed out, since the server
closes the connection before the answer can arrive.

Documented in docs/Features/Admin-Panel/Problems/Password-length.md, the
settings on every platform. The new strings are in every locale in
English for now. Tests: passwordPolicy, passwordChangeRequired and
shortPasswordLogins (node); password-min-length-signup-login,
password-length-report and password-change-other-sessions (Playwright,
Chromium and WebKit), which change each setting on the page and check
what it does, and that sign-up with no mail server registers and signs
the user in.

Thanks to xet7 !
2026-10-31 14:54:22 +02:00
Lauri Ojansivu f165a24192 docs: Admin Panel / People / Audit log.
Thanks to xet7 !
2026-10-31 00:44:14 +02:00
Lauri Ojansivu 2017f2871f Audit log: its own Admin Panel URL and title, and right-to-left table alignment.
/admin/people/audit-log opens the pane and names it in the title bar like
every other People pane, and the table aligns to the start of the line,
so it reads right to left in Arabic, Hebrew, Persian and the rest.

Thanks to xet7 !
2026-10-31 00:32:33 +02:00
Lauri Ojansivu b4e5f90afb Card links never launch an OS handler with a code-execution history.
Custom URL schemes in card text link only when Admin Panel -> Features ->
URL lists them, and never javascript:, data: or another scheme that runs
code in the page. A listed operating-system handler was linked, though, and
some have a public code-execution history: ms-msdt (Follina,
CVE-2022-30190), search-ms and search, ms-officecmd, ms-appinstaller,
ms-cxh and ms-cxh-full, and telnet - the last step of the 2026 Edge full
chain, where Windows ran telnet.exe from a user-writable PATH directory.
An administrator listing one would let any board member plant a link that
runs something on whoever clicks it.

- They are never linked, listed or not, by the app's parser and by the
  markdown package's copy of the rule (the two lists must be identical),
  and a click on one does nothing should such a link reach the page.
- A card title or description, comment, checklist or checklist item that
  ADDS a link to one is shown in Admin Panel -> Problems (HandlerLink,
  CWE-939), with who wrote it; the text is saved as written and the writer
  is told nothing. The word in prose, or a link the text already had, is
  not recorded.
- docs/Features/Admin-Panel/Settings/Visibility.md lists both kinds of
  never-linked scheme, and says what a listed file: does on Windows (an SMB
  link outside the network sends the clicker's NTLM hash).

Tests: tests/urlSchemeAllowlist.test.cjs (positive and negative, through
the real DOMPurify, both sanitizer passes and the detector; fails on the
old code), and custom-url-schemes.e2e.js, which passes in Chromium and
WebKit: listed handlers do not link, a description saved with one through
the editor is recorded for its writer, and a kept link or a comment that
only mentions the word is not. Firefox does not start in this
environment.

Thanks to xet7 !
2026-10-30 01:19:16 +02:00
Lauri Ojansivu 9dc3fc3f31 Data health: custom fields whose boards were all deleted get their own check.
customfield-dead-board counted every custom field that names a deleted
board, but its Repair only removes dead boards from a field that is also on
a live board; a field whose boards were ALL deleted was skipped on purpose,
since removing them would leave it on no board. So those fields stayed in
the count with a Repair button that could never change it: one instance
showed 510 and "Repaired: 0" after every click.

- New check customfield-boards-deleted: fields whose boards no longer exist
  (through boardIds or the legacy boardId). No existing board shows them.
  Repair deletes the definitions after the confirming second click, like
  the other repairs that delete data. A value a card on a live board still
  holds is left alone and stays listed under card-dead-field-values.
- customfield-dead-board counts only fields that still have a live board,
  so its count goes to zero when its Repair has run.

docs/Features/Admin-Panel/Problems/Data-Health.md lists the new check.
Test: tests/dataHealth6818.test.cjs.

Thanks to Raymond-Voogt and xet7 !

Fixes #6829.
2026-10-28 13:50:49 +02:00
Lauri Ojansivu 30e960e129 Docs: Show empty fields of Card Settings, the [label] title switch of Member Settings, and Data health repairs.
Thanks to xet7 !
2026-10-27 19:42:20 +02:00
Lauri Ojansivu d9cf523f9e Admin Panel / Problems: Data health, a nightly scan of broken references, with repairs.
The scan counts what points at nothing: cards on a missing or another board's
list or swimlane, ghost and duplicate board members, boards with no active
admin, custom field values and boards that no longer exist, checklist items
without a checklist, and subtasks or linked cards whose card is gone. It runs
nightly when the CPU is low, claimed so that only one pod scans or repairs.

A repair that only moves data reuses the upgrade steps; one that deletes asks
twice, and the server refuses it without that confirmation. Every repair edits
with $pull or a guarded update, so a concurrent edit is not overwritten, and a
member dedupe keeps the admin entry. Legacy scalar boardId and string boardIds
of custom fields are understood.

Translations done with low confidence: ay, csb, lld, kok, wa-RR, ve-CC, bi, tpi.

Thanks to Raymond-Voogt and xet7 !

Fixes #6818.
2026-10-27 17:08:15 +02:00
Lauri Ojansivu 567f99227b A failed import or export disables an account only when it was an attempt.
secureTransfer recorded every throw as a high "blocked" event naming the
user, and that disables the account. sanitize-html 2.18.0 threw
RangeError on nested <iframe>, so a card planted with the payload locked
out the next member who exported the board; any library fault would do
the same. The transfer boundary now marks its own refusals
(TransferBoundaryError: a cycle, a non-plain object, a size limit), and
models/lib/transferFailure.cjs records only those as blocked - high, and
medium for a size limit as before. Anything else is ImportExportFailure:
detected, still naming whose import or export it was, and never blocking
an account.

tests/transferFailure.test.cjs drives real boundary refusals and a
sanitizer that throws through blockOnSecurityEvent's own decision; with
the old secureTransfer it fails. It also fails if the boundary throws a
plain Error or anything but secureTransfer calls it.

Thanks to xet7 !
2026-10-22 04:56:40 +03:00
Lauri Ojansivu e0189bc53b Meteor 3.5.3's six security fixes on Meteor 3.6-rc.0.
Meteor published the same six fixes as 3.3.3, 3.4.2 and 3.5.3 on
2026-10-16. 3.6-rc.0, which WeKan runs, does not have them: its tag is
behind 3.5.3 by exactly the 15 security commits, and neither it nor
Meteor's devel branch has the fixed code. Going back to 3.5.3 would also
mean Node.js 24 and Rspack 1, so WeKan carries the fixes in packages/,
where a package overrides the release's own: 3.6-rc.0's allow-deny,
ddp-server, dynamic-import, facebook-oauth, google-oauth and shell-server
with Meteor's 3.5.3 patch applied unchanged, each marked with a _1 wrap
number. ddp-server leaves out uWebSockets.js, as Meteor's published
package does; it is only for the opt-in uws transport and came from a git
URL.

- DDP: a message named __proto__, toString or not a string, and an async
  handler that rejected, crashed the server.
- allow/deny: a top-level __proto__ in an update was missing from the field
  list the validators saw.
- dynamic import(): requests are bounded in size (413) and complexity
  (400).
- Google and Facebook direct-token login: a token issued to another
  application was accepted; identity now comes from the provider, and a
  Facebook verification error no longer carries the app secret.
- meteor shell: its history and compile cache are private.
- Deploy discovery is in meteor-tool, which an app cannot replace; WeKan is
  never deployed with meteor deploy, and a test keeps it so.

Refused attempts are in Admin Panel / Problems as Meteor DDP, Meteor
allow-deny, Meteor dynamic-import and Meteor OAuth, through a hook the
packages call (server/lib/meteorSecurityHook.js); none blocks an account.
An oversized import() body, an unknown ordinary DDP name and a network
failure are refused without a row, because ordinary use can reach them.

tests/meteor353SecurityFixes.test.cjs runs each fix's code cut out of the
fork and searches the tree for the shape of each fault; all of them fail on
the unfixed 3.6-rc.0 packages. It also fails when .meteor/release moves
on, because a fork would silently keep 3.6-rc.0's code.
tests/playwright/specs/meteor-353-security.e2e.js sends each attack to a
running WeKan built with the forks. The forked files are recorded in the
source audit baseline, and the suites that check WeKan's own methods and
subscriptions skip Meteor's package tests.

Thanks to Meteor developers, Vext Labs and xet7 !
2026-10-22 04:12:13 +03:00
Lauri Ojansivu 4d8df83fdd Admin Panel validation audit lists the Retention pane.
tests/adminPanelDocsMatchMenu.test.cjs requires docs/Features/Admin-Panel/Validation.md
to inventory every live pane; Attachments / Retention (#6789) was missing
from it. It also crashed on the commit before this one, whose test run was
read by its last line only.

Thanks to xet7 !
2026-10-20 14:05:36 +03:00
Lauri Ojansivu 5c2fc294f3 Tests and docs follow this round's changes: Retention documented, a real regex escape, the snap guards.
The full node suite found three breakages in this round's own work:
- adminPanelDocsMatchMenu: Admin Panel / Attachments / Retention was in the
  product and not in docs/Features/Admin-Panel/Attachments/README.md. It has
  its section and its row in the URL table now.
- incompleteEscaping: the test from the #6788 patch hand-wrote a regex escape.
  It now uses the same escape-string-regexp package the server imports.
- releaseSnapArches: two guards read the Launchpad step's text. infra_failure()
  (the v12.56 Launchpad-outage retries) also reads snapcraft-wekan-*.txt
  before the loop, and the fixed `sleep 60` became `sleep "$pause"`. The
  guards now look for the log printing after the no-snap message, and for the
  pause variable, and say why. These two had been failing since that change.
  An earlier run's report that this suite was only "load-related" was wrong:
  the FAIL lines had been filtered out of the output that was read.

Thanks to xet7 !
2026-10-20 14:05:13 +03:00
Lauri Ojansivu 06b7ce994a Backup status: every run recorded, failures and overdue backups shown, old local backups pruned - in Admin Panel too.
The result of an Admin Panel backup lived only in the memory of the server
process that ran it. After a restart, or on another pod of a multi-pod
deployment, the Backup pane said "idle", so nobody could see whether last
night's backup had worked. There was no history, no warning when a scheduled
backup failed or did not run, and local backups were never cleaned up.

- Every run, scheduled or "Backup now", is recorded in backupRuns: start and
  end, who or the schedule, scope, storage, status, size, duration and error.
  Records are kept 400 days. A run still "running" after 24 hours counts as
  failed, and a scheduled run skipped for another backup is recorded as failed.
- Admin Panel / Problems / Backup shows the last success (overdue when its
  period has passed), the last failure with its error, the next run and the
  last 30 runs. The same status is above the schedule in Attachments / Backup.
- Problems / Summary lists "Backup failed" and "Backup overdue". /metrics adds
  wekan_backup_last_success_timestamp_seconds,
  wekan_backup_last_run_timestamp_seconds, wekan_backup_last_run_success and
  wekan_backup_overdue for alerting.
- BACKUP_KEEP_DAYS: after a successful backup, local backups of the same scope
  older than N days are deleted. The newest is always kept, and only a
  directory inside the backup root holding a backup.zip is removed.

From the patch by Raymond-Voogt, and in Admin Panel as well:
- BACKUP_KEEP_DAYS is the catalog's backup section
  (models/lib/authConfigCatalog.js), overridable beside the schedule in
  Admin Panel / Attachments / Backup and resolved through authEnv at every
  backup.
- It is listed on every platform: the compose files, start-wekan.sh and .bat,
  the snap config and help, and the Dockerfile, which the patch's version
  had missed. Its platform hunks, written before #6789, were re-applied by
  hand.

tests/backupRuns.test.cjs (21): the run records, overdue and pruning against a
real directory, and the Admin Panel override, including that nothing reads the
variable around it. tests/authConfigCatalog.test.cjs checks every platform.
admin-attachment-retention.e2e.js checks the Backup pane shows the field.

Thanks to Raymond-Voogt and xet7 !

Fixes #6790.
2026-10-20 13:30:00 +03:00
Lauri Ojansivu dcf91e4471 Attachment retention: delete card attachments after N days, purge deleted ones after M days - in Admin Panel too.
Organisations that must not keep files forever (storage limits, GDPR data
minimisation) had only scripts working on MongoDB directly, which leave card
references behind, do not know which storage holds a file, and break on
upgrades. Soft-deleted attachments were also kept forever.

A nightly job (server/attachmentRetention.js, rules in
models/lib/attachmentRetention.js), off unless configured:
- ATTACHMENTS_RETENTION_DAYS soft-deletes card attachments older than N days
  through the card's own Delete (cover unset, activity, card history Restore).
- ATTACHMENTS_RETENTION_MODE: uploaded (default), or card-inactive, which also
  requires the card to have had no activity for as long.
- ATTACHMENTS_PURGE_DELETED_AFTER_DAYS removes attachments soft-deleted at
  least M days ago for good, file included, through Attachments.removeAsync
  so every storage backend is cleaned. The write re-checks deletedAt, so one
  restored meanwhile stays.
- ATTACHMENTS_RETENTION_DRY_RUN only logs; ATTACHMENTS_RETENTION_MAX_PER_RUN
  caps each step per night.
Never touched: attachments without a card, template boards and cards, and
attachments with an unknown upload date.

From the patch by Raymond-Voogt, and in Admin Panel as well:
- The five settings are a section of models/lib/authConfigCatalog.js.
  Admin Panel / Attachments / Retention (/admin/attachments/retention) draws
  the shared override form on it: a value there wins, an empty field falls
  back to the variable, and the source in effect is shown.
- The job resolves them through authEnv on every run, so a change needs no
  restart.
- The new menu label attachment-retention is translated in all 246 locales.
- The catalog's platform test then found that the patch had missed the
  Dockerfile and the snap help; both list all five now, beside
  docker-compose, start-wekan and the snap config.

tests/attachmentRetention.test.cjs (14) covers the rules and the Admin Panel
wiring; negative: the job never reads the environment around the Admin Panel.
tests/authConfigCatalog.test.cjs checks every platform lists each variable.
tests/playwright/specs/admin-attachment-retention.e2e.js opens the pane, and
checks that an ordinary user cannot.

Thanks to Raymond-Voogt and xet7 !

Fixes #6789.
2026-10-20 13:28:17 +03:00
Lauri Ojansivu e76f12bcdd docs/Features: guides for the eleven helpers that need no AI, filed by menu, and how each is hidden.
Thanks to xet7 !
2026-10-17 13:57:56 +03:00
Lauri Ojansivu 2b8a33022b Problems / Instrumentation: the whole picture as one bar per area of WeKan.
The page listed up to 200 methods and publications on 11 pages. Above the
table there is now one bar per area - Settings, Login, People, Boards,
Cards, Lists, Notifications, Attachments, Problems and the rest - its length
the area's calls against the busiest area's, its red end the share that
failed, with the calls, the average duration and the errors beside it.
Clicking an area shows only its operations in the table; clicking it again
shows all.

models/lib/instrumentationAreas.js groups an operation by the first area
whose pattern matches its name, and weights an area's average duration by
completed calls. tests/instrumentationAreas.test.cjs checks the grouping
against all of WeKan's 614 methods and publications - "Other" holds three -
and the summary arithmetic. The labels are existing translation keys, apart
from the new instrumentation-other ("Other"), translated into every
language. The Finnish "cards" heading read "Korttia" (the form after a
number) in every place it is used as a heading or label; it is "Kortit".

The bars are HTML and CSS, one grid with subgrid rows so names, bars and
numbers line up; the viewport matrix checks the page at every device size.

Thanks to xet7 !
2026-10-17 06:03:55 +03:00
Lauri Ojansivu 8c3e80240a Shared labels: the board labels' colour picker, at the size it has on a card.
Admin Panel / Settings / Shared labels chose a colour from a dropdown of
colour names. The board label popup's picker - the palette swatches and the
custom colour wheel - is its own template now, labelColorPicker, and the
popup and every Shared labels row use it; a custom colour is saved as its
hex, which the server already accepted. A pick reaches the row through an
object with set(), because Spacebars calls a function argument instead of
passing it.

The swatches grow to fill their row, so on the wide admin page their width
followed the window. There they have the size they have in an opened card's
label popup, 54 x 33px, at every window width.

tests/sharedLabelsColorPicker.test.cjs pins the shared picker and the size;
tests/playwright/specs/shared-labels-color-picker.e2e.js picks a swatch and a
custom colour, saves both, and measures the swatches at 1366, 1920 and
2560px.

Thanks to xet7 !
2026-10-17 05:32:38 +03:00
Lauri Ojansivu b6a3d98736 Admin Panel / Problems: Memory Usage and Disk usage reports.
The RAM and Disk usage pages were designs that no code carried out. Both
monitors now run beside the CPU one and, like it, record only the start and
the end of a sustained high-usage period:

- Memory: RAM from /proc/meminfo's MemAvailable - os.freemem() is MemFree,
  which counts the page cache as used and shows a healthy server as full -
  and swap, which alone also starts a period. Without /proc/meminfo it uses
  what Node reports. Settings WEKAN_RAM_* and WEKAN_SWAP_HIGH_PERCENT.
- Disk: each filesystem WeKan writes to - the files root, its temp directory
  for uploads and the snap's data directory - measured once per device, with
  an episode of its own. WEKAN_DISK_WATCH_PATHS replaces the paths.
- Each report has a live line above its rows and a URL
  (/admin/problems/ram, /admin/problems/disk). The menu and titles use the
  existing memory-usage and sandstorm-disk-usage keys, which every language
  translates, and the Summary names the cpu, ram and disk streams.

Observe and report only, as designed. The three monitors share one event
recorder (server/lib/resourceLog.js) and one intEnv (models/lib/intEnv.js),
which three files had each defined. The pure halves are tested without
Meteor, and resourceMonitorsWiring checks the wiring and that no copy of
either shared piece returns.

Thanks to xet7 !
2026-10-17 01:14:16 +03:00
Lauri Ojansivu b7589737f5 File docs/Features the way the WeKan menus are.
A feature's guide now sits under the menu it is used from, read from the
templates rather than from where somebody once put it: Header-Bar,
All-Boards, Pages, Board-Header (search and every Board View), Board-Sidebar
(Board Settings, Filter, Members, Labels, Archive), Rules, Swimlanes, Lists,
Cards, Member-Settings, Admin-Panel and Login-Page. Topics used from several
menus (editor, history, theme, translations, email, troubleshooting) are in
General.

- Each area has a README mapping its menu entries to their guides and to the
  template or server module that implements them. Board-Header/README.md
  lists all 40 Board View entries with the template boardBody.jade renders;
  five views (Table, Calendar Mode, Time, Timeline, Map) have no guide of
  their own yet and say so.
- Board-View.md listed 25 views while BOARD_VIEWS has 40; it now lists all
  of them with their English labels.
- Pages keep their file names. Where a released changelog links an old path,
  a "# Moved" page stays there and points to the new one; test-menu.sh skips
  folders holding only those.
- Undo was filed under Login; it is General/Undo now.
- Every link in docs, code comments, scripts and tests follows the move;
  docsLinksResolve checks them all.

Tests that read a doc follow it to its new place. boardViewSettings now fails
when Board-View.md or Board-Header/README.md differs from BOARD_VIEWS, or
names a template boardBody.jade does not render.

Thanks to xet7 !
2026-10-16 23:45:35 +03:00
Lauri Ojansivu 70dcc2a134 Remove files nothing uses, and point their tests at the code that runs.
Found by following every import, re-export and require from client/main.js
and server/main.js (meteor.mainModule: only what that reaches is loaded) and
each local package's package.js:

- packages/wekan-fullcalendar/fullcalendar/ (88 files) and update.sh: the
  vendored FullCalendar v3 and the script that updated it as a git
  submodule; the package takes FullCalendar 5 from npm (Npm.depends).
- packages/wekan-fontawesome/fontawesome-free: js/, sprites/, metadata/,
  attribution.js and the .min.css copies; package.js adds only the plain CSS
  and the webfonts. Its package.json no longer names the removed files.
- models/lib/propagateMembers.js and models/lib/teamBoardMemberSync.js:
  replaced on 2026-10-14 by models/lib/groupBoardMembership.js (memberships
  that record their team or org, #4178), which the server uses. Their tests
  exercised the old modules; the same scenarios now run against planGain /
  groupChanges, and the wiring checks are kept.
- models/lib/recoveryPlan.js and its test: a model of the SQLite recovery
  policy that nothing ran; releases/ferretdb/sqlite-recovery.mjs carries it
  out and tests/sqliteVerifiedRecovery.test.cjs tests that. The Recovery
  design page says so.
- packages/wekan-ldap/server/configResolver.js: a vendored copy nothing
  imported - the LDAP package reads its settings through the app's
  accessor. Its sync test is now a check that no package carries such a
  copy; three comments that sent readers to it explain the reason instead.

docs/Design/TeamsOrganizations.md also said losing a team removes nothing;
since #4178 a team takes back what it gave.

Thanks to xet7 !
2026-10-16 22:54:48 +03:00
Lauri Ojansivu 4234081752 WebDAV attachment storage: attachments in Nextcloud Files or any WebDAV server.
Admin Panel -> Attachments has a WebDAV (Nextcloud Files) pane beside S3,
Azure and Google Cloud Storage: a folder address, a user and an app
password, with the same read and write switches, connection test, default
storage choice and Move Attachment target as the others. Files are stored
with PUT, read with GET and removed with DELETE; the folders are made with
MKCOL as needed, and the connection test is a PROPFIND of the folder.

The password is a secret field like the other backends' keys: never
published, and kept when the form is saved without it. The address is the
site admin's own setting, so a Nextcloud on WeKan's own network can be
used; redirects are refused so it cannot be pointed somewhere else.

With the CalDAV sync of the previous commit this completes #687. The
Nextcloud ExApp is built by .github/workflows/nextcloud.yml; publishing it
to the Nextcloud app store is the maintainer's own step.

Thanks to scriptPilot and xet7 !

Fixes #687.
2026-10-14 16:46:59 +03:00
Lauri Ojansivu 86ec72fade Optional APIs: Trello, Jira, GitHub and Gitea compatible, off by default.
Admin Panel / Settings / Visibility / Optional APIs ticks REST APIs in the
shape of Trello's, Jira's, GitHub's and Gitea's, so apps and scripts written
for them - Trellinator, Jira and GitHub clients - can use WeKan as their
backend, each described on the page with the address a client uses.

None has rights of its own. Each is a translator: a request in the other
tracker's shape becomes the matching request to WeKan's own REST API, made
with the caller's own WeKan token on the loopback address, and the answer
becomes the other tracker's JSON. Every WeKan permission therefore applies
unchanged - read-only and comment-only roles, assigned-only members, list
permissions - and an API that is not ticked answers 404. All of them sit
under /api, so WITH_API turns them off with the rest of the REST API.

Trello: members, boards, lists, cards, labels, comments and checklists; a
deleted card is archived. Jira: projects are boards (the board key), issues
are cards (DEV-7), statuses are lists and transitions move cards, with
simple JQL. GitHub and Gitea: repositories are boards and issues are cards,
open or closed as in Board View / Version Control, with labels, milestones,
comments and Link pagination.

The new strings are translated in every locale. A runtime test over real
HTTP shows a read-only member refused and a Trello-created card readable
through Jira and GitHub.

Thanks to xet7 !

Fixes #4930.
2026-10-14 13:11:23 +03:00
Lauri Ojansivu 7595727917 Boards every signed-in user may edit, per-list permissions, board keys and shared labels.
A board can let every signed-in user edit it, when Admin Panel / Settings /
Visibility allows it: a non-member then comments, edits and moves cards like
a Normal member, and their first edit makes them one. Board-wide settings
stay members-only.

Lists can allow or refuse editing and commenting per role (list menu, board
admins): built-in roles, custom roles and, on an open board, everyone signed
in. The server applies it to card, checklist, attachment and comment writes
and to moves into the list; the REST API honours a list's refusal on every
per-list card route. A direct client write of the setting is refused and
shown in Admin Panel / Problems.

Every board has a key, unique in the instance, suggested from its title and
set for existing boards at startup; card numbers show as KEY-7 and searching
for KEY-7 finds the card.

Shared labels in Admin Panel / Labels, for every board or an organization's
or team's boards; a board uses them beside its own, and a rename or colour
change reaches every board that uses it.

New strings translated in all languages. Low confidence: chr, tlh, wal, ay,
ace, nah, gn, iu, zgh, rup, wo, se, ff, tig, vo, bm, kl, bua, nap, mai, to,
ks, qu, bo, fj, haw, om, ss, ts, ve-PP, lld, scn.

Thanks to Meeques, justinr1234, tiFabGrinch, ejlkemper and xet7 !

Fixes #4575,
Fixes #4527,
Fixes #2449,
Fixes #2450,
Fixes #2761,
Fixes #4698.
2026-10-14 09:28:45 +03:00
Lauri Ojansivu 79ac301675 Email bounces with Amazon SES, Problems -> Email, and runtime tests for sending and receiving.
Email that comes back is handled in WeKan instead of landing in the
administrator's own inbox. Admin Panel -> E-mail -> Bounces: hard bounces
(a mistyped invited address), soft bounces, automatic replies and spam
complaints, each with actions done after a count - list it, stop emailing
the address, tell who invited it, delete the invitation, ignore - as
listmonk and Discourse do it, and the list of addresses that bounced, which
an admin clears to email them again. Amazon SES Bounce and Complaint
notifications and mail SES receives arrive through SNS at
/api/inbound-email/ses, signature-checked against Amazon's certificate and
accepted only from the topics the Admin Panel lists. Delivery status
notifications and out-of-office replies arriving by IMAP, a provider
webhook or a Sandstorm grain are bounces too, never cards.

Admin Panel -> Problems -> Email shows what went wrong sending or
receiving: a send the mail server refused or could not reach, an IMAP
mailbox that cannot be read (never with its password), an unreadable
email, an email for a board that no longer takes email. Attempts stay in
Problems -> Security. The inbound email endpoints no longer need the whole
REST API turned on.

Runtime tests run WeKan's own transports against local SMTP and IMAP
servers over TCP, a real Sandstorm Maildir, the webhook over HTTP and
signed SNS messages; UI tests drive the Email and Chat views, the
email-to-card settings and the Bounces page. They found and this fixes:
every whole-message parse failed in the server bundle (nodemailer's
addressparser resolved to its ESM build), the Chat view's drag and drop
never started (dragscroll took the mouse) and dropped a card into the
target's previous list, and the Chat panes did not follow a change to a
right-to-left language.

Thanks to xet7 !

Fixes #794,
Fixes #3948.
2026-10-13 18:27:25 +03:00
Lauri Ojansivu d6bee1bdaf Fix the six failing node suites.
Validation.md links models/eventLog.js by its real name, the README counts
the 166 catalogs now over 90% translated, and CHANGELOG.md is wrapped at 80
columns (whitespace only). The guards follow deliberate changes: Supported.md
belongs directly in docs/Databases, build.sh's macOS branch writes
export PATH="$HOME/.meteor:$PATH" since the npx meteor change, and the
wasmagic pin is checked against the lockfile instead of a copied number.

Thanks to xet7 !
2026-10-11 13:36:32 +03:00
Lauri Ojansivu ab785acd1e Choose what notifications contain, and group and schedule them, for e-mail, the tray and webhooks.
One delivery model (models/lib/notificationDelivery.js) for every channel -
e-mail, the notification tray and outgoing webhooks - set in the
Notifications popup at Admin Panel, board and Member Settings scope, and in
each one-way webhook's form:

- Content. Webhooks: the text on or off, and which property groups are sent
  (standard, ids, names, links, actor, people, details), with checklist,
  list, swimlane and board names, before/after values, the card description
  and links added as opt-in groups. E-mail and tray: the classic sentence or
  a clear layout - the board as a link first, the actor and card in bold -
  and which parts are shown.
- Grouping: one message per event, per card, per board, or all together.
- Scheduling: immediately, every 5 minutes to 24 hours, or once a day, with
  optional quiet hours, in the time zone of whoever set it.

The most specific explicit setting wins: member, then board, then Admin
Panel for e-mail and tray; the webhook, then its board, then Admin Panel for
webhooks. A member can only leave their own name out of outgoing webhooks.
The built-in defaults keep today's behaviour. Grouped and scheduled webhooks
and e-mail go through durable queues that survive restarts; queued webhooks
use the SSRF guard and log a blocked address in Admin Panel - Problems.
Values of admin-only custom fields now reach only board admins in e-mail and
tray, and never a webhook.

The new strings are in English in every locale, pending Transifex.

Thanks to HaKePlan, Meeques and xet7 !

Fixes #3695,
Fixes #5171.
2026-10-10 15:28:26 +03:00
Lauri Ojansivu d56bb34490 Make large boards fast on 8 GB: load only what is visible, stop re-running every minicard, and stream big attachments.
On a seeded board of 4,000 cards, 12,000 comments, 20,000 checklist items
and 32,000 activities, served from FerretDB with SQLite as the Snap is:
closing a card took about 4 s and now takes 0.1 s, the board finished
loading in 17.6 s and now in 6.2 s, a card open re-ran about 7,700
computations and now about 1,000, and FerretDB used 13 s of CPU and 724 MB
while the board opened, now 3 s and 280 MB.

- The client data cache queued one more stop callback each time a helper
  re-ran, and every check asked Tracker for dependents, which on close was
  710 walks of 3 ms; there is now one callback per computation and key and
  one timer per key (imports/lib/dataCache.js).
- Opening a card writes the user's last-viewed card, and avatars, @mentions,
  list and swimlane headers read the whole user, so everything re-ran; they
  read only the fields they show. The comment, checklist and subtask indexes
  were one object every minicard depended on and are now per card
  (imports/lib/keyedGroupIndex.js).
- The lazy card window kept a full copy of every card of a list per window;
  it keeps the id and the sort keys, runs its id query once instead of five
  times, and sends comment text only when minicards show comments
  (server/publications/cardsWindow.js, models/lib/cardWindowFields.js).
  Comment reactions and legacy attachment lookups come with the opened card
  only.
- Attachment downloads honour Range for filesystem, GridFS and cloud
  storage (206, 416, If-Range); uploads to cloud storage are spooled to a
  temporary file and streamed instead of gathered into one Buffer; the API
  copy checks the size limit before reading; the legacy attachment
  migration streams, and no longer stores "[object Promise]" as a file; the
  board export no longer concatenates buffers quadratically.

The FerretDB fork pushes dotted-path filters such as meta.cardId down to
SQLite (they were each a full scan) and decodes only the fields a count
filters on. docs/Features/Admin-Panel/Problems/Large-Boards.md describes how
a large board loads and how to size a server with 8 GB of RAM.

Thanks to markusst1982 and xet7 !

Fixes #6745.
2026-10-09 13:28:08 +03:00
Lauri Ojansivu e8fac5e85c Record every board import before its first write, and let an admin keep or discard an interrupted one.
An import that stopped halfway - a restart, a hung writer, a failure - left a
partial board nothing recorded. Each board import, copy, Trello zip and
Trello API import now writes an importRuns record first, with the id its new
board will have; every creator inserts the board with that id and the run's
id stamped on it. A heartbeat keeps the run alive; the method deadline now
stops the writer at its next step. A scan at startup and every 5 minutes
flags a stopped or failed run once and records it in Admin Panel ->
Problems -> Recovery, where an administrator keeps the partial board or
discards it: only the board carrying the run's stamp is removed, through the
board's own removal, then whatever else has its id. Discard is refused while
the import's Scrum stage is still written or recovered, and the Scrum resume
is refused while the import is discarded. Resume is deliberately not
offered: the source file is not kept, for its size, its personal data and
the anonymization it would undo; discard and import again instead.

Thanks to xet7 !
2026-10-08 12:44:46 +03:00
Lauri Ojansivu cb69750b17 Let administrators discard List Sync operations that cannot be replayed.
A saved List Sync operation whose list was removed, recreated or
reconfigured, or whose actor lost full-list write access, could never be
replayed. The replay job refused it every minute with only a console line,
and because a new Sync of the list resumes the pending operation first,
that list could not Sync at all.

Such an operation is now marked once, with one Recovery event, and a
blocked manual Sync says why. Admin Panel / Problems / Recovery lists these
operations with what they applied and their live replayability. An
instance administrator can discard one only while the same checks a replay
makes still fail. The discard holds the list's Sync lease, records one
immutable decision, then removes the saved steps and the operation marker,
reading each removal back. It writes no card: changes already applied stay
and the rest are never written. A retry or a second administrator only
finishes the removal, and a replay that finds a decided discard completes it
instead of resuming. The intent stays as evidence and the discard is
audited in Recovery.

Unit, source-wiring and negative tests cover recording once, refusing
non-administrators and replayable operations, and discarding twice; an app
test and a Playwright case cover the Recovery page. New interface text is
in English, pending Transifex.

Thanks to xet7 !
2026-10-08 11:53:06 +03:00
Lauri Ojansivu c1d1a04790 Implement automatic logout with LOGOUT_WITH_TIMER and the LOGOUT_* settings.
The snap, Docker, docker-compose and start scripts have offered
LOGOUT_WITH_TIMER, LOGOUT_IN, LOGOUT_ON_HOURS and LOGOUT_ON_MINUTES since
2018, but the code that read them was removed with the job queue it ran on,
so they did nothing. Per the maintainer decision of 2026-10-08 they work
again. A login ends LOGOUT_IN days after it was made, or at
LOGOUT_ON_HOURS:LOGOUT_ON_MINUTES server time (the next such time, or on the
day LOGOUT_IN days later). Once a minute the server removes the expired
login tokens in one query, and Meteor signs out the browsers that used them.
The four settings are overridable in Admin Panel / People / Login like the
other login variables, with hours and minutes bounded. An unusable
combination signs nobody out and the server log says why.

tests/logoutTimer.test.cjs checks the one-query cutoff against the
per-login deadline over two years of logins in three time zones, two of them
with daylight saving time.

Thanks to xet7 !
2026-10-08 11:03:50 +03:00
Lauri Ojansivu 95147fc999 Add an LDAP Sync now button to Admin Panel / People / LDAP.
ldap_sync_now existed in packages/wekan-ldap/server/syncUser.js, but nothing
imported that file, so the method was never defined. It also imported every
directory user regardless of the sync settings. Per the maintainer decision of
2026-10-08 it is now loaded and wired to a Sync LDAP users now button. It runs
the background sync once with the settings in effect: import new users and/or
keep existing users updated, without needing the schedule to be on. Only an
active site administrator can run it, and with neither setting on it says
there is nothing to sync. The button and the scheduled job share one run at a
time, and a failed run reports the directory's error instead of success.

New text is added in English, pending Transifex.

Thanks to xet7 !
2026-10-08 10:59:21 +03:00
Lauri Ojansivu 9d27f4be4d Make LDAP Test connection prove the directory answered without a service account.
ldapts opens its socket lazily. Test connection connected and then bound only
when LDAP_AUTHENTIFICATION was set, so without a service account no operation
ran and it reported success even for a host that does not resolve. Per the
maintainer decision of 2026-10-08 a test without credentials now does an
anonymous base-scope search of LDAP_BASEDN (no attributes, one entry at most)
and succeeds only when the directory answered; otherwise the directory's own
error is shown. Without a base DN it says nothing could be tested.

Thanks to xet7 !
2026-10-08 10:43:35 +03:00
Lauri Ojansivu a7628504f1 Make header login settings environment-only.
With HEADER_LOGIN_ID and HEADER_LOGIN_TRUSTED_IPS set, a proxy at a trusted
address signs in as anyone it names. Since the 2026-10-05 Admin Panel work a
site administrator could switch that on from People / Header login, so a
stolen administrator session was enough to open every account. Per the
maintainer decision of 2026-10-08 the six HEADER_LOGIN_* settings now come
from the environment only: a value stored in the Admin Panel is ignored, the
pane shows the values in effect read-only with no Save button, and a save sent
by hand is refused and recorded in Admin Panel / Problems as ProxyBleed
(authn.header-login-env-only). Every other login section stays overridable.

New text is added in English only, pending Transifex.

Thanks to xet7 !
2026-10-08 10:41:59 +03:00
Lauri Ojansivu 03975e7d31 Admin Panel / Settings / Visibility / Features: choose which card fields every board shows.
Above the Features Save there is now one tick per field type of Board
Settings / Card - the same icons and names, all ticked by default. An
unticked field, for example Pomodoro, is not offered in any board's Board
Settings / Card and is not shown on any opened card or minicard, whatever
the board chose. Visibility only: no card data and no board setting is
written, so ticking it again shows every board as it was set. There is no
ordering here; the order stays each board's own.

The catalog and the decisions are models/lib/cardFieldVisibility.js. It is
stored as cardFieldStates on the settings document (missing means shown,
so existing installs are unchanged), published with the other Features
settings and saved by the site-admin-only saveInstanceFeatures method,
which refuses unknown field keys. The hiding happens at the client's one
board read (ReactiveCache.getBoard), which returns the board with the
hidden fields' flags reading false - the same object when nothing is
hidden - so every card and minicard template that reads a board flag agrees
without being edited; the per-card List title and the user's Labels text
override check the same mark. The server's reads, publications and the REST
API are unchanged. A whole-column tick in Board Settings / Card leaves the
hidden fields' board values alone.

Thanks to xet7 !
2026-10-08 08:54:04 +03:00
Lauri Ojansivu 63a7fba465 Let the login-cookie refresh limit be raised, and stop browser tests signing themselves out.
Meteor answers more than 30 refreshes of the HttpOnly login cookie per 10
seconds from one client address with 429, and every page load is one. The
whole browser suite comes from localhost, so a quick run signed its own
test users out: 18-rtl-layout "list pages" landed on the sign-in page on
its third full navigation in one of two runs with two workers. TODO Later
recorded this on 2026-10-05.

ACCOUNTS_COOKIE_REFRESH_RATE_LIMIT now sets the allowance per 10 seconds;
anything but a positive whole number keeps Meteor's default, and the
limiter is never switched off. The browser test server sets it to 100000.
It also helps an install where many users reach WeKan through a proxy
that is not trusted for the client address and so share one address;
documented beside the login settings. loginWithToken additionally waits
out a 429 on the cookie refresh and resumes once, and only for that.

tests/cookieRefreshRateLimit.test.cjs runs the parser with valid and
invalid values and pins the server, build.sh, docs and helper wiring.

Thanks to xet7 !
2026-10-07 17:29:08 +03:00
Lauri Ojansivu 0314406656 Read login secrets from *_FILE, and let the Admin Panel override DEFAULT_AUTHENTICATION_METHOD.
LDAP_AUTHENTIFICATION_PASSWORD_FILE and OAUTH2_SECRET_FILE were listed in the
Dockerfile, snap, docker-compose and start-wekan since #5724, but no code read
them, so a password could only be an environment variable. WeKan now reads the
file itself, at each login (a rotated secret applies at once), so the password
is never in the environment - the point of Docker / Kubernetes secrets. The
order is the Admin Panel, the variable, then its file; one trailing line break
is dropped. The application code is the same on every platform, so it works in
Docker, the snap (files under /var/snap/wekan/common), bundles, AppImage,
Flatpak, Mac and Windows. Admin Panel / People shows when the password comes
from the file and when the file cannot be read, never the path or content.
The path itself is an environment variable only: settable in the Admin Panel
next to the LDAP host or the OAuth2 token endpoint, it would let the server
read any file and send it there as the password.

DEFAULT_AUTHENTICATION_METHOD is a field of the Login pane's settings form.
Choosing a method overrides the variable; Default gives it back. Before, the
variable rewrote the stored method at every start, so an administrator's
choice was lost on restart. The published method is now the override, else the
variable, else password. On upgrade, a method chosen with the old dropdown
while no variable was set becomes the override, once, so it is not lost. The
old dropdown is removed: two controls for one setting on one page.

The snap key for the LDAP password file was spelled
ldap-authentification-password-file, unlike its siblings and its help text;
it is ldap-authentication-password-file now (it had no effect before).
MAIL_SERVICE_PASSWORD_FILE, MONGO_PASSWORD_FILE and S3_SECRET_FILE are still
not read by WeKan; secrets/README.md says so.

Thanks to Roemer, CrashOverride-lab, ww-daniel-mora and xet7 !
2026-10-05 21:11:12 +03:00
Lauri Ojansivu eb49a4b1b6 Override every login environment variable in Admin Panel / People.
Each login method has its own section: LDAP, OAuth2 (OpenID Connect and
Oracle OIM), CAS and Header login are new or complete, and the Login pane
gets PASSWORD_LOGIN_ENABLED and ACCOUNTS_COMMON_LOGIN_EXPIRATION_IN_DAYS.
models/lib/authConfigCatalog.js lists all 103 variables; server/lib/
authConfig.js resolves each one at run time (Admin Panel value first, then
the environment variable) for the app and, through
globalThis.__wekanAuthEnv, for the wekan-ldap, wekan-oidc and
wekan-accounts-cas packages. OAuth2 and CAS service configurations and LDAP
background sync are reapplied when their settings are saved.

Passwords and client secrets (LDAP_AUTHENTIFICATION_PASSWORD, OAUTH2_SECRET)
never reach the browser: the page learns only whether one is set and where
it comes from. Other values have credentials written inside a URL masked on
the server, and a URL field refuses them.

Fixed on the way:
- The Admin Panel LDAP overrides never applied. The bridge read Settings with
  a server-side findOne(), which Meteor 3 refuses; the error was swallowed and
  the environment used. The settings are now held in an observed cache.
- LDAP Test connection never existed on the server (testConnection.js was not
  imported) and used the synchronous Meteor.user().
- PROPAGATE_OIDC_DATA=false turned the feature on.
- CAS read only the misspelled CASE_VALIDATE_URL; CAS_VALIDATE_URL now works
  and the old name still does.
- snap help named three group-filter keys snap set does not know.

Every catalog variable is now listed in the Dockerfile, the docker-compose
files, snap config and help, and start-wekan.sh/.bat; 14 were missing.
tests/authConfigCatalog.test.cjs pins resolution, secrets, input validation,
platform coverage, and - across the whole tree - that no code reads a login
variable outside the catalog or straight from process.env, and that no
server code calls the synchronous Settings.findOne() or Meteor.user().
tests/playwright/specs/admin-login-env-overrides.e2e.js drives the pages.

Thanks to xet7 !
2026-10-05 19:48:48 +03:00
Lauri Ojansivu 6b9bc496b1 Add continuous backup to the Admin Panel: stream every change, restore to a moment.
Admin Panel / Attachments / Continuous backup, beside Backup, streams the
database, filesystem attachments, avatars and logs to a target directory
within seconds, so less work is lost between scheduled archives and a
restore can return to a chosen moment. Design: docs/Backup/Continuous-Backup.md.

The built-in engines use only copyfree code - WeKan's own (MIT), Node.js
and its node:sqlite (SQLite, public domain) - and start no other program:

- oplog: tails local.oplog.rs like Meteor's oplog driver, on a MongoDB
  replica set or FerretDB with its oplog; a base archive (the existing
  wekan-instance-backup format), then whole documents and deletions,
  continued from the saved position after a restart, a new generation
  when the oplog has lost it.
- sqlite: SQLite's online backup API on FerretDB's WAL database at an
  interval, storing only the pages that changed.
- files: attachments, avatars and logs by watching and by rescans, each
  content once under its SHA-256, links never followed.

Litestream (Apache-2.0) is an optional engine for SQLite: WeKan writes its
configuration and supervises a binary the administrator installed.

Segments are written whole, flushed and listed in append-only indexes.
A restore checks every index line, segment and blob first and writes
nothing when one is missing or changed; a database restore replays up to
the chosen moment, a file restore puts files back (add-missing or
replace-all), and a SQLite restore builds a checked file beside the stream,
never over the live one. Site administrators only; failures of the
unattended stream show in Admin Panel / Problems.

Tests: tests/continuousBackup.test.cjs; tests/integration/continuousBackup.test.cjs
against a real replica set, a real WAL database, real files and a stand-in
Litestream; server/lib/tests/continuousBackup.tests.js through the methods;
tests/playwright/specs/admin-continuous-backup.e2e.js on Chromium and WebKit.

Thanks to xet7 !
2026-10-03 01:47:42 +03:00
Lauri Ojansivu 6ed28c41f7 Run manual and scheduled Sync through the durable journal, with replay.
The write-ahead Sync journal and its stored stages - History, activities,
rules, notifications, webhooks - were built and tested but never called:
manual and scheduled Sync wrote cards directly, so a restart mid-run lost
whatever effects had not happened yet. Maintainer decision of 2026-09-30:
finish the Scrum/Sync handoff by calling the stored stages from both.

A run now takes the durable path when it can do everything
(server/lib/listSyncSteps.js): the board enabled Sync effects (and the
instance enabled cron effects for a scheduled run), the list has a versioned
scope, and every rule action on the board - extra actions included - has a
durable adapter. Only sending email has one, so a board with any other rule
action keeps the direct writes, exactly as before; nothing that ran stops
running.

server/lib/listSyncApplication.js turns the reconcile plan into saved steps
from the stored cards, refusing a local edit made since the fetch as the
direct update would, and carrying custom fields only with a mapped estimate.
Effects are planned once with the shared planner and each step is applied
through the hooked card and activity adapters and durable delivery, as the
actor and trigger that started it. Both paths build the same new card.

A new run first finishes the list's unfinished operation, reusing the list
lease it holds instead of taking it twice. The wekan-list-sync-replay cron
job resumes interrupted operations every minute; intents now record their
trigger (version 2) so a replay runs under the same activation switches. An
operation still preparing has written nothing and is discarded - a replay has
no source data to rebuild it from.

Tests: eligibility for every condition, and step building including the
local-edit refusal (node); the direct-path suite now pins that it never calls
the durable one; a full-app run creates, updates and archives through the
journal, crashes at a card write, replays without duplicate cards or
activities, and falls back to the direct path for a rule action without an
adapter. 565 server tests pass.

Thanks to xet7 !
2026-09-30 11:29:44 +03:00
Lauri Ojansivu 778e9ed957 Show private rule email delivery reports in administrator Recovery 2026-09-28 16:06:59 +03:00
Lauri Ojansivu f1e572db5c Integrate cancelled notification payload cleanup and terminal receipt reporting 2026-09-28 15:13:39 +03:00
Lauri Ojansivu 8814dbf1f4 Compact cancelled activity payloads into permanent terminal receipts 2026-09-28 15:08:00 +03:00
Lauri Ojansivu b590d3d76f Cancel remaining activity notification delivery with permanent replay protection 2026-09-28 15:04:37 +03:00
Lauri Ojansivu 7fef4f4cdc Add administrator pause and resume controls for activity notifications 2026-09-28 15:00:31 +03:00