Editing a provider's upstream URL meant retyping its API key. The key is the
one field the API never returns, so there was nothing to retype from, and the
dashboard had to demand it because a provider_id request resolved the upstream
from the stored row — listing the old endpoint while the form showed the new
one.
The request now reads as a partial edit of the record it names, the same way
PUT on a provider already does: an upstream in the body overrides the stored
one, an omitted upstream keeps it. The credential and the catalog entry still
come from the record only, so a caller cannot aim a stored key at a vendor of
their choosing.
Pressing "Load models from provider" against a key the vendor refuses answered
"internal server error". Every outcome on that path is the operator's own key
or upstream, so a 500 was wrong twice over: it told them the server broke, and
it named nothing they could act on.
The failures are already typed and already have sentences written for them —
the save-time check translates the same set. Discovery now runs them through
the same classifier, so the button reports a refused credential or an
unreachable url in the words the form uses elsewhere.
ErrNoDiscovery and ErrInvalidRequest pass through untouched. The handler maps
both already, and a provider with no listing endpoint is a fact about the
catalog entry rather than a failure — the caller falls back to the catalog's
own models instead of showing an error at all.
The dial-time guard reported every non-public address as ErrPrivateHost, which
the credential check reads as "this upstream cannot be reached from here, so
save it unchecked". checkPublicHost has already cleared the target by the time
anything is dialled, so an address refused at the socket is never the
operator's upstream — it is a rebinding attempt, or an HTTP proxy the
management server egresses through. A deployment behind such a proxy would
install this feature and have it silently do nothing on every provider.
It now reports an ordinary failure, which classifies as unreachable and blocks.
Only the resolve-stage check still means "cannot be checked", and that one
knows it is looking at the operator's own host.
This is also why the three fixtures below passed locally and failed in CI: a
sandbox that egresses through a loopback proxy skipped the check entirely,
while CI reached the real api.openai.com and had the dummy key refused. They
want a provider row rather than a working vendor, so they move to a private
address and no longer depend on where a hostname resolves or whether the runner
has egress.
The SSRF guard resolves the host before any request is built, so a name that
does not resolve fails there rather than at the transport, and that error
reached the caller unclassified — an operator with a typo in the hostname was
told the provider could not be checked rather than that the url could not be
reached. It is the commonest way for an upstream to be wrong.
The live suite is what caught it: the unit tests construct the transport
errors directly and so never went through the guard.
The management fixture moves to a private upstream in the same change. It
wants a provider row to hang a policy off, not a working vendor, and it was
pointing a dummy key at the real api.openai.com — which the credential check
now correctly refuses. A private address is left unchecked whether or not the
run has vendor keys, and covers that path while it is there.
The OpenAPI change is the 422 both provider routes can now answer, plus what
decides it: the create path checks the pair before storing, the update path
checks only when the upstream or the key moved, and an update omitting the key
is checked against the stored one.
The live tests cover what a unit test structurally cannot. Mocked refusals
prove the classifier maps a status to a message; they cannot show that these
vendors refuse a bad key on their listing endpoint at all, which is the
assumption the feature rests on. The good-key case earns its place beside the
bad one — a check that refused everything would satisfy a test asserting only
the refusal.
The rotation case pins the state worth the most: after a rejected key, an edit
that reuses the stored one still passes. The API never returns a key, so that
is the only way to show the working credential is still there.
The provider form accepted anything and found out later. A typo in the
upstream, a key pasted a character short, an AWS access key in a field that
wants a Bedrock API key — all saved cleanly, then surfaced minutes later as a
failed request or an empty model picker, with nothing pointing back at the
record that caused it.
CreateProvider now spends the credential once against the vendor's own model
listing, and UpdateProvider does the same when the upstream or the key
changed — only then, so renames, model rows and price edits neither wait on a
vendor nor fail because one is having a bad day. Both run before the store
write, so a rejected rotation leaves the working key exactly where it was.
The check reuses the discovery Fetch rather than a lighter status probe. It
exercises the path the model picker will take, so a URL answering 200 with a
login page fails here instead of passing a status check and producing an empty
picker later.
Failures that mean "we cannot ask" are not failures: a gateway with no listing
endpoint, a Bedrock record behind a proxy where no control-plane host can be
derived, and a self-hosted endpoint on a private network that the proxy
reaches through the tunnel but management cannot. None of those are evidence
the record is wrong, and refusing them would make this a lockout.
Discovery failures are typed for it. The message an operator sees carries no
status code and never echoes their URL — WriteError lowercases it, and paths
are case-sensitive, so an echoed URL would come back altered and describe
something they did not type. The vendor's status is logged instead.
[proxy,management] Serve Bedrock model discovery from the control plane
A Bedrock provider could never answer a model-discovery request. The router
sent GET /inference-profiles to the record's upstream, which has to be
bedrock-runtime.<region> for InvokeModel to work, and that host does not
implement the operation. ListInferenceProfiles is a control-plane operation on
bedrock.<region>.amazonaws.com, and one provider record carries one upstream,
so the two hosts genuinely differ.
The route now carries a discovery host, taken from the catalog's declaration
with the region read back out of the configured upstream, and the listing — and
only the listing — goes there. Inference is untouched. A proxied or self-hosted
Bedrock endpoint gets no discovery host at all rather than a guessed one, since
inventing a host would send the operator's credential somewhere they never
configured.
Two things had to follow for the listing to be usable once it arrives. The
response filter only understood OpenAI's {"data":[{"id":…}]}, so a Bedrock
listing fell through it untouched, offering every profile in the account
whatever the policy said. And discoverableModels intersected by exact string,
so a record registering the raw profile id while a guardrail names the catalog
key intersected to nothing — bounding a working provider's listing down to
empty.
Normalisation is the third. The geography in front of a cross-region profile
was matched against a hardcoded list of four, so every profile issued under jp,
au, ca, sa or us-gov carried its prefix into the pricing key, matched no
catalog entry and metered at zero. It is now recognised by either the geography
or the vendor that follows it, so an id has to be new on both axes at once to
slip through — a live eu-central-1 listing returned "global.xai.grok-4.6" days
after the vendor list was first written.
[management] Offer a provider's live model list in the config form
Adds POST /api/agent-network/catalog/providers/models, which asks a vendor
which models an operator's own credential can actually reach, so the provider
form can offer a live list instead of only the compiled-in catalog. The catalog
goes stale, and it cannot see an account: which OpenAI models an org is
entitled to, which Bedrock inference profiles an account and region hold, which
Vertex models a project has enabled.
The endpoints, auth headers and response shapes come from probing the live APIs
(#7244); each vendor invented its own envelope and none can be guessed from the
request. Bedrock shaped the design: its listing lives on the control plane
while inference must go to the runtime host, so Discovery carries its own host
rather than reusing the record's upstream, and profile ids are taken verbatim
because the region prefix is what AWS requires at invoke time.
A caller supplies either the key they are typing or the id of a saved record
whose stored credential is reused — never both, since accepting both would run
an arbitrary credential under the identity of a record the caller may only be
permitted to read. Gated on Create rather than Read, because this spends the
operator's credential against a third party.
Management has not made outbound calls on an operator's behalf before and it
holds a credential for every provider, so every resolved address must be public
— covering loopback, RFC1918, the cloud metadata address and NetBird's own
100.64/10 range — and redirects are not followed, since a redirect moves the
request to a host the check never saw.
The vendor is authoritative for the id; the catalog stays authoritative for
pricing. A discovered model the shipped table cannot price returns
pricing_known: false so the operator must set rates rather than being
registered at a silent zero.
[proxy,management] Bound model discovery to the caller's own policies
GET /v1/models was bounded by the provider record's enumerated models, which is
the right bound only while a single policy reaches a provider. Where two teams
share one provider under different allowlists, every caller was offered the
union — each model outside their own policy being a request the guardrail
refuses a moment later. A gateway record enumerating nothing was worse: it
offered the upstream's entire catalogue however narrow the policy was.
Each route now carries one rule per authorising policy — its source groups and
the models it permits — instead of a single flattened list. At request time the
router keeps the rules whose groups intersect the caller's, unions their
models, and intersects that with what the provider serves.
nil and [] stay distinct end to end: a policy setting no allowlist reaches the
router as nil and lifts the restriction for the groups it binds, while an
enabled allowlist with no models arrives as [] and permits nothing. Collapsing
them would let a listing that should offer nothing fall open to everything.
The guardrail's own per-provider allowlist is untouched. It is a fail-closed
backstop that cannot tell who is asking, so discovery is now narrower than the
backstop rather than wider.
[proxy,management] Conform the Agent Network endpoint to the LLM gateway protocol
Reviewed the proxy against Claude Code's published gateway contract. The
transport layer already held up; fourteen gaps sat one layer up, in the model
catalog and in the non-inference endpoints clients call.
Two of them cost money. The catalog carried no claude-opus-5 or
claude-sonnet-5, so an operator could not authorise the models coding agents
default to — those requests denied as not-routable, or priced at zero where a
catch-all carried them. And gateway records pin ParserID "openai" while the
same record serves /v1/messages, so Anthropic responses were read with the
OpenAI parser, which never looks at message_start where input tokens live:
input metered as roughly zero on every stream and cost was skipped entirely.
The rest fix requests refused for structural rather than policy reasons: model
discovery denied for every account with a model allowlist, token counting
denied on Bedrock and mis-parsed on Vertex, startup probes refused and written
into the access log at every session start, and denials rendered in a shape no
LLM client parses. Two changes are additive by design — the deny body keeps
every field it had and adds the vendor's error object alongside, and body-level
identity injection is now gated on the request's dialect so it stops sending
OpenAI-shape fields into Anthropic bodies that reject them.
The end-to-end work turned up one more: the discovery filter treated any slash
in a model id as a gateway prefix, which would have dropped every self-hosted
"Qwen/..." model from the picker.
* [misc] Let gomobile init fetch the toolchain gobind needs
The previous commit's CI run confirmed the failure on a comment-only diff
off main, so the cause is not any branch's changes:
Android / Build failure
iOS / Build failure
`gomobile init` re-installs gobind from x/mobile@latest whatever gomobile is
pinned to, and setup-go sets GOTOOLCHAIN=local, so the install dies the
moment @latest declares a newer Go than go.mod does:
gomobile: go install golang.org/x/mobile/cmd/gobind@latest failed: exit status 1
go: golang.org/x/mobile@v0.0.0-20260821190718-4776eadac327
requires go >= 1.26.0 (running go 1.25.12; GOTOOLCHAIN=local)
GOTOOLCHAIN=auto on that step alone lets the install fetch what it asks for.
Scoped to the step deliberately: the repo's Go version and every build below
it stay on go.mod's toolchain, so this buys the mobile jobs nothing except
the ability to run gobind.
Pinning gobind next to gomobile does not work — init re-installs @latest
regardless. A durable fix is to stop `init` reaching the network at all, or
to track x/mobile's Go requirement in go.mod; both are larger changes than a
red CI warrants right now.
The updater runs as LocalSystem and started netbird-ui via
CreateProcessAsUser with a nil environment, so the UI inherited the
SYSTEM environment (USERPROFILE, APPDATA pointing at systemprofile)
while running under the user's token. The WebView2-based UI exits
immediately in that state, so the UI never came back after an update.
Build the environment from the user's token with CreateEnvironmentBlock
and pass it to CreateProcessAsUser.
Android never reported its local network interfaces, so PeerNetworkRange posture checks could not be evaluated: NetworkAddresses always arrived empty.
net.Interfaces() is unusable on Android 11+ (SELinux blocks netlink), so the addresses are parsed from the interface description the host app already provides via stdnet.ExternalIFaceDiscover. The MAC filter is skipped, mirroring #5906
for iOS, since Android does not expose MACs either and nothing reads Mac server side.
Stop the UI before a silent Windows update and suppress the installer reboot
On silent MSI updates msiexec could reboot the machine on its own. The running UI holds a lock on its own exe, and since msiexec runs as LocalSystem it cannot close the interactive user's UI via Restart Manager, so the MSI scheduled the
file replacement for the next reboot and marked the install restart-required.
Terminate netbird-ui.exe before launching the installer and wait until its image file is released; the existing deferred restart brings it back after the install on every exit path
Run msiexec with /norestart REBOOT=ReallySuppress so it never reboots on its own
Treat exit codes 3010/1641 as success with a warning instead of a failure
---------
Co-authored-by: Viktor Liu <viktor@netbird.io>
* [client] Clear stale installer result before starting update
The installer result file could survive a previous update attempt (e.g.
when the updater wrote it after the restarted daemon already ran its
startup check). A new install attempt left the old file in place, so the
GUI progress window's first GetInstallerResult poll read the outdated
result: a stale success made the GUI quit mid-install, which cancelled
the TriggerUpdate context and aborted the artifact verification; a stale
error surfaced a bogus failure dialog for a succeeding update.
Remove any leftover result file at the start of RunInstallation, before
the download begins, so result watchers only see the current attempt's
outcome.
* [client] Align stale-result warning with log message style
On network changes the client restarted the whole engine. That is heavy-handed and slow: it tears down working state to recover from a transition the engine could handle itself. This replaces the restart with proper network event handling.
Suspend the retry loops while no network is available. Instead of burning through backoff intervals against an unreachable network, the reconnection loops park until the OS reports a usable network again.
Reconnect immediately on a network switch. When the OS hands us a new network, connections bound to the old one are swept and re-dialed right away, rather than waiting for a timeout to notice they are dead.
Adds an SSHClient gomobile binding so the Android app can run an SSH session over the tunnel with a PTY, exposed through a listener interface for the in-app terminal.
Server type is auto-detected from the SSH banner, which selects the auth path: JWT device-code flow, NetBird key, or a regular server (NetBird key first, then password). Host keys are verified against the peer registry for NetBird servers and trust-on-first-use for regular ones.
* [docs] Update agent-network docs for management-owned pricing
The docs still described the retired proxy-side pricing: pricing.Loader,
pricing_path, MiddlewareDataDir, embedded defaults_pricing.yaml, and the
symlink-safe Unix loader. Rewrite them for the current design — management
synthesizes the whole table and ships it in cost_meter's ConfigJSON, so the
proxy carries no price list and has nothing to reload.
* [client] Pass stored email as login hint from UI and keep it on logout
Follow the CLI pattern: the Wails UI now reads the account email from the
user-owned profile state file and passes it as the OIDC login_hint on login
and session extend, since the daemon-side fallback runs as root and cannot
see the user's state file. Logout no longer deletes the stored email, so a
later login preselects the account at the IdP; profile removal remains the
operation that deletes it.
* [client] Log ignored profile lookup errors in extend-session hint fallback
* [client] Skip IPv6 route tests when the default nexthop is unusable
ensureIPv6DefaultRoute treated a successful netlink RouteAdd as proof that
a usable IPv6 nexthop exists. Installing ::/0 via loopback can succeed while
the kernel still rejects that nexthop for a concrete prefix, which surfaced
on ubuntu22/20260810.260 runners as:
add route to table: netlink add route: invalid argument
Probe the resolved nexthop by installing and removing a discard-prefix route
through the same code path the tests use, and skip when it fails. EEXIST
means the nexthop already carries a route, so it counts as usable.
* [client] Probe the IPv6 nexthop through raw netlink
addRoute swallows EAFNOSUPPORT and EOPNOTSUPP via isOpErr, so a nil return
did not prove the probe route was installed. Call netlink directly so an
unsupported operation skips the test instead of passing as usable.
Align Android logout semantics with the desktop UI and CLI: logging out no
longer deletes the stored account email, so the next login passes it as the
OIDC login_hint and the IdP preselects the account. Removing the profile is
now the operation that deletes the email; previously RemoveProfile left the
account file behind, which the fixed-name default profile would have
inherited on recreation.
English (en) is the source of truth for UI translation keys; the other
nine locales rely on runtime English fallback for any missing key, so a
gap never surfaces in CI. Add a dependency-free Node check that fails
when any locale declared in _index.json does not carry the exact same
key set as en (missing or orphaned keys), wired into a dedicated
UI Translations workflow that runs on locale changes.
Also close the one existing gap the check found: ja was missing
daemon.outdated.download ("Download Latest").
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Prepares the repository for the release-branch process agreed internally:
one long-lived release-0.N branch per minor, with fixes backported by
cherry-pick and patch releases tagged from the branch.
Pushes to release-* branches now run the Release workflow and publish
immutable sha-* container images, the way pushes to main already do, so
a release branch can be tested before it is tagged. Release branches
never publish the floating "main" image tag. The push-to-main CI
workflows (Go tests on all platforms, frontend UI, install script,
mobile/wasm validation, infrastructure files, license check) also run
on release-* pushes; pull request triggers were already unfiltered, so
backport PRs were covered — this closes the post-merge gap.
Releases are no longer marked latest before signing: make_latest is
now false in all four goreleaser configs, so a release stays published
but not latest until the signing pipeline uploads the signed Windows
and macOS artifacts and marks it latest itself. Previously the release
became GitHub's "Latest release" at publish time, and the download
endpoints that resolve through the latest-release API could serve a
release whose signed installers did not exist yet. prerelease: auto
additionally labels rc tags as prereleases, so a release candidate can
never take the latest slot.
The trigger_sync_tag job is removed: it dispatched a downstream
image build on every v* tag (release candidates included), which would
race the deliberate release-branch build on every release. The android
and ios submodule bumps are unchanged.
Also sets perennial-regex = "^release-" so git-town never syncs or
ships a release branch into main.
The NSIS installer deleted HKLM/HKCU CurrentVersion\Run values it never
writes, which matches common AV heuristics for unwanted Run-key
manipulation and is suspected to contribute to Windows Defender and
third-party antivirus false positives on the installer.
Drop all autostart registry deletions from both the install and
uninstall sections so the installer only touches keys it creates
itself. Cleanup of the legacy machine-wide entry written by old
installers is left to documentation.
Extends the approach of the closed PR #6735, which only removed the
per-user deletion on uninstall.
e2e/harness documents itself as feature-agnostic, but three details
assumed the caller lives in this repo, so the terraform provider's
acceptance suite would otherwise carry a second harness for the same
product.
repoRoot took the first module root above the working directory as the
Docker build context, which from another module is the caller's own
root, with no combined/Dockerfile.multistage in it. It now requires that
ancestor to be this module, and otherwise asks the go tool for the
source: for a dependent, the extracted directory of the version it pins,
so the server matches the client library it was compiled against. That
lookup uses -mod=readonly, since automatic vendor mode otherwise reports
an empty Dir.
Geolocation was disabled unconditionally. Agent-network ingest does not
use it, but location-based posture checks need the database, and a rule
management cannot evaluate fails rather than passing.
StartClient pinned one network alias and set no hostname, so a second
agent could not start and a peer's name was arbitrary. Management
records that hostname, making it the peer's name in the API.
The client entrypoint is copied with an explicit mode: git tracks it
100755, but the module cache extracts 0444, so a dependent's build
produced a container exiting with "permission denied".
Adds CombinedOption, WithGeolocation, WithServerEnv, ClientOption and
WithClientName.
- Align default names and reuse same environment variables
- With the uploads now targeting the same stable/yum paths as the GTK4
packages, two packages named netbird-ui with the same version and arch
would collide in the repo indexes. Give the GTK3 variant its own
package name and mark the two as conflicting alternatives.
---------
Co-authored-by: Zoltan Papp <zoltan.pmail@gmail.com>
