19f6a18b8 moved credential-env resolution behind an explicit lookup
(ApplicationConfig.ProxyAPIKeyEnvLookup / config.WithProxyAPIKeyEnvLookup),
wired only at the CLI boundary (core/cli/run.go). The e2e suite builds its
Application in-process without that option, so the failover prober could
never resolve a remote target's api_key_env, remote liveness never passed,
and "fails over ... and fails back" hung waiting for chain-remote to
recover. Pass config.WithProxyAPIKeyEnvLookup(os.Getenv) there too, same as
the CLI. worker/federated commands don't serve proxy/failover configs and
tests/e2e-ui never sets api_key_env, so neither needs the lookup.
Also make the misconfiguration itself easier to diagnose: the prober now
logs a one-time xlog.Warn per api_key_env when a remote target sets it but
no lookup is configured, instead of only surfacing it as a per-probe
"is unset" error indistinguishable from a genuinely empty env var.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The failover prober gate calls TryAcquire with the application context,
which never ends. A database that stopped answering blocked the
scheduler goroutine for good, with the lock's mutex held. Bound the
session open and the lock query with the same timeout as Verify.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Every frontend recorded localai_failover_switches_total for the switches
it adopted from the leader, so a cluster of N frontends counted each
switch N times. Only the leader, which decides the switch, records it.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Pin and Unpin apply the change locally first for read-your-writes. When
the shared write then failed, the local pin stayed, so this frontend
served a target the others did not. Restore the previous pin on error,
unless a newer change arrived meanwhile.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A NATS reconnect re-hydrates the pins map from the DB without OnApply, so
a frontend that missed an unpin kept serving the old pin and flip-flopped
with the leader's republish. Every frontend now reconciles the manager's
pins with the shared set every ten ticks and after a reconnect, and the
pins map re-reads the DB every 30 s to repair a delta dropped without a
reconnect.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
ResourceExhausted is retried and trips the target, which is right for a
rate limit or an out-of-memory backend. A payload over the gRPC message
cap is ResourceExhausted too, but every target rejects it the same way,
so it tripped the whole chain. Classify it as a request error.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The non-OpenAI endpoints (depth, detection, face_*, voice_*, images,
video, 3d) map a backend's gRPC Unimplemented to an echo 501 without the
gRPC status. The retry loop did not see a capability gap, and IsRetryable
is false for 501, so the client got 501 and the next target was never
tried. Treat a returned or written 501 as a capability gap.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The failover chain editor leaves the warm toggle enabled on every row
(the model list has no backend field to gate on) and relies on the
server warning instead, so the docs describing it as disabled for
remote targets were wrong. Separately, syncstate's hydrate() returns
early with no Store or Loader, so a Reconcile tick is a no-op rather
than one that empties the map — correct that claim everywhere it was
repeated (contributor guide, distsync comment, design spec).
No behavior change.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Treating ResourceExhausted as a capability gap skipped the target
without counting a failure, so a target that stays rate limited or out
of memory kept its traffic. It is now an ordinary retryable failure:
the request moves to the next target and the exhausted one trips.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Rerank no longer sends top_n 0, which the upstream rejects. A
mid-stream upstream error frame now fails the call instead of ending
it as a short success. Temperature 0 is forwarded. An upstream 429
becomes ResourceExhausted, which failover skips like Unimplemented.
A localai-proxy config sends its own name upstream when upstream_model
is unset, and a chat proxy defaults to the tokenizer template so chat
reaches the upstream as messages.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A leader whose host died without closing its connection kept the
advisory lock for about two hours of OS keepalive defaults, and no other
frontend could probe. The lock session now sets short TCP keepalives and
tcp_user_timeout, so the server drops it within about 30 seconds.
Shutdown now closes the lock for good, so a tick that runs after it
cannot take the lock back.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A lock taken per tick passed between frontends on almost every tick, so
several frontends probed at once and each change of leader re-sent the
warm set and all state. The leader now holds a dedicated PostgreSQL
session with the advisory lock and keeps it until it shuts down or the
session dies.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A pin arrives from the sync layer once. Dropping it when the chain or
target is unknown here left this frontend routing differently from the
cluster whenever its config lagged or a chain was re-created.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Pass the API key environment lookup through ApplicationConfig to satisfy
core configuration lint. Keep credential resolution dynamic and exclude
the callback from serialization.
Handle the five close results reported by errcheck.
Assisted-by: Codex:gpt-6 golangci-lint
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The spec promised a load-time warning when a chain marks a remote
target warm, where the flag does nothing; the loader now logs it. The
remote-backend test moves into ModelConfig.IsRemoteProxy so the loader
and the failover manager agree on what is remote.
The spec now says what ships: a load blocked by pinned warm targets
proceeds over the limit after eviction retries, without an error that
names them.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Go resends custom headers such as x-api-key when it follows a redirect,
also to another host, so a redirecting upstream could receive the
target's API key elsewhere. The probe client now treats a redirect as
the response, which fails the probe as a non-2xx status.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A chain request reached a cloud-proxy target with the client's model,
the chain name, whenever the target set no upstream_model: passthrough
forwards the body's model and translate falls back to it. The upstream
answered 404, which neither retries nor trips, while the liveness
probe, which checks the target's own name, kept passing.
PrepareTarget now sets the upstream model of a remote target to
proxy.upstream_model or the target name, the same name the probe uses.
The request pipeline and realtime chain stages both call it.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A warm target's liveness probe called ModelLoader.Load, which blocked
until the model finished loading (while the warm preload loaded it
too). Tick waited for every probe, so all probing froze, and the probe
then ran HealthCheck on an expired context and tripped the target at
every startup.
The prober now takes a function that returns the running backend
without loading it. A target that is not loaded passes liveness; its
recovery is neither confirmed nor failed and it returns to healthy
after min_dwell, like a cold target. Tick no longer waits for probes:
each probe applies its own result and a target whose probe is running
is skipped.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The request path called HasChains on every request, and with no chains
it scanned the config source each time: the loader's lock plus a copy
and sort of every config, forever, on every installation without
chains. Sync now keeps an atomic flag and HasChains reads only that.
A chain added since the last sync is still served because Plan syncs
on a miss; only in-request retry waits for the next tick (at most 1s).
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A missing file is not a reliable signal: models download on first use,
some backends need no file, and dotted names like Phi-3.5-mini look like
paths. Marking such a fallback down removed the retry a chain exists
for. Cold targets are now judged only by real requests.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
An admission rejection or a disabled target moves the request to the
next target through Attempt.Skip, which records no failure. A 4xx
response no longer counts as a success. Requests skip body recording
when no chain is configured, and stop it once the model is known not
to be a chain.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Warm local targets are pinned in the watchdog and preloaded. Switches
and target health are exported as metrics, skipped attempts as traces.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Remote liveness uses /v1/models, which every OpenAI-compatible upstream
serves. Recovery sends one minimal request for the target's usecase.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Idle targets get a liveness probe each interval, recovering targets an
inference probe. Cold local targets are never loaded to be probed.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
recomputeLocked only fired the event on an active-target change or on
entering degraded. When the active target itself recovered while every
target was down, the chain silently left degraded with no event, so
SSE/realtime consumers tracking chain.switched.state got stuck on
"degraded".
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
errcheck flagged two side-effect-only m.Plan calls in tests, and unused
flagged close(), which Task 5's probe scheduler wires in.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Health is tracked per target and the active target per chain. Fail-back
waits for recovery probes and a minimum time on the fallback.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Reject chains whose targets are missing or are chains, at load and on
create or edit, and warn when the targets share no usecase.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
On a single-node install nothing in Operate listed the models loaded on
this machine or let an admin stop one. The System page that did was
retired in #11548, and its replacements (the Nodes workbench) only work
in distributed mode. The Nodes page also mis-detected single-node mode:
the cluster routes are not registered there, so /api/nodes answers 404,
but only 503 was treated as "distributed off", which sent every
single-node install to the empty worker-registration card. The rail hid
the entry anyway.
Nodes route on a single node becomes "This machine":
- the Nodes page's VRAM / RAM / CPU / models-disk gauges, fed from this
host by mapping /api/resources onto the worker heartbeat fields
- a memory bar splitting host RAM by running model
- a running-models table (backend, RSS, CPU share, uptime, PID) with
search, sorting, logs and a confirmed Stop
- the distributed setup behind an "Add machines" button
The Operate overview gains a "Running now" preview (heaviest five, with
Stop) on single node and a pointer to Nodes > Running models on a
cluster. The rail shows "This machine" in Runtime with a running count.
Backend, additive only:
- /system: each loaded model carries a `process` block (pid, rss_bytes,
memory_percent, cpu_percent, started_at). A sampler keeps one gopsutil
handle per PID so CPU is the share since the previous poll rather than
the lifetime average; it is omitted on the first reading.
- /api/resources: host `cpu` and models-path `disk`, the same readings
workers send in their heartbeat.
Also fixes the fleet tables widening the page on phones: the headers'
absolutely positioned sr-only labels escaped the scroll wrapper.
Assisted-by: Claude:claude-opus-5 [Playwright]
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
* feat(oci): pull ORAS artifacts into a directory
Galleries published to an OCI registry arrive as an ORAS artifact whose
layers carry their tree path in org.opencontainers.image.title. Nothing in
pkg/oci could read a non-image manifest, so add PullArtifact: it lays the
layers out under a destination directory and returns the manifest digest so
callers can pin and verify signatures on it.
A registry is remote input and this writes files, so the manifest is
validated before any byte is fetched. A title that is absolute, escapes the
destination or is missing is refused, an artifact of an unexpected
artifactType is refused, and both the layer count and the total size are
capped with caller-settable limits.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Assisted-by: Claude:claude-opus-5 [Claude Code]
* feat(gallery): fetch and verify galleries published as OCI artifacts
A gallery whose URL is oci://host/repo:tag is now pulled as an ORAS
artifact and its index.yaml read from the unpacked tree, alongside the
existing HTTP and file sources. Mirrors, the per-candidate timeout, the
failure cooldown and the last known good copy work the same for both
schemes, so a gallery can mirror an OCI primary with an HTTP fallback.
When the gallery declares a verification policy the artifact reference is
resolved to its manifest digest, the publisher signature is checked
against that digest, and only then is the same digest pulled. Nothing
unverified is written to disk or parsed. With strict integrity mode on,
an OCI gallery without a verification policy is refused instead of
warned about, which is what the flag already does for backend installs.
The pull lands in a staging directory that is renamed into the cache only
once the whole tree is on disk and the index reads back as an index, so a
failed or interrupted pull leaves nothing a later fetch would serve. A
fetch that finds a fresh cache entry does not contact the registry.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Assisted-by: Claude:claude-opus-5 [Claude Code]
* feat(gallery): resolve entry urls relative to the gallery root
A gallery published as a self-contained tree, which is what an OCI
gallery is, names its base configs by their place in that tree. Such a
url reached the HTTP client verbatim and no entry of the gallery could
be installed.
A scheme-less entry url now resolves against the root the index came
from: the unpacked artifact for an oci:// gallery, and the directory of
the index URL otherwise. The relative path may not climb out of that
root, reusing the guard the artifact puller already applies to layer
titles, and an entry that does is dropped from the listing rather than
failing the whole gallery. An entry that names its own scheme is
untouched, so .ref indirection and local github: resolution behave as
before.
Reading an entry of an OCI gallery also needs the unpacked artifact
directory as its trusted root, since the cache is deliberately a sibling
of the models directory the downloader otherwise confines a file read
to.
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Assisted-by: Claude:claude-opus-5 [Claude Code]
* fix(oci): restrict artifact file permissions
Create artifact files with owner-only permissions to satisfy G302.
Document the existing path validation for the G304 scanner finding.
Check the permissions of downloaded layers in the artifact test.
Assisted-by: Codex:gpt-6 gosec
* fix(oci): confine artifact writes to their root
Use root-relative filesystem operations to reject escapes through existing
directory symlinks. Retain owner-only permissions and add a symlink escape
regression test, without suppressing the path traversal finding.
Assisted-by: Codex:gpt-6
---------
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
Co-authored-by: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com>
* fix(vulkan): preserve host ICD discovery for packaged backends
Add bundled Mesa manifests through VK_ADD_DRIVER_FILES instead of replacing the system driver list. Merge inherited and model-specific additive paths while preserving explicit operator overrides, with regression coverage.
Assisted-by: Codex:gpt-5 golangci-lint
Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Richard Palethorpe <io@richiejp.com>
* feat(3d): add Kimodo CPU and Vulkan animation backend
Introduce a distinct animation capability and model-described 3D operations, with a typed /3d/animate API, RPC transport, distributed media staging, permissions, and tracing.
Add a persistent kimodo.cpp adapter, skeleton GLB export, CPU/Vulkan packages, model and backend galleries, importer support, CI builds, and documentation. Adapt Studio inputs to each model and provide real-time skeleton playback, seeking, and history.
Cover backend validation, packaging, API behavior, importer inventories, distributed staging, and Studio workflows. Validate real-model CPU/Vulkan generation and deploy the integration to the local QA instance.
Assisted-by: Codex:gpt-5 golangci-lint
Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Richard Palethorpe <io@richiejp.com>
* feat(kimodocpp): adopt monolithic encoders and resident inference
Update upstream for resident weights, packed execution paths, and cached motion graphs. Default to all 32 text layers while retaining configurable streaming and legacy bundle support.
Use monolithic Q8_0 encoders by default and offer all six published quantizations through the gallery and importer. Refresh pinned hashes, tests, and documentation; remove the obsolete thread patch and ensure cached source checkouts follow the upstream pin.
Validated CPU and Vulkan generation, lower-bit streaming, gallery/importer suites, packaging, lint, and cold/warm Studio generation on localai-dev.
Assisted-by: Codex:gpt-5 golangci-lint
Assisted-by: Codex:gpt-5.6-sol
Signed-off-by: Richard Palethorpe <io@richiejp.com>
---------
Signed-off-by: Richard Palethorpe <io@richiejp.com>
The LocalAGI bump in #11985 changed state.NewAgentPool to take a
SkillsProvider and a PoolLimits value. The call site here was not
updated, so master stopped compiling and every Go job went red.
Pass the limits explicitly, mirroring LocalAGI's own defaults, so the
pool prunes conversation dumps and scheduler run history instead of
growing without end.
Assisted-by: Claude Code:claude-opus-5 [Bash] [Edit]
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
pinned: true was only honoured by the per-node watchdog. Every distributed
eviction path was pinned-blind: the router's LRU eviction (EvictLRU,
evictLRUAndFreeNode) and the replica reconciler's idle scale-down would
happily unload a pinned model — and since eviction is gated on
in_flight = 0, a pinned model became eviction-eligible the instant each
response completed. Under capacity pressure that surfaces as the backend
being freed immediately after every request (#11101).
Wire the model config loader into the router and reconciler through a new
PinnedModelResolver seam (mirroring ConcurrencyConflictResolver):
- EvictLRU passes the pinned set into FindLRUModel's query so the
next-oldest unpinned model is selected instead of the attempt failing
- evictLRUAndFreeNode filters pinned models inside its locked selection
- scaleDownIdle skips pinned models entirely: trimming to the floor still
means requests beyond the survivor's capacity pay a cold reload
Deliberate teardown (admin unload, model delete, node drain) intentionally
still applies to pinned models, as does dead-row reaping (state correction,
not eviction).
Regression specs verified to fail with the exclusion disabled.
Addresses the cluster-side eviction gap in #11101
Assisted-by: Claude Code:claude-fable-5 [Claude Code]
Signed-off-by: Adira Denis Muhando <dennisadira@gmail.com>
backend.stop was the one lifecycle subject a worker never answered. The
controller published and returned nil as soon as the local publish
succeeded, so a stop that killed nothing, and a stop that failed
outright, were indistinguishable from one that worked.
The unload endpoint calls model.unload and then StopBackend. Only the
first is acknowledged, so the endpoint answered 200 while the backend
kept running and held its VRAM, and its own "backend stop failed" branch
could never run. The worker logged the failure and nobody saw it.
Give the subject a reply. The worker now enumerates the process keys it
terminated and reports any per-process error, so StopBackend fails when
the stop failed. Resolving to nothing stays a success: stopping a backend
that is not running leaves the caller in the state it asked for, and
eviction paths stop already-gone models routinely. The empty list is what
says nothing matched, and ReportsStoppedProcesses is what makes that
emptiness trustworthy, the same way BackendDeleteReply handles it.
A worker built before this reply still receives the request and still
stops the backend, it only stays silent, so a timeout degrades to the old
assumption rather than failing every stop on a fleet mid-upgrade. Only
silence degrades: a transport error is still reported, because
UnloadRemoteModel skips its registry cleanup for a node it could not
reach and needs to keep hearing about that.
Assisted-by: Claude:claude-opus-5 golangci-lint
Saved profiles previously resolved to one audio path and transcript, so
cloning backends could not use several examples of one personality.
Store ordered audio and transcript pairs while preserving the legacy
first-reference fields. Fish Speech and audio.cpp receive all pairs,
including on distributed workers. Other backends retain their
single-reference behavior.
Assisted-by: Codex:gpt-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Add a NATS event contract and exact-residency provider for backend KV cache reports. Keep guessed request observations as the default routing source while maintaining the reported index for future producers.
Assisted-by: Codex:gpt-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Broadcast forced-disturb events and successful scale-up resets so every frontend shares the same rolling autoscale signal. Deduplicate NATS echoes, expose an origin-only Prometheus counter, and document cluster behavior.
Closes#10083
Assisted-by: Codex:gpt-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Route prefix-cache replica selection through composable filters, weighted scorers, and a replaceable picker. Preserve the existing load guard and deterministic selection policy while exposing per-model scorer weights through scheduling configuration.
Assisted-by: Codex:gpt-5 [go-vet]
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
JPEG base64 starts with /9j and was classified as an absolute path in distributed mode. Detect valid JPEG base64 by its decoded magic bytes so vision requests reach the backend unchanged while malformed path-like inputs retain existing staging behavior.
Assisted-by: Codex:gpt-5
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
Capacity guards reject symlink components. On macOS, temporary paths
start with /var, which links to /private/var, so the new staging tests
fail before exercising cleanup or capacity accounting.
Resolve the fixture directories before building guarded paths. Keep
explicit symlinks within the fixtures for containment tests.
Assisted-by: Codex:gpt-6
Co-authored-by: localai-org-maint-bot <306269227+localai-org-maint-bot@users.noreply.github.com>
Backend processes shared the host temporary directory, so crashes could leave request images and audio behind until the filesystem filled. Give each process a locked LocalAI-owned runtime, remove scratch on exit, and sweep only marked abandoned runtimes at the next start.
Also close known request error-path leaks in the Python media backends, CrispASR, LongCat Video, and stable-diffusion.cpp.
Assisted-by: Codex:gpt-5
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
* docs: design ephemeral staging retention
High-frequency camera and audio inputs can fill a worker before the current six-hour cleanup window expires.
Define a one-hour retention policy that preserves recently modified request payloads.
Assisted-by: Codex:gpt-6
* docs: make ephemeral staging request-owned
Time-based retention can still fill a worker under bursty or high-rate input. Define request-lifecycle cleanup with capacity reservation and crash recovery.
Assisted-by: Codex:gpt-6
* feat(distributed): release exact staged keys
Request inputs need transport-neutral cleanup after backend calls. Add authenticated exact-key deletion for HTTP and coordinated cache eviction before shared-object deletion for S3/NATS.
Preserve URL metacharacters as filename data, reject unsafe keys, and remove upload sidecars while pruning empty request directories.
Assisted-by: Codex:gpt-6
* fix(distributed): release staged request inputs
Ephemeral inputs remained on workers after inference completed. Release each exact key after synchronous and streaming calls, including partial staging failures.
Use a bounded cleanup context so caller cancellation cannot suppress release. Preserve caller requests and backend results when cleanup fails.
Assisted-by: Codex:gpt-6
* feat(worker): bound ephemeral staging capacity
Concurrent staging can otherwise exceed its byte limit or consume reserved filesystem headroom. Explicit states keep bytes charged through each reservation, write, and commit transition.
Use a synchronized waiter count to prove Commit blocks until bounded writers close, and retain committed baselines across re-reservation.
Assisted-by: Codex:gpt-6
* feat(worker): enforce ephemeral staging bounds
Share capacity accounting across HTTP and S3 request inputs so workers
reject uploads before exhausting their filesystem. Reconcile exact release
and crash recovery with the same guard.
Assisted-by: Codex:gpt-6
* fix(distributed): make staged release race-safe
Pin each release path component before removing request-owned inputs and sidecars. Stop pruning when a directory identity changes.
Assisted-by: Codex:gpt-6
* fix(worker): retain staged input ownership
Keep committed request inputs protected from age recovery until exact release ends their ownership. Startup-scanned files remain reclaimable and can acquire ownership through reservation.
Assisted-by: Codex:gpt-6
* fix(worker): claim cached ephemeral inputs
Keep startup-scanned cache hits owned while inference uses them and reconcile their actual size against capacity.
Assisted-by: Codex:gpt-6
* fix(distributed): enforce staging admission
Propagate multimodal staging failures before inference and claim matching ephemeral HTTP cache entries. Fall back to PUT when an older worker does not support claims.
Assisted-by: Codex:gpt-6
* fix(distributed): close staging accounting gaps
Keep unknown-length reservations charged until bytes reach disk and bound NATS release waits by the lifecycle cleanup deadline.
Assisted-by: Codex:gpt-6
* fix(distributed): restage swept cache hits
Treat files removed between cache probing and ownership claims as misses so HTTP and S3 workers can stage them again.
Assisted-by: Codex:gpt-6
* fix(distributed): release staged inputs by request
Release every input from one inference with one fixed-size worker coordination request. Fence request ingress against cleanup, bound staging capacity and cleanup state, and retain exact-key release for rolling upgrades.
Assisted-by: Codex:gpt-6
---------
Co-authored-by: Ettore Di Giacinto <mudler@localai.io>
