rcourtman 5c65f65a90 Pass keep_alive=30s to Ollama so the model unloads between Patrol runs ...

Ollama keeps the loaded model in RAM for 5 minutes by default after
each request, and every new request refreshes that 5-minute window.
Pulse never passed keep_alive, so any Ollama traffic (Patrol, alert
analysis, AI chat) within 5 minutes of the previous request kept the
model warm — and on a server with continuous Pulse activity that
meant the model never unloaded, even with Patrol set to a 24-hour
interval (Fixes #1425).

Pass keep_alive=30s on every Chat and ChatStream request. Short
enough that the model unloads shortly after a Patrol burst or
one-shot analysis ends, long enough to span the small gaps between
sequential calls within a single analysis session (so the model is
not reloaded mid-burst).

Tests assert that both the streaming and non-streaming Chat paths
include the keep_alive field in the Ollama request body.

2026-04-30 10:59:04 +01:00

..

adapters

fix(ai): fsync knowledge store temp file before rename to prevent empty reads

2026-02-18 13:27:47 +00:00

approval

test: add memory regression coverage for AI stores

2026-02-04 19:56:12 +00:00

baseline

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

chat

Add Ollama basic auth support

2026-04-01 12:00:38 +01:00

circuit

test: expand backend coverage

2026-01-25 21:08:44 +00:00

context

fix: Update runtime config when toggling Docker update actions setting

2026-01-03 11:14:17 +00:00

correlation

test: add AI memory regression coverage

2026-02-04 19:46:20 +00:00

cost

Fix Gemini cost estimation tiers (#1360)

2026-03-25 09:55:17 +00:00

eval

fix(test): remove flaky content type test in eval

2026-02-02 19:26:24 +00:00

finding

refactor(ai): extract shared Finding type to eliminate duplicate structs

2026-02-01 23:26:20 +00:00

forecast

test: expand backend coverage

2026-01-25 21:08:44 +00:00

investigation

Route patrol investigations through patrol model (#1360)

2026-03-26 09:16:38 +00:00

knowledge

Harden storage path handling for CodeQL

2026-03-31 08:55:45 +01:00

learning

test: add AI memory regression coverage

2026-02-04 19:46:20 +00:00

memory

Harden alert history and tenant storage paths

2026-03-31 09:23:03 +01:00

patterns

test: add AI memory regression coverage

2026-02-04 19:46:20 +00:00

providers

Pass keep_alive=30s to Ollama so the model unloads between Patrol runs

2026-04-30 10:59:04 +01:00

proxmox

Tighten integer and allocation bounds for CodeQL

2026-03-31 09:50:11 +01:00

remediation

fix(patrol): improve stream subscriber safety and detect stuck runs

2026-02-01 18:07:51 +00:00

safety

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

tools

Fail closed on unknown AI read commands

2026-04-09 14:33:54 +01:00

unified

feat(api): add POST /api/ai/patrol/undismiss endpoint to revert suppressed findings (#1300)

2026-03-01 22:29:36 +00:00

alert_adapter_test.go

feat: add AI Intelligence system with investigation and forecasting

2026-01-24 22:41:43 +00:00

alert_adapter.go

feat: enhance AI patrol with streaming and improved findings

2026-01-22 22:30:35 +00:00

alert_provider_additional_test.go

test: expand backend coverage

2026-01-25 21:08:44 +00:00

alert_provider_test.go

fix: AI Commands toggle now updates immediately in UI. Related to #952

2025-12-29 13:56:29 +00:00

alert_provider.go

feat: enhance AI patrol with streaming and improved findings

2026-01-22 22:30:35 +00:00

alert_threshold_adapter_test.go

fix: Docker container update detection showing false positives

2025-12-29 13:49:04 +00:00

alert_threshold_adapter.go

refactor(ai): Remove over-engineered URL discovery service

2025-12-10 08:35:24 +00:00

alert_triggered_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

alert_triggered.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

baseline_adapter_test.go

fix: Docker container update detection showing false positives

2025-12-29 13:49:04 +00:00

baseline_adapter.go

Enhance devcontainer and CI workflows

2026-01-01 22:29:15 +00:00

baseline_exports.go

feat(ai): Wire baseline learning loop into router startup

2025-12-12 11:29:47 +00:00

config.go

AI Problem Solver implementation and various fixes

2025-12-06 23:46:08 +00:00

correlation_exports.go

feat(ai): Add multi-resource correlation detection (Phase 6)

2025-12-12 14:26:10 +00:00

cost_persistence_test.go

fix: Docker container update detection showing false positives

2025-12-29 13:49:04 +00:00

cost_persistence.go

feat(ai): Add enriched context with historical trends and predictions

2025-12-12 09:45:57 +00:00

coverage_increase_test.go

test(ai): update tests for license gate removals and DeepSeek cleanup

2026-02-01 18:08:02 +00:00

demo_test.go

feat: add AI Intelligence system with investigation and forecasting

2026-01-24 22:41:43 +00:00

demo.go

feat: add AI Intelligence system with investigation and forecasting

2026-01-24 22:41:43 +00:00

discovery_adapter_test.go

Unify Proxmox discovery results

2026-02-04 13:46:56 +00:00

discovery_adapter.go

Unify Proxmox discovery results

2026-02-04 13:46:56 +00:00

exports_test.go

test(ai): improve AI package test coverage from 59.7% to 69.5%

2025-12-19 21:53:06 +00:00

findings_additional_test.go

test: expand backend coverage

2026-01-25 21:08:44 +00:00

findings_clustering_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

findings_coverage_test.go

fix(patrol): address multiple issues in patrol service

2026-01-02 12:45:00 +00:00

findings_dedup_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

findings_investigation_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

findings_persistence_test.go

fix: AI Commands toggle now updates immediately in UI. Related to #952

2025-12-29 13:56:29 +00:00

findings_persistence.go

feat: add AI Intelligence system with investigation and forecasting

2026-01-24 22:41:43 +00:00

findings_test.go

feat(patrol): implement patrol findings, evaluation, and investigation logic

2026-01-31 16:23:08 +00:00

findings_timeout_test.go

feat(patrol): add timeout handling and shorter retry cooldown

2026-02-01 10:12:23 +00:00

findings_tools_adapter_test.go

refactor(ai): Rename findings adapter and add chat patrol alias

2026-01-19 19:20:49 +00:00

findings_tools_adapter.go

refactor(ai): Rename findings adapter and add chat patrol alias

2026-01-19 19:20:49 +00:00

findings.go

feat(api): add POST /api/ai/patrol/undismiss endpoint to revert suppressed findings (#1300)

2026-03-01 22:29:36 +00:00

incident_coordinator_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

incident_coordinator.go

feat: add AI Intelligence system with investigation and forecasting

2026-01-24 22:41:43 +00:00

intelligence_coverage_test.go

Improve internal package test coverage

2025-12-29 17:25:21 +00:00

intelligence_test.go

refactor: fix lint issues in internal/ai package

2026-01-02 19:53:01 +00:00

intelligence.go

feat: add AI Intelligence system with investigation and forecasting

2026-01-24 22:41:43 +00:00

investigation_adapter.go

refactor(ai): extract shared Finding type to eliminate duplicate structs

2026-02-01 23:26:20 +00:00

kubernetes_analysis_test.go

Enhance devcontainer and CI workflows

2026-01-01 22:29:15 +00:00

kubernetes_analysis.go

fix: Allow Host Agent thresholds to be set to 0 to disable alerting. Related to #864

2025-12-20 20:25:20 +00:00

memory_exports.go

feat(ai): Add operational memory (Phase 3) - change detection and remediation logging

2025-12-12 13:49:37 +00:00

metadata_provider_test.go

fix: Docker container update detection showing false positives

2025-12-29 13:49:04 +00:00

metadata_provider.go

fix: add missing HandleLicenseFeatures method and related changes

2025-12-19 22:59:52 +00:00

metrics_history_adapter_test.go

fix: Docker container update detection showing false positives

2025-12-29 13:49:04 +00:00

metrics_history_adapter.go

feat(ai): Add enriched context with historical trends and predictions

2025-12-12 09:45:57 +00:00

mock_test.go

Improve internal package test coverage

2025-12-29 17:25:21 +00:00

patrol_ai_additional_test.go

docs: add comprehensive deep-dive documentation for AI subsystems

2026-02-02 10:29:07 +00:00

patrol_ai_eval_additional_test.go

Add tests for guest intelligence and reachability signals

2026-02-04 14:12:50 +00:00

patrol_ai_intelligence_test.go

test(ai): add patrol AI analysis and intelligence tests

2026-02-02 11:15:07 +00:00

patrol_ai_more_test.go

Adapt Patrol retry budgets to provider context limits (#1370)

2026-03-27 10:57:14 +00:00

patrol_ai_test.go

fix(ai): filter DeepSeek DSML internal function-call format from responses

2026-02-01 18:07:41 +00:00

patrol_ai.go

Adapt Patrol retry budgets to provider context limits (#1370)

2026-03-27 10:57:14 +00:00

patrol_alerts_review_test.go

test(ai): add patrol alert review, resource state, and scoped run tests

2026-02-02 11:15:19 +00:00

patrol_context_additional_test.go

Retry Patrol with reduced seed context on provider window errors (#1370)

2026-03-26 23:16:28 +00:00

patrol_coverage_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

patrol_e2e_integration_test.go

refactor(ai): extract shared Finding type to eliminate duplicate structs

2026-02-01 23:26:20 +00:00

patrol_findings_additional_test.go

docs: add comprehensive deep-dive documentation for AI subsystems

2026-02-02 10:29:07 +00:00

patrol_findings.go

refactor(ai): extract shared Finding type to eliminate duplicate structs

2026-02-01 23:26:20 +00:00

patrol_history_memory_regression_test.go

test: add memory regression coverage for AI stores

2026-02-04 19:56:12 +00:00

patrol_history_persistence_test.go

Cap patrol history and event correlation allocations

2026-03-31 09:37:49 +01:00

patrol_history_persistence.go

Tighten integer and allocation bounds for CodeQL

2026-03-31 09:50:11 +01:00

patrol_init_additional_test.go

Skip patrol triggers when patrol is disabled (#1258)

2026-03-25 11:33:34 +00:00

patrol_init.go

Skip patrol triggers when patrol is disabled (#1258)

2026-03-25 11:33:34 +00:00

patrol_integration_test.go

fix(patrol): improve service lifecycle, graceful shutdown, and concurrency

2026-02-01 16:27:25 +00:00

patrol_intelligence_test.go

Add tests for guest intelligence and reachability signals

2026-02-04 14:12:50 +00:00

patrol_intelligence.go

fix: probe all guest IPs for reachability, not just first

2026-02-10 21:46:11 +00:00

patrol_lifecycle_integration_test.go

fix(patrol): improve service lifecycle, graceful shutdown, and concurrency

2026-02-01 16:27:25 +00:00

patrol_metrics_additional_test.go

docs: add comprehensive deep-dive documentation for AI subsystems

2026-02-02 10:29:07 +00:00

patrol_metrics.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

patrol_prober.go

Add unified guest intelligence to patrol seed context

2026-02-04 14:08:57 +00:00

patrol_reconcile_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

patrol_resource_state_test.go

test(ai): add patrol alert review, resource state, and scoped run tests

2026-02-02 11:15:19 +00:00

patrol_run_additional_test.go

Add patrol trigger disabled-state coverage

2026-04-05 23:35:54 +01:00

patrol_run_test.go

fix: address AI patrol concurrency and streaming issues

2026-02-03 14:39:00 +00:00

patrol_run.go

fix(ui): add Pro badge to Reporting tab and reduce patrol trigger log noise (#1285, #1258)

2026-02-26 21:09:13 +00:00

patrol_scoped_run_test.go

test(ai): add patrol alert review, resource state, and scoped run tests

2026-02-02 11:15:19 +00:00

patrol_signals_test.go

Filter virtual disks from SMART checks to prevent false positives (#1329)

2026-03-08 22:16:24 +00:00

patrol_signals.go

Filter virtual disks from SMART checks to prevent false positives (#1329)

2026-03-08 22:16:24 +00:00

patrol_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

patrol_thresholds_test.go

feat: add AI Intelligence system with investigation and forecasting

2026-01-24 22:41:43 +00:00

patrol_triggers_test.go

fix: address AI patrol concurrency and streaming issues

2026-02-03 14:39:00 +00:00

patrol_triggers.go

fix: address AI patrol concurrency and streaming issues

2026-02-03 14:39:00 +00:00

patrol.go

fix(patrol): cap per-run tokens and reset patrol session history

2026-02-24 11:29:47 +00:00

pattern_exports.go

feat(ai): Add failure pattern detection for predictive intelligence (Phase 5)

2025-12-12 14:11:28 +00:00

resource_context_test.go

Improve internal package test coverage

2025-12-29 17:25:21 +00:00

resource_context.go

refactor: fix lint issues in internal/ai package

2026-01-02 19:53:01 +00:00

routing_test.go

Improve internal package test coverage

2025-12-29 17:25:21 +00:00

routing.go

Enhance devcontainer and CI workflows

2026-01-01 22:29:15 +00:00

sanitize_test.go

refactor(ai): Remove over-engineered URL discovery service

2025-12-10 08:35:24 +00:00

service_coverage_imp_test.go

Fix oversized AI discovery responses

2026-04-12 22:33:48 +01:00

service_extended_test.go

feat(backend): implement AI Patrol, Investigation, and system-wide refactors

2026-01-30 19:02:14 +00:00

service_remediation_test.go

Enhance devcontainer and CI workflows

2026-01-01 22:29:15 +00:00

service_test.go

Enhance devcontainer and CI workflows

2026-01-01 22:29:15 +00:00

service_tools_test.go

Cover blocked AI fetch hosts

2026-02-04 13:54:32 +00:00

service.go

Align infra discovery with Patrol interval

2026-04-17 11:10:14 +01:00

target_host_test.go

Enhance devcontainer and CI workflows

2026-01-01 22:29:15 +00:00