From 6f481b9d98855aba6ce40af742765d999f3ae431 Mon Sep 17 00:00:00 2001 From: Buildkite Date: Tue, 13 Oct 2026 20:05:30 +0000 Subject: [PATCH] =?UTF-8?q?=F0=9F=A4=96=20sync'ing=20generated=20docs?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- docs/admin/code-hosts/github.mdx | 2 +- docs/admin/code-hosts/gitlab.mdx | 6 +- docs/admin/config/settings.mdx | 1 + docs/admin/config/site-config.mdx | 11 +- docs/ai/models.mdx | 23 ++-- docs/cody/capabilities/supported-models.mdx | 10 +- docs/self-hosted/observability/alerts.mdx | 21 ++- docs/self-hosted/observability/dashboards.mdx | 130 ++++++++++++++++-- 8 files changed, 155 insertions(+), 49 deletions(-) diff --git a/docs/admin/code-hosts/github.mdx b/docs/admin/code-hosts/github.mdx index b44e2e5bf..eff935acd 100644 --- a/docs/admin/code-hosts/github.mdx +++ b/docs/admin/code-hosts/github.mdx @@ -462,7 +462,7 @@ GitHub connections support the following configuration options, which are specif { // If non-null, enforces GitHub repository permissions. This requires that there is an item in the [site configuration json](https://sourcegraph.com/docs/admin/config/site_config#auth-providers) `auth.providers` field, of type "github" with the same `url` field as specified in this `GitHubConnection`. "authorization": { - "groupsCacheTTL": 72, + "groupsCacheTTL": -1, "markInternalReposAsPublic": false, "syncInternalRepoPermissions": false }, diff --git a/docs/admin/code-hosts/gitlab.mdx b/docs/admin/code-hosts/gitlab.mdx index 8e55fe81d..a14cc678d 100644 --- a/docs/admin/code-hosts/gitlab.mdx +++ b/docs/admin/code-hosts/gitlab.mdx @@ -278,10 +278,12 @@ See [Internal rate limits](/admin/code-hosts/rate-limits#internal-rate-limits). // A GitLab access token with "api" scope. Can be a personal access token (PAT) or an OAuth token. If you are enabling permissions with identity provider type "username", this token should also have "sudo" scope. "token": null, - // The OAuth token expiry (Unix timestamp in seconds) + // ⚠️ DEPRECATED: DEPRECATED: This setting has no effect. The OAuth token expiry (Unix timestamp in seconds) + // DEPRECATED: This setting has no effect. The OAuth token expiry (Unix timestamp in seconds) "token.oauth.expiry": 0, - // The OAuth refresh token + // ⚠️ DEPRECATED: DEPRECATED: This setting has no effect. The OAuth refresh token + // DEPRECATED: This setting has no effect. The OAuth refresh token "token.oauth.refresh": null, // The type of the token diff --git a/docs/admin/config/settings.mdx b/docs/admin/config/settings.mdx index c27cf95b2..3f5d3257c 100644 --- a/docs/admin/config/settings.mdx +++ b/docs/admin/config/settings.mdx @@ -98,6 +98,7 @@ Settings options and their default values are shown below. "batchChangeAgentsDisabledDefaultSkills": [], "batchChangeAgentsGuidance": [], "boostRelevantRepositories": true, + "changesetsReviewPage": false, "clientSearchResultRanking": "by-zoekt-ranking", "codeInsightsRepoUI": "single-search-query", "deepSearchGuidance": [], diff --git a/docs/admin/config/site-config.mdx b/docs/admin/config/site-config.mdx index ee6c29947..b09cda700 100644 --- a/docs/admin/config/site-config.mdx +++ b/docs/admin/config/site-config.mdx @@ -577,8 +577,8 @@ All advanced configuration options and their default values are shown below. // The number of concurrent external service syncers that can run. "repoConcurrentExternalServiceSyncers": 3, - // Interval (in minutes) for checking code hosts (such as GitHub, Gitolite, etc.) for new repositories. - "repoListUpdateInterval": 1, + // Minimum interval (in minutes) between syncs of each code host (such as GitHub, Gitolite, etc.). Code hosts with no changes or with sync errors are synced less often, up to every 8 hours. + "repoListUpdateInterval": 2, // Configuration for repository purge worker. "repoPurgeWorker": { @@ -643,6 +643,11 @@ All advanced configuration options and their default values are shown below. // EXPERIMENTAL: Use content-based language detection for lang: filters (Zoekt only). "searchContentBasedLanguageDetection": false, + // Policies the Sourcegraph instance applies to the src CLI. These are served unauthenticated at /.api/src-cli/config so the CLI can read them before authenticating, so do not place sensitive values here. The allowAccessTokenAuth policy is enforced server-side. + "srcCli": { + "allowAccessTokenAuth": true + }, + // Syntax highlighting configuration // Other example values: // - { @@ -778,7 +783,7 @@ All advanced configuration options and their default values are shown below. // Enables users to change their username after account creation. Warning: setting this to be true has security implications if you have enabled (or will at any point in the future enable) repository permissions with an option that relies on username equivalency between Sourcegraph and an external service or authentication provider. Do NOT set this to true if you are using non-built-in authentication OR rely on username equivalency for repository permissions. "auth.enableUsernameChanges": false, - // When enabled, users are required to connect at least one external account to their Sourcegraph account. Site admins are exempt from this requirement. + // When enabled, users are required to connect at least one external account to their Sourcegraph account. Site admins are exempt from this requirement. To require an account for a specific code host authentication provider, set `enforceAccountConnection` on that provider in `auth.providers`. "auth.enforceExternalAccountConnection": false, // Enables OAuth 2.0 Dynamic Client Registration (RFC 7591) for the Sourcegraph identity provider. When enabled, OAuth clients can self-register programmatically instead of requiring manual pre-configuration. This is used for MCP clients; if `mcp.enabled` is `false`, this setting is treated as `false`. diff --git a/docs/ai/models.mdx b/docs/ai/models.mdx index 3aab9f801..2db4b2a80 100644 --- a/docs/ai/models.mdx +++ b/docs/ai/models.mdx @@ -5,16 +5,17 @@ This reference lists the models currently used by [Sourcegraph AI features](/ai) {/* BEGIN GENERATED: ai-models-reference */} {/* DO NOT EDIT: generated via: sg generate ai-docs */} -| Model | Used by | -| --------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -| Claude Sonnet 5 | [Deep Search](/deep-search) | -| GPT-6 Luna | Deep Search categorization, Deep Search answer summaries, Agentic Batch Changes categorization, Agentic Batch Changes diagnostics, Coding agent low Finder, Coding agent medium Finder, Coding agent high Finder | -| GPT-5.6 Luna | Deep Search titles, Deep Search compaction, [Code Finder](/api/mcp#code-finder), [Diff Tour](/diff-tour), Agentic Batch Changes titles, Agentic Batch Changes workspace analysis, Agentic Batch Changes schema help | -| GPT-5.4 Nano | [Smart Hover](/code-navigation/smart-hover) | -| Gemini 3.5 Flash-Lite | [Query Assist](/code-search/query-assist), Agentic Batch Changes repository search | -| Claude Opus 5.5 | [Agentic Batch Changes](/agentic-batch-changes), Agentic Batch Changes compaction | -| GPT-6.1 Sol | Coding agent low, Coding agent medium | -| GPT-6 Astra | Coding agent high, Coding agent low Oracle, Coding agent medium Oracle, Coding agent high Oracle | -| GPT Transcribe | [Dictation](https://sourcegraph.com/changelog/agent-dictation) | +| Model | Used by | +| --------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | +| GPT-5.6 Luna | Change request file grouping, Deep Search compaction, [Code Finder](/api/mcp#code-finder), [Diff Tour](/diff-tour), Agentic Batch Changes workspace analysis, Agentic Batch Changes schema help | +| Claude Sonnet 5 | [Deep Search](/deep-search) | +| GPT-6 Luna | Deep Search categorization, Deep Search answer summaries, Deep Search titles, Agentic Batch Changes titles, Agentic Batch Changes categorization, Agentic Batch Changes diagnostics, Coding agent low Finder, Coding agent medium Finder, Coding agent high Finder | +| GPT-5.4 Nano | [Smart Hover](/code-navigation/smart-hover) | +| Gemini 3.5 Flash-Lite | [Query Assist](/code-search/query-assist), Agentic Batch Changes repository search | +| Claude Opus 5.5 | [Agentic Batch Changes](/agentic-batch-changes), Agentic Batch Changes compaction | +| GPT-6.1 Sol | Coding agent low, Coding agent medium | +| Gemini 4 Argon | Coding agent high | +| GPT-6 Astra | Coding agent low Oracle, Coding agent medium Oracle, Coding agent high Oracle | +| GPT Transcribe | [Dictation](https://sourcegraph.com/changelog/agent-dictation) | {/* END GENERATED: ai-models-reference */} \ No newline at end of file diff --git a/docs/cody/capabilities/supported-models.mdx b/docs/cody/capabilities/supported-models.mdx index 94ec54820..9012b3787 100644 --- a/docs/cody/capabilities/supported-models.mdx +++ b/docs/cody/capabilities/supported-models.mdx @@ -18,8 +18,8 @@ Cody supports a variety of cutting-edge large language models for use in chat an | Anthropic | [Claude Opus 5](https://platform.claude.com/docs/en/about-claude/models/whats-new-opus-5) | ✅ | ✅ | | Anthropic | [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview) | ✅ | ✅ | | Anthropic | [Claude Sonnet 5](https://platform.claude.com/docs/en/about-claude/models/whats-new-sonnet-5) | ✅ | ✅ | -| Anthropic | [Claude Haiku 4.5](https://platform.claude.com/docs/en/about-claude/models/overview) | ✅ | ✅ | -| Anthropic | [Claude Haiku 4.5 with Thinking](https://platform.claude.com/docs/en/about-claude/models/overview) | ✅ | ✅ | +| Anthropic | [Claude Haiku 5.5](https://platform.claude.com/docs/en/models/haiku-5-5/overview) | ✅ | ✅ | +| Google | [Gemini 4 Argon](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/4-argon) | ✅ | ✅ | | Google | [Gemini 3.8 Flash](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-8-flash) | ✅ | ✅ | | Google | [Gemini 3.5 Flash-Lite](https://docs.cloud.google.com/gemini-enterprise-agent-platform/models/gemini/3-5-flash-lite) | ✅ | ✅ | | Google | [Gemini 3.1 Pro](https://cloud.google.com/vertex-ai/generative-ai/docs/models/gemini/3-1-pro) | ✅ (beta) | ❌ | @@ -39,8 +39,6 @@ Cody supports a variety of cutting-edge large language models for use in chat an Cody uses a set of models for autocomplete which are suited for the low latency use case. | **Provider** | **Model** | **Status** | **Hosting region** | | :----------- | :-------- | :--------- | :----------------- | -| Anthropic | [Claude Haiku 4.5](https://platform.claude.com/docs/en/about-claude/models/overview) | ✅ | global | -| Anthropic | [Claude Haiku 4.5 with Thinking](https://platform.claude.com/docs/en/about-claude/models/overview) | ✅ | global | | Fireworks.ai | DeepSeek V2 Lite Base | ✅ | US Iowa | | Fireworks.ai | Autoedits Fine-Tune DeepSeek Coder V2 | ✅ | US Iowa | @@ -57,6 +55,6 @@ The following models are used by default for each feature when no specific model | **Feature** | **Default Model** | | :---------- | :---------------- | -| Chat | Claude Sonnet 5 | +| Chat | Claude Sonnet 5.5 | | Autocomplete | DeepSeek V2 Lite Base | -| Fast Chat | Claude Haiku 4.5 | +| Fast Chat | Claude Haiku 5.5 | diff --git a/docs/self-hosted/observability/alerts.mdx b/docs/self-hosted/observability/alerts.mdx index af4f85787..0d6b41e12 100644 --- a/docs/self-hosted/observability/alerts.mdx +++ b/docs/self-hosted/observability/alerts.mdx @@ -3198,6 +3198,7 @@ Generated query for critical alert: `(min((sum(src_worker_jobs{job=~"^worker.*", **Next steps** - Completed, unexpired SCIP uploads finished 1-96h ago are not represented in the persisted commit graph fast path. +- Uploads whose commit is no longer reachable from any ref (for example, on a deleted branch) are excluded, so the remaining uploads are expected to become visible. - Check whether the codeintel commit graph updater worker is running and healthy. - Check worker logs for commit graph update failures. - Inspect codeintel commit graph freshness and upload processing panels for related backlog or errors. @@ -3876,7 +3877,7 @@ Generated query for warning alert: `min((max by (name) (src_github_rate_limit_re **Descriptions** -- critical worker: less than 30 remaining calls to GitLab rest API before hitting the rate limit +- warning worker: less than 30 remaining calls to GitLab rest API before hitting the rate limit **Next steps** @@ -3886,7 +3887,7 @@ Generated query for warning alert: `min((max by (name) (src_github_rate_limit_re ```json "observability.silenceAlerts": [ - "critical_worker_gitlab_rest_rate_limit_remaining" + "warning_worker_gitlab_rest_rate_limit_remaining" ] ``` @@ -3895,7 +3896,7 @@ Generated query for warning alert: `min((max by (name) (src_github_rate_limit_re
Technical details -Generated query for critical alert: `min((max by (name) (src_gitlab_rate_limit_remaining{resource="rest"})) <= 30)` +Generated query for warning alert: `min((max by (name) (src_gitlab_rate_limit_remaining{resource="rest"})) <= 30)`
@@ -7368,8 +7369,7 @@ Custom query for critical alert: `max(((sum(increase(src_deepsearch_question_pro **Descriptions** -- warning deepsearch: 3+ questions taking longer than 5 minutes over 15m -- critical deepsearch: 10+ questions taking longer than 5 minutes over 15m +- warning deepsearch: 3+ questions taking longer than 5 minutes over 15m for 10m0s **Next steps** @@ -7381,8 +7381,7 @@ Custom query for critical alert: `max(((sum(increase(src_deepsearch_question_pro ```json "observability.silenceAlerts": [ - "warning_deepsearch_deepsearch_slow_questions", - "critical_deepsearch_deepsearch_slow_questions" + "warning_deepsearch_deepsearch_slow_questions" ] ``` @@ -7391,9 +7390,7 @@ Custom query for critical alert: `max(((sum(increase(src_deepsearch_question_pro
Technical details -Generated query for warning alert: `max((sum(increase(src_deepsearch_question_processing_duration_seconds_count{operation="question"\}[15m])) - sum(increase(src_deepsearch_question_processing_duration_seconds_bucket\{le="300",operation="question"}[15m]))) > 3)` - -Generated query for critical alert: `max((sum(increase(src_deepsearch_question_processing_duration_seconds_count{operation="question"\}[15m])) - sum(increase(src_deepsearch_question_processing_duration_seconds_bucket\{le="300",operation="question"}[15m]))) > 10)` +Generated query for warning alert: `max((sum(increase(src_deepsearch_question_processing_duration_seconds_count{operation="question"\}[15m])) - sum(increase(src_deepsearch_question_processing_duration_seconds_bucket\{le=~"300([.]0)?",operation="question"}[15m]))) > 3)`
@@ -7405,7 +7402,7 @@ Generated query for critical alert: `max((sum(increase(src_deepsearch_question_p **Descriptions** -- critical deepsearch: 3+ truncated questions with no answer produced over 15m +- critical deepsearch: 5+ truncated questions with no answer produced over 15m for 10m0s **Next steps** @@ -7427,7 +7424,7 @@ Generated query for critical alert: `max((sum(increase(src_deepsearch_question_p
Technical details -Generated query for critical alert: `max((sum(increase(src_deepsearch_question_outcome_total{outcome="response_truncated_no_answer"}[15m]))) > 3)` +Custom query for critical alert: `max(((sum(increase(src_deepsearch_question_outcome_total{outcome="response_truncated_no_answer"\}[15m]))) and (sum(increase(src_deepsearch_question_outcome_total\{outcome="response_truncated_no_answer"\}[5m])) >= 2) and (sum(increase(src_deepsearch_question_outcome_total\{outcome="response_truncated_no_answer"}[15m])) / sum(increase(src_deepsearch_question_outcome_total[15m])) >= 0.05) and (sum(increase(src_deepsearch_question_outcome_total[15m])) >= 20)) >= 5)`
diff --git a/docs/self-hosted/observability/dashboards.mdx b/docs/self-hosted/observability/dashboards.mdx index b7d236dbc..07c44a2ff 100644 --- a/docs/self-hosted/observability/dashboards.mdx +++ b/docs/self-hosted/observability/dashboards.mdx @@ -11139,6 +11139,78 @@ sum(increase(src_codeintel_upload_processor_errors_total{job=~"^precise-code-int
+#### precise-code-intel-worker: codeintel_upload_errored_backlog + +

SCIP uploads in the errored state, including uploads waiting to retry

+ +These uploads remain unfinished even when retry backoff temporarily excludes them from the eligible queue age and depth metrics. The count includes all uploads in the errored state, regardless of when their next attempt is due. + +This panel has no related alerts. + +To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100020` on your Sourcegraph instance. + +*Managed by the Sourcegraph Code Understanding team.* + +
+Technical details + +Query: + +``` +max(src_workerutil_queue_errored{domain="precise_code_intel_upload_worker"}) +``` +
+ +
+ +#### precise-code-intel-worker: codeintel_upload_completion_attempts + +

SCIP upload completion attempts over 30m

+ +Successful store calls, which can include duplicate or no-op reports and therefore do not necessarily count distinct uploads. + +This panel has no related alerts. + +To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100021` on your Sourcegraph instance. + +*Managed by the Sourcegraph Code Understanding team.* + +
+Technical details + +Query: + +``` +sum(increase(src_workerutil_dbworker_store_total{domain="codeintel_upload",op="MarkComplete"}[30m])) +``` +
+ +
+ +#### precise-code-intel-worker: codeintel_upload_error_attempts + +

SCIP upload error attempts over 30m

+ +Successful store calls, including retryable errors and possible duplicate or no-op reports. Compare with completion attempts and errored backlog before treating a falling eligible queue age as recovery. + +This panel has no related alerts. + +To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100022` on your Sourcegraph instance. + +*Managed by the Sourcegraph Code Understanding team.* + +
+Technical details + +Query: + +``` +sum(increase(src_workerutil_dbworker_store_total{domain="codeintel_upload",op="MarkErrored"}[30m])) +``` +
+ +
+ #### precise-code-intel-worker: codeintel_upload_processor_active_upload_runtime

Runtime of the oldest SCIP upload actively being processed by each worker instance

@@ -11148,7 +11220,7 @@ to see how much headroom uploads have before hitting PRECISE_CODE_INTEL_WORKER_M This panel has no related alerts. -To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100020` on your Sourcegraph instance. +To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100030` on your Sourcegraph instance. *Managed by the Sourcegraph Code Understanding team.* @@ -11171,7 +11243,7 @@ max by(instance) ((time() - src_codeintel_upload_processor_oldest_start_time_sec Refer to the [alerts reference](alerts#precise-code-intel-worker-codeintel_upload_processor_runtime_limit_exceeded) for 1 alert related to this panel. -To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100021` on your Sourcegraph instance. +To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100031` on your Sourcegraph instance. *Managed by the Sourcegraph Code Understanding team.* @@ -11193,7 +11265,7 @@ sum(increase(src_codeintel_upload_processor_maximum_runtime_exceeded_total{job=" Refer to the [alerts reference](alerts#precise-code-intel-worker-codeintel_upload_processor_runtime_over_limit) for 1 alert related to this panel. -To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100022` on your Sourcegraph instance. +To see this panel, visit `/-/debug/grafana/d/precise-code-intel-worker/precise-code-intel-worker?viewPanel=100032` on your Sourcegraph instance. *Managed by the Sourcegraph Code Understanding team.* @@ -19356,16 +19428,18 @@ Query:
-#### searcher: store_zip_prepare_duration +#### searcher: store_archive_prepare_duration -

Average archive preparation duration by cache outcome over 5m

+

Average archive preparation duration by format and cache outcome over 5m

This graph measures the average time needed to make a repository archive -available to a search. Cache misses include fetching code from gitserver and -should be slower than hits. Compare this with the fetch queue to distinguish -dependency or cache pressure from local search execution. The average is used -because the current histogram`s highest finite bucket is 10 seconds, while a -large archive fetch can take minutes. +available to a search, split by archive format. Structural search always uses +ZIP archives; other unindexed searches use paired archives when +SEARCHER_USE_PAIRED_ARCHIVES is enabled and ZIP archives otherwise. Cache misses +include fetching code from gitserver and should be slower than hits. Compare +this with the fetch queue to distinguish dependency or cache pressure from +local search execution. The average is used because the histograms` highest +finite bucket is 10 seconds, while a large archive fetch can take minutes. This panel has no related alerts. @@ -19379,7 +19453,35 @@ To see this panel, visit `/-/debug/grafana/d/searcher/searcher?viewPanel=100111` Query: ``` -sum by (cache_hit) (rate(searcher_store_zip_prepare_duration_sum{instance=~`${instance:regex}`}[5m])) / sum by (cache_hit) (rate(searcher_store_zip_prepare_duration_count{instance=~`${instance:regex}`}[5m])) +label_replace(sum by (cache_hit) (rate(searcher_store_zip_prepare_duration_sum{instance=~`${instance:regex}`}[5m])) / sum by (cache_hit) (rate(searcher_store_zip_prepare_duration_count{instance=~`${instance:regex}`}[5m])), "format", "zip", "", "") or label_replace(sum by (cache_hit) (rate(searcher_store_archive_prepare_duration_sum{instance=~`${instance:regex}`}[5m])) / sum by (cache_hit) (rate(searcher_store_archive_prepare_duration_count{instance=~`${instance:regex}`}[5m])), "format", "paired", "", "") +``` + + +
+ +#### searcher: store_archive_rebuilds + +

Paired search archives rebuilt after failing validation every 5m (per instance)

+ +This graph should remain zero. Each rebuild means a cached paired archive was +truncated or corrupt, so searcher discarded it and fetched the repository again. +Occasional rebuilds after a crash or node failure are expected. Sustained +rebuilds point to disk problems or an archive writer bug; inspect searcher logs +for `rebuilding invalid search archive`. + +This panel has no related alerts. + +To see this panel, visit `/-/debug/grafana/d/searcher/searcher?viewPanel=100112` on your Sourcegraph instance. + +*Managed by the Sourcegraph Code Understanding team.* + +
+Technical details + +Query: + +``` +sum by (instance) (increase(searcher_store_archive_rebuild_total{instance=~`${instance:regex}`}[5m])) ```
@@ -36956,7 +37058,7 @@ sum by (entitlement) (increase(src_deepsearch_question_usage_limited_total[15m]) Number of deep search questions that took longer than 5 minutes in the last 15 minutes. -Refer to the [alerts reference](alerts#deepsearch-deepsearch_slow_questions) for 2 alerts related to this panel. +Refer to the [alerts reference](alerts#deepsearch-deepsearch_slow_questions) for 1 alert related to this panel. To see this panel, visit `/-/debug/grafana/d/deepsearch/deepsearch?viewPanel=100040` on your Sourcegraph instance. @@ -36968,7 +37070,7 @@ To see this panel, visit `/-/debug/grafana/d/deepsearch/deepsearch?viewPanel=100 Query: ``` -sum(increase(src_deepsearch_question_processing_duration_seconds_count{operation="question"}[15m])) - sum(increase(src_deepsearch_question_processing_duration_seconds_bucket{le="300",operation="question"}[15m])) +sum(increase(src_deepsearch_question_processing_duration_seconds_count{operation="question"}[15m])) - sum(increase(src_deepsearch_question_processing_duration_seconds_bucket{le=~"300([.]0)?",operation="question"}[15m])) ``` @@ -38090,7 +38192,7 @@ histogram_quantile(0.95, sum(rate(src_batch_change_agent_wake_duration_seconds_b

Pending agent wakes

-Number of non-idle threads with unread inbox items whose wake is retried once a minute. Most are threads waiting on an ask_user question, which only the user can unblock; pending wakes well above "active message handlers" means threads are stuck in processing with no running job, so check the "Stalled job resetters" group. +Number of mid-turn threads with unread inbox items whose wake is retried once a minute. A wake for a thread waiting on an ask_user question or tool approval is dropped instead: an answer, an approval, or a new message resumes the agent with a hint to read the inbox, and a cancel resumes nothing. Pending wakes well above "active message handlers" means threads are stuck in processing with no running job, so check the "Stalled job resetters" group. This panel has no related alerts.