Skip to content

docs: rework the home page and user guide entry points - #2868

Merged
killme2008 merged 21 commits into
mainfrom
docs/home-and-user-guide-entry-points
Sep 13, 2026
Merged

docs: rework the home page and user guide entry points#2868
killme2008 merged 21 commits into
mainfrom
docs/home-and-user-guide-entry-points

Conversation

@killme2008

@killme2008 killme2008 commented Sep 11, 2026 •
edited
Loading

Copy link
Copy Markdown
Member

What changed

Three documentation entry points were table-of-contents pages that restated the sidebar in prose. This rewrites them around what a reader is trying to do, and fixes the navigation and accuracy defects found along the way.

Home page. Four intent cards, an inline Ask AI panel (embedded biel-bot, reusing the existing docusaurus-biel configuration), and Why GreptimeDB claims that lead with the conclusion. Adds the agent-facing surface — MCP Server, Skills, llms.txt, semantic layer — which was previously one bullet in a list.

User guide overview. Reorganised by task. Covers the sections it had omitted: Manage Data, semantic layer, vector storage, time zone, protocols, integrations.

Ingest Data. Replaces four unequal bullets with a table mapping each data source to the protocol it writes with and the page that covers it. Each row was verified against the target page.

Section overviews. Logs, Query Data, Deployments & Administration, Metadata Storage, WAL, Protocols, and the two ingestion scenario pages. The logs page now explains what a pipeline does — processors, transform, indexes — because no concept page covers it and this is where a reader meets the idea.

Ingestion navigation. The observability group linked a 22-word forwarding page for the InfluxDB line protocol while the 1078-word guide sat under For IoT; Kafka had the mirror problem. Both scenario groups now point at the page that holds the content. A doc id can appear in two sidebar categories, so no forwarding page is needed for reachability — the files stay in place and no URL changes.

Protocol boundaries. The Elasticsearch, Loki, Splunk, and OpenTSDB protocol pages were one sentence pointing elsewhere. Each now states its endpoints, the mapping onto tables, and what is not implemented.

Duplicated category. Two categories were both labelled "Manage Data". The maintenance one is now "Table and Region Operations", and the data-operations page is named after what it covers instead of "Overview".

Agent RCA Bench. Cited in Why GreptimeDB (40% fewer wrong diagnoses, 48% fewer input tokens read, about 45% lower cost to run) and, for Nightly only, in the semantic layer concept page. The 1.2 pages omit the semantic graph arm, which that release does not have.

Two new components, HomeCards and AskAI, are stripped by the llms plugin so no JSX reaches llms.txt, llms-full.txt, or the .md endpoints. AskAI renders nothing when BIEL_PROJECT_ID is unset, which is how the build and link-check workflows run.

Follow-up, not in this PR: the same ingestion material is split across four directories and produces a set of forwarding pages. Consolidating it changes URLs and needs its own redirect plan.

Scope

  • Documentation versions: Nightly, 1.2. Four 1.1 Chinese files are also touched: 对账 is an accounting term and was replaced with 核对两边数据 wherever it appeared, which includes the 1.1 migration guides.
  • Languages: English, Chinese

Verification

  • DOCS_LINK_CHECK=true builds for both locales (strict links and anchors)
  • A build without BIEL_PROJECT_ID, matching how CI runs: succeeds, and the Ask AI panel degrades to nothing while the section heading remains
  • pnpm test (59 passed, including new cases for the two MDX strip regexes)
  • markdownlint and typos with the CI configs; front matter parsed with the library Docusaurus uses
  • git diff --check
  • Build output checked: new and aliased anchors resolve, forwarding-page URLs still resolve, no JSX in index.md, llms-full.txt, or user-guide/overview.md
  • Breadcrumbs verified for the pages listed in two categories: both resolve to the observability group in each locale
  • A guard comparing every touched file against the base commit confirms no Nightly/1.2 difference was flattened
  • Rendering checked in the browser at 1440x1000 and 390x700, light and dark

Checklist

  • I verified the content against the applicable GreptimeDB version.
  • I updated the relevant documentation versions and languages, or explained why not.
  • I checked changed links and anchors. Renamed headings keep their published anchors through AnchorAlias; anchors whose section has no successor are deliberately not aliased. No page URL changes.
  • I updated navigation when the document structure changed.

@killme2008
docs: rework the documentation home page
ac0e898 
Replace the table-of-contents style intro with an intent-based entry
point: four navigation cards, an inline Ask AI panel, sharpened Why
GreptimeDB claims, and a section for the agent-facing surface.

Add HomeCards for the card grid and AskAI for the embedded biel-bot,
both stripped from llms.txt and the .md endpoints so no JSX reaches
agent consumers. The floating Ask AI button is hidden on pages that
embed the panel.
@killme2008
docs: cite Agent RCA Bench results
698c911 
Add the end-to-end measurement to Why GreptimeDB: 40% fewer wrong
diagnoses, 48% fewer input tokens read, and about 45% lower cost than
the same investigations run against Prometheus, Loki, and Tempo. The
1.2 page omits the semantic graph arm, which that release does not
have.

Record the semantic layer results on the concepts page for Nightly
only: fewer rows returned in all 35 eligible Discovery results, fewer
rows and calls in 11 of 12 dependency-retrieval results, and 112 of
120 correct diagnoses on service and dependency faults.
@killme2008
docs: rewrite the user guide and ingestion entry points
647cf38 
The user guide overview restated the sidebar in prose, repeating
"refer to the X Documentation" eight times. Reorganise it by task and
cover the sections it had omitted: Manage Data, the semantic layer,
vector storage, time zone, protocols, and integrations.

Ingest Data now maps each source to the protocol it writes with and
the page that covers it, so choosing a path does not require knowing
how for-observability, for-iot, protocols, and integrations divide the
same material. Published heading anchors are kept with AnchorAlias.

Concepts: use the indefinite article, matching the rest of the docs.
@cloudflare-workers-and-pages

cloudflare-workers-and-pages Bot commented Sep 11, 2026 •
edited
Loading

Copy link
Copy Markdown

Deploying greptime-docs with  Cloudflare Pages  Cloudflare Pages

Latest commit: c134dd8
Status: ✅  Deploy successful!
Preview URL: https://67be55f5.greptime-docs.pages.dev
Branch Preview URL: https://docs-home-and-user-guide-ent.greptime-docs.pages.dev

View logs

@killme2008
docs: fix ingestion navigation and a duplicated category name
1f65280 
The observability group linked a 22-word forwarding page for the
InfluxDB line protocol while the 1078-word guide sat under For IoT;
Kafka had the mirror problem. Point both scenario groups at the page
that holds the content. A doc id can appear in two sidebar categories,
so neither scenario needs a forwarding page to be reachable. The
forwarding files stay in place, so no URL changes.

Two categories were both labelled "Manage Data", one for data
operations and one for table and Region maintenance, giving readers
two identical sidebar and search entries. Rename the maintenance one.
@killme2008
docs: rewrite four more section overviews
65eff82 
For observability: group the paths by signal instead of listing them
in the order the file happened to be written, and link the InfluxDB
line protocol guide rather than the forwarding page. For IoT: replace
three paragraphs that never mention GreptimeDB behaviour, and match
the Chinese page, which already had a written list where the English
one auto-generated cards.

Query Data: add the Jaeger API, which the page omitted; fold views and
CTEs into the SQL entry they belong to; drop the "since v0.9" framing.

Deployments & Administration: flatten four levels of nesting into
four task groups, and add the maintenance and troubleshooting sections
the page never linked.
@killme2008
docs: clean up the remaining section overviews
cddc2ea 
Logs: replace the feature list and the paragraphs restating the
architecture diagram with an explanation of what a pipeline actually
does — processors parse and rewrite fields, transform decides column
types and indexes — since this page is where a reader meets the
concept and no concept page covers it.

Metadata storage: drop the generic best-practice list that held for
any database, keep the RDS recommendation and the point that the
metadata store needs its own backup and monitoring.

WAL: retitle from "Overview", and fix a Datanode link that pointed at
why-greptimedb rather than the architecture page.

Protocols: the directory holds connection protocols and ingestion
protocols, which the page now says out loud.

Manage Data: this was the only "Overview" in the user guide that is
operational content rather than a section introduction, so name it
after what it covers. The published #manage-data anchor is preserved
for the 40 references to it.
@killme2008
docs: fix calques and factual errors in the Chinese overviews
8a1b6a5 
Reading the rewritten Chinese pages line by line turned up errors a
regex pass had missed.

Two factual ones, in both locales: the user guide and Ingest Data both
pointed at Manage Data for compaction, which lives on a different page
and under a title that has since changed; Query Data claimed three
query interfaces while listing four.

Terminology: table reconciliation was rendered as 表对账, an accounting
term, where the target page already says 表元数据修复; sequence
management likewise already has a Chinese title. The same accounting
calque appeared as 对账 in the migration guides, replaced there with
核对两边数据.

The rest are calques and broken constructions: 存在同一个引擎里,
日志特有的部分在入库之前, 将其转换成行, a document as the subject of
使用, a link label starting with a stray space, and a duplicated
paragraph restating the diagram caption.
@killme2008
docs: document what each ingestion protocol actually covers
a35d182 
The Elasticsearch, Loki, Splunk, and OpenTSDB protocol pages were one
sentence pointing elsewhere, so a reader arriving from search could not
tell whether GreptimeDB supports the part of the protocol they need.

Each page now states the endpoints, the mapping onto tables, and the
boundary: the Elasticsearch integration inserts only and reads just the
_index field, Loki ingestion provides no LogQL, Splunk omits indexer
acknowledgment and ignores the channel parameter, and OpenTSDB has no
compatible query endpoint. Every claim is taken from the corresponding
ingestion guide, which the pages still link for configuration.
@killme2008
docs: address review findings on the entry-point rewrite
54cd956 
Elasticsearch: the `${index}` path segment is the default, not the
source of truth — an `_index` field in the request body overrides it.
The page said the opposite, which would mislead anyone picking a
target table for a migration.

Fluent Bit and Vector were described as carrying all three signals
through one output. Fluent Bit's HTTP output takes logs only; metrics
and traces go through the OpenTelemetry output, and metrics can also
use Remote Write. Vector has separate metrics and logs sinks and no
traces sink at all.

OpenTSDB: the ingestion table omitted the `/v1` prefix, so the path as
written returns 404. The source page had the same problem, stating the
path without the prefix and correcting it in a following note that the
Chinese page did not carry at all; both now give the full path.

The user guide claimed tables are created as data arrives without
qualification, contradicting the exception list on the ingestion page.

Renamed sections lost published anchors. Aliases are restored for the
headings that still represent the same concept; sections with no
successor are left out rather than pointed somewhere unrelated.
@killme2008
docs: tighten overclaims and terminology in the entry-point copy
82b86d9 
"Migrate without rewriting queries" promised more than the protocols
deliver: Loki ingestion has no LogQL, the Elasticsearch integration has
no Query DSL, and MySQL and PostgreSQL are wire-protocol compatible
rather than SQL compatible. The claim now names the write path and the
read path separately and points at the protocol pages for the limits.

Other corrections: a log line is unstructured text only when it is
unstructured, which the pipeline section contradicted two paragraphs
later; collectors write through an exporter, sink, or output that the
operator configures, not one the tool picks; storing traces needs no
extra processing step; and the Reference card promised configuration,
command lines, and HTTP endpoints while linking only to the SQL
reference.

Terminology: cohort is a sample group, not a queue; "the lightest
option to run" became an unquantified cost claim in Chinese; the repo
writes 链路追踪 rather than 链路; Markdown is a proper noun; and
"无 schema 写入" reads as though the database has no schema at all.
@killme2008
docs: link OTLP alongside the other protocols on the home page
c4bfcca 
Every other protocol named in that paragraph carries a link to its
ingestion or query guide; OTLP was the only one without.
@killme2008
killme2008 requested a lite review from Copilot September 13, 2026 03:42
Copilot AI reviewed Sep 13, 2026
View reviewed changes

Copilot AI left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

🟡 Changes recommended

Unresolved AskAI/HomeCards TypeScript errors and AskAI plugin-data build failure remain, along with a broken Nightly semantic-layer link and unscoped version-1.1 edits.

Get a fresh assessment by requesting another Copilot review.

Pull request overview

Reworks documentation entry points around reader intent, improves ingestion navigation, and adds embedded AI/agent-facing surfaces for Nightly and 1.2.

Changes:

  • Reorganized home, user guide, query, ingestion, protocol, and operations overviews.
  • Updated English and Chinese navigation, translations, aliases, and protocol summaries.
  • Added HomeCards, AskAI, and llms output stripping support.
File summaries
File Change
versioned_sidebars/version-1.2-sidebars.json Updated 1.2 navigation.
versioned_docs/version-1.2/user-guide/query-data/overview.md Reorganized query entry point.
versioned_docs/version-1.2/user-guide/protocols/splunk.md Added protocol summary.
versioned_docs/version-1.2/user-guide/protocols/overview.md Clarified protocol groups.
versioned_docs/version-1.2/user-guide/protocols/opentsdb.md Added protocol summary.
versioned_docs/version-1.2/user-guide/protocols/loki.md Added protocol summary.
versioned_docs/version-1.2/user-guide/protocols/elasticsearch.md Added protocol summary.
versioned_docs/version-1.2/user-guide/overview.md Reorganized user-guide entry point.
versioned_docs/version-1.2/user-guide/manage-data/overview.md Clarified data operations.
versioned_docs/version-1.2/user-guide/logs/overview.md Reorganized logs overview.
versioned_docs/version-1.2/user-guide/ingest-data/overview.md Added source-to-protocol ingestion map.
versioned_docs/version-1.2/user-guide/ingest-data/for-observability/overview.md Updated observability grouping.
versioned_docs/version-1.2/user-guide/ingest-data/for-iot/overview.md Updated IoT grouping.
versioned_docs/version-1.2/user-guide/ingest-data/for-iot/opentsdb.md Updated OpenTSDB guidance.
versioned_docs/version-1.2/user-guide/deployments-administration/wal/overview.md Updated WAL overview.
versioned_docs/version-1.2/user-guide/deployments-administration/overview.md Reorganized administration entry point.
versioned_docs/version-1.2/user-guide/deployments-administration/manage-metadata/overview.md Refined metadata overview.
versioned_docs/version-1.2/user-guide/deployments-administration/manage-data/overview.md Clarified table and Region operations.
versioned_docs/version-1.2/user-guide/concepts/why-greptimedb.md Added benchmark-backed rationale.
versioned_docs/version-1.2/user-guide/concepts/overview.md Updated concepts entry point.
versioned_docs/version-1.2/index.md Reworked 1.2 homepage.
versioned_docs/version-1.2/faq-and-others/vibecoding.md Added agent references.
src/theme/MDXComponents.js Registered new MDX components.
src/plugins/llms-txt-generator.ts Stripped interactive MDX from generated output.
src/plugins/__tests__/llms-txt-generator.test.ts Added MDX stripping tests.
src/css/custom.css Updated embedded assistant styling behavior.
src/components/HomeCards/styles.module.css Styled homepage cards.
src/components/HomeCards/index.tsx Added homepage card component.
src/components/AskAI/styles.module.css Styled AI panel.
src/components/AskAI/index.tsx Added embedded AI assistant.
sidebars.ts Updated Nightly navigation.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/query-data/overview.md Reorganized Chinese query entry point.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/protocols/splunk.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/protocols/overview.md Clarified Chinese protocol groups.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/protocols/opentsdb.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/protocols/loki.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/protocols/elasticsearch.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/overview.md Reorganized Chinese user guide.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/migrate-to-greptimedb/migrate-from-postgresql.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/migrate-to-greptimedb/migrate-from-mysql.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/migrate-to-greptimedb/migrate-from-clickhouse.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/manage-data/overview.md Updated data operations.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/ingest-data/overview.md Added Chinese ingestion map.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/ingest-data/for-observability/overview.md Updated Chinese observability grouping.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/ingest-data/for-iot/overview.md Updated Chinese IoT grouping.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/ingest-data/for-iot/opentsdb.md Updated Chinese OpenTSDB guidance.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/deployments-administration/wal/overview.md Updated Chinese WAL overview.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/deployments-administration/overview.md Reorganized Chinese administration entry point.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/deployments-administration/manage-metadata/overview.md Refined Chinese metadata overview.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/deployments-administration/manage-data/overview.md Clarified Chinese operations.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/concepts/why-greptimedb.md Added Chinese benchmark rationale.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/index.md Reworked Chinese 1.2 homepage.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/faq-and-others/vibecoding.md Added Chinese agent references.
i18n/zh/docusaurus-plugin-content-docs/version-1.2/db-cloud-shared/migrate/migrate-from-influxdb.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/version-1.2.json Updated localized version labels.
i18n/zh/docusaurus-plugin-content-docs/version-1.1/user-guide/migrate-to-greptimedb/migrate-from-postgresql.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/version-1.1/user-guide/migrate-to-greptimedb/migrate-from-mysql.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/version-1.1/user-guide/migrate-to-greptimedb/migrate-from-clickhouse.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/version-1.1/db-cloud-shared/migrate/migrate-from-influxdb.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/query-data/overview.md Reorganized Chinese query entry point.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/protocols/splunk.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/protocols/overview.md Clarified Chinese protocol groups.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/protocols/opentsdb.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/protocols/loki.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/protocols/elasticsearch.md Added Chinese protocol summary.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/overview.md Reorganized Chinese user guide.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/migrate-to-greptimedb/migrate-from-postgresql.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/migrate-to-greptimedb/migrate-from-mysql.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/migrate-to-greptimedb/migrate-from-clickhouse.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/manage-data/overview.md Updated data operations.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/logs/overview.md Reorganized Chinese logs overview.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/ingest-data/overview.md Added Chinese ingestion map.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/ingest-data/for-observability/overview.md Updated Chinese observability grouping.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/ingest-data/for-iot/overview.md Updated Chinese IoT grouping.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/ingest-data/for-iot/opentsdb.md Updated Chinese OpenTSDB guidance.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/deployments-administration/wal/overview.md Updated Chinese WAL overview.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/deployments-administration/overview.md Reorganized Chinese administration entry point.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/deployments-administration/manage-metadata/overview.md Refined Chinese metadata overview.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/deployments-administration/manage-data/overview.md Clarified Chinese operations.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/concepts/why-greptimedb.md Added Chinese benchmark rationale.
i18n/zh/docusaurus-plugin-content-docs/current/user-guide/concepts/semantic-layer.md Updated Chinese semantic-layer content.
i18n/zh/docusaurus-plugin-content-docs/current/index.md Reworked Chinese Nightly homepage.
i18n/zh/docusaurus-plugin-content-docs/current/faq-and-others/vibecoding.md Added Chinese agent references.
i18n/zh/docusaurus-plugin-content-docs/current/db-cloud-shared/migrate/migrate-from-influxdb.md Updated migration terminology.
i18n/zh/docusaurus-plugin-content-docs/current.json Updated localized Nightly labels.
docs/user-guide/query-data/overview.md Reorganized query entry point.
docs/user-guide/protocols/splunk.md Added protocol summary.
docs/user-guide/protocols/overview.md Clarified protocol groups.
docs/user-guide/protocols/opentsdb.md Added protocol summary.
docs/user-guide/protocols/loki.md Added protocol summary.
docs/user-guide/protocols/elasticsearch.md Added protocol summary.
docs/user-guide/overview.md Reorganized Nightly user guide.
docs/user-guide/manage-data/overview.md Clarified data operations.
docs/user-guide/logs/overview.md Reorganized logs overview.
docs/user-guide/ingest-data/overview.md Added source-to-protocol ingestion map.
docs/user-guide/ingest-data/for-observability/overview.md Updated observability grouping.
docs/user-guide/ingest-data/for-iot/overview.md Updated IoT grouping.
docs/user-guide/ingest-data/for-iot/opentsdb.md Updated OpenTSDB guidance.
docs/user-guide/deployments-administration/wal/overview.md Updated WAL overview.
docs/user-guide/deployments-administration/overview.md Reorganized administration entry point.
docs/user-guide/deployments-administration/manage-metadata/overview.md Refined metadata overview.
docs/user-guide/deployments-administration/manage-data/overview.md Clarified table and Region operations.
docs/user-guide/concepts/why-greptimedb.md Added benchmark-backed rationale.
docs/user-guide/concepts/semantic-layer.md Added semantic-layer context.
docs/user-guide/concepts/overview.md Updated concepts entry point.
docs/index.md Reworked Nightly homepage.
docs/faq-and-others/vibecoding.md Added agent references.
Review details

Suppressed comments (19)

docs/index.md:59

  • This is a Nightly page, but the root-relative link resolves under the stable site root (/user-guide/...). Version 1.2 has no user-guide/semantic-layer/overview.md, so this homepage link leads to a 404 instead of the Nightly semantic-layer guide. Keep the link version-relative (for example ./user-guide/semantic-layer/overview.md).
- **[Semantic layer](/user-guide/semantic-layer/overview.md)** — what each table means and how the entities behind it relate, queried with ordinary SQL.

docs/user-guide/ingest-data/for-iot/overview.md:10

  • The linked gRPC SDK overview also provides Rust, .NET, Erlang, and TypeScript clients, so saying this page only offers Go and Java clients incorrectly narrows the documented support. Please list the supported client languages or describe this as a subset.
- [gRPC SDKs](./grpc-sdks/overview.md) — Go and Java clients for writing from your own application.

docs/user-guide/ingest-data/overview.md:42

  • The exception list does not make this statement true for "all ... integrations": the integrations section also contains query/visualization tools such as Grafana and MCP that do not create tables or ingest rows. Scope the claim to the ingestion paths listed above, otherwise readers may expect unrelated integrations to auto-create schema.
All protocols and integrations support this except [SQL](./for-iot/sql.md), [Apache Flink](/user-guide/integrations/flink.md), and [Apache Spark](/user-guide/integrations/spark.md).

docs/user-guide/ingest-data/overview.md:29

  • The linked gRPC SDK overview also lists Rust, .NET, Erlang, and TypeScript clients, so limiting this entry to Go or Java incorrectly suggests those are the only available SDKs. Please list the supported client languages or describe this as a subset.
| Your own application | gRPC SDK for Go or Java | [gRPC SDKs](./for-iot/grpc-sdks/overview.md) |

docs/user-guide/overview.md:10

  • The previous ## Understanding GreptimeDB Concepts heading exposed the published #understanding-greptimedb-concepts fragment, but this rewrite adds aliases for the other removed sections without preserving this one. Add an AnchorAlias for that fragment so existing links to the concepts entry point do not break.
<AnchorAlias id="ingesting-data-based-on-your-use-case" />

i18n/zh/docusaurus-plugin-content-docs/current/index.md:59

  • This is a Nightly page, but the root-relative link resolves under the stable site root (/user-guide/...). Version 1.2 has no user-guide/semantic-layer/overview.md, so this homepage link leads to a 404 instead of the Nightly semantic-layer guide. Keep the link version-relative (for example ./user-guide/semantic-layer/overview.md).
- **[语义层](/user-guide/semantic-layer/overview.md)** —— 用普通 SQL 查询每张表的含义，以及它背后的实体和关系。

i18n/zh/docusaurus-plugin-content-docs/current/user-guide/ingest-data/for-iot/overview.md:10

  • 链接的 gRPC SDK 概览还列出了 Rust、.NET、Erlang 和 TypeScript 客户端，因此这里只写 Go 和 Java 客户端会误导读者，以为文档只支持这两种语言。请列出支持的客户端语言，或明确说明这里只是部分示例。
    i18n/zh/docusaurus-plugin-content-docs/current/user-guide/ingest-data/overview.md:42
  • 例外列表并不能让“所有协议和集成”这一表述成立：集成目录还包含 Grafana、MCP 等仅用于查询或可视化、不会创建表或写入数据的工具。请把范围限定为上面列出的写入路径，否则读者可能误以为其他集成也会自动建表。
    i18n/zh/docusaurus-plugin-content-docs/current/user-guide/ingest-data/overview.md:29
  • 链接的 gRPC SDK 概览还列出了 Rust、.NET、Erlang 和 TypeScript 客户端，因此这里只写 Go 和 Java 会误导读者，以为只有这两种 SDK 可用。请列出支持的客户端语言，或明确说明这里只是部分示例。
    i18n/zh/docusaurus-plugin-content-docs/current/user-guide/overview.md:10
  • The previous ## 理解 GreptimeDB 的概念 heading exposed the published #理解-greptimedb-的概念 fragment, but this rewrite adds aliases for the other removed sections without preserving this one. Add an AnchorAlias for that fragment so existing links to the concepts entry point do not break.
    i18n/zh/docusaurus-plugin-content-docs/version-1.1/user-guide/migrate-to-greptimedb/migrate-from-mysql.md:52
  • The PR declares its documentation scope as Nightly and 1.2, but this commit also changes this version-1.1 page and three other version-1.1 Chinese migration pages. That adds an unannounced stable-version change and was not covered by the stated scope; please revert the version-1.1 edits or explicitly expand the scope and validation to include 1.1.
    i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/ingest-data/for-iot/overview.md:10
  • 链接的 gRPC SDK 概览还列出了 Rust、.NET、Erlang 和 TypeScript 客户端，因此这里只写 Go 和 Java 客户端会误导读者，以为文档只支持这两种语言。请列出支持的客户端语言，或明确说明这里只是部分示例。
    i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/ingest-data/overview.md:42
  • 例外列表并不能让“所有协议和集成”这一表述成立：集成目录还包含 Grafana、MCP 等仅用于查询或可视化、不会创建表或写入数据的工具。请把范围限定为上面列出的写入路径，否则读者可能误以为其他集成也会自动建表。
    i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/ingest-data/overview.md:29
  • 链接的 gRPC SDK 概览还列出了 Rust、.NET、Erlang 和 TypeScript 客户端，因此这里只写 Go 和 Java 会误导读者，以为只有这两种 SDK 可用。请列出支持的客户端语言，或明确说明这里只是部分示例。
    i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/overview.md:10
  • The previous ## 理解 GreptimeDB 的概念 heading exposed the published #理解-greptimedb-的概念 fragment, but this rewrite adds aliases for the other removed sections without preserving this one. Add an AnchorAlias for that fragment so existing links to the concepts entry point do not break.
    versioned_docs/version-1.2/user-guide/ingest-data/for-iot/overview.md:10
  • The linked gRPC SDK overview also provides Rust, .NET, Erlang, and TypeScript clients, so saying this page only offers Go and Java clients incorrectly narrows the documented support. Please list the supported client languages or describe this as a subset.
    versioned_docs/version-1.2/user-guide/ingest-data/overview.md:42
  • The exception list does not make this statement true for "all ... integrations": the integrations section also contains query/visualization tools such as Grafana and MCP that do not create tables or ingest rows. Scope the claim to the ingestion paths listed above, otherwise readers may expect unrelated integrations to auto-create schema.
    versioned_docs/version-1.2/user-guide/ingest-data/overview.md:29
  • The linked gRPC SDK overview also lists Rust, .NET, Erlang, and TypeScript clients, so limiting this entry to Go or Java incorrectly suggests those are the only available SDKs. Please list the supported client languages or describe this as a subset.
    versioned_docs/version-1.2/user-guide/overview.md:10
  • The previous ## Understanding GreptimeDB Concepts heading exposed the published #understanding-greptimedb-concepts fragment, but this rewrite adds aliases for the other removed sections without preserving this one. Add an AnchorAlias for that fragment so existing links to the concepts entry point do not break.
  • Files reviewed: 108/108 changed files
  • Comments generated: 12
  • Review effort level: Lite

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

Comment thread src/components/AskAI/index.tsx Outdated
Comment thread src/components/AskAI/index.tsx
Comment thread src/components/AskAI/index.tsx
Comment thread src/components/HomeCards/index.tsx Outdated
Comment thread docs/index.md Outdated
Comment thread i18n/zh/docusaurus-plugin-content-docs/current/user-guide/protocols/overview.md Outdated
Comment thread i18n/zh/docusaurus-plugin-content-docs/version-1.2/index.md Outdated
Comment thread i18n/zh/docusaurus-plugin-content-docs/version-1.2/user-guide/protocols/overview.md Outdated
Comment thread versioned_docs/version-1.2/index.md Outdated
Comment thread versioned_docs/version-1.2/user-guide/protocols/overview.md Outdated
killme2008 and others added 7 commits September 13, 2026 12:00
@killme2008
docs: make the Ask AI welcome message say something
aab8fdd 
The welcome message spent the first line users see introducing the
assistant by the name already printed in the panel header, then
restating that a chat box accepts questions. Around it, the header
gives the name, the suggested questions give the scope, and the footer
covers fallibility and how to reach a human; the only thing left
unsaid was where answers come from, which is also what bounds them.

The home-page line above the panel now states scope rather than
repeating the same point, so the two do not say the same thing on the
same screen.
@killme2008
docs: add the WeChat group to the Chinese Ask AI footer
dda9caa 
The Chinese footer pointed only at Github Discussions and Slack. WeChat
comes first for readers on the Chinese site; both existing channels and
their links are unchanged. The English footer is untouched.
@killme2008
docs: address PR review findings
e2875a0 
The MCP Server was described as read-only for agents, but mcp.md states
the opposite in as many words: the guarantee covers execute_sql only,
while the pipeline and dashboard tools write regardless of
--allow-write. gRPC was grouped with the protocols that carry both
writes and queries, though its page documents SDK ingestion and no
query interface.

In the components, the return types referenced the global JSX
namespace that @types/react 19 no longer declares, and only `project`
was narrowed before the sibling fields were read.
@sunchanglong
feat: keep homecard much same with sidebar item card
69e0ec7
@sunchanglong
Merge branch 'docs/home-and-user-guide-entry-points' of github.com:Gr…
e6bde79 
…eptimeTeam/docs into docs/home-and-user-guide-entry-points
@sunchanglong
style: ai card
30fd403
@sunchanglong
style: ai card use system theme
230e253
@killme2008
docs: separate query endpoints from wire protocols, and widen two ent…
f7d586c 
…ries

The home page read as though PromQL and the Jaeger API were served over
the MySQL and PostgreSQL wire protocols. They have their own HTTP
endpoints, /v1/prometheus and /v1/jaeger; SQL is what arrives over the
wire protocols.

Restore #self-managed-greptimedb-deployment and #greptimedb-architecture
on the operations overview. I had judged these as having no successor
heading, but the old h2 covered the whole page, so the alias belongs at
the top of it rather than nowhere.

The gRPC SDK entries named only Go and Java; the page they link to
lists six languages.
@killme2008
Merge branch 'docs/home-and-user-guide-entry-points' of github.com:Gr…
9819c65 
…eptimeTeam/docs into docs/home-and-user-guide-entry-points
@killme2008
chore: restore the lockfile to main's version
c134dd8 
A local pnpm 9 install rewrote pnpm-lock.yaml from lockfileVersion 6.1
to 9.0. CI pins pnpm 8.6.0, which cannot read that format: the build
log shows `WARN Ignoring not compatible lockfile`, after which it
resolves every dependency afresh. The build stays green while the
lockfile no longer pins anything.

This is a docs change, so it should not touch the lockfile at all.
Moving the repo to pnpm 9 is a separate decision — the pinned 8.6.0
also warns that it contains a regression.
@killme2008
killme2008 merged commit 23adc9a into main Sep 13, 2026
7 checks passed
@killme2008
killme2008 deleted the docs/home-and-user-guide-entry-points branch September 13, 2026 15:25
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

Copilot code review Copilot
Copilot review effort, defaults to Lite
Applies to this pull request for everyone.Learn more about Copilot code review.
Copilot left review comments
@sunchanglong sunchanglong sunchanglong approved these changes
@fengjiachun fengjiachun Awaiting requested review from fengjiachun fengjiachun is a code owner
@sunng87 sunng87 Awaiting requested review from sunng87

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@killme2008 @sunchanglong