refactor: sanitize cache config in DatanodeOptions and StandaloneOptions

Signed-off-by: zyy17 <zyylsxm@gmail.com>

.github/scripts/deploy-greptimedb.sh

@@ -7,8 +7,6 @@ KUBERNETES_VERSION="${KUBERNETES_VERSION:-v1.32.0}"
 ENABLE_STANDALONE_MODE="${ENABLE_STANDALONE_MODE:-true}"
 DEFAULT_INSTALL_NAMESPACE=${DEFAULT_INSTALL_NAMESPACE:-default}
 GREPTIMEDB_IMAGE_TAG=${GREPTIMEDB_IMAGE_TAG:-latest}
-GREPTIMEDB_OPERATOR_IMAGE_TAG=${GREPTIMEDB_OPERATOR_IMAGE_TAG:-v0.5.1}
-GREPTIMEDB_INITIALIZER_IMAGE_TAG="${GREPTIMEDB_OPERATOR_IMAGE_TAG}"
 GREPTIME_CHART="https://greptimeteam.github.io/helm-charts/"
 ETCD_CHART="oci://registry-1.docker.io/bitnamicharts/etcd"
 ETCD_CHART_VERSION="${ETCD_CHART_VERSION:-12.0.8}"
@@ -60,7 +58,7 @@ function deploy_greptimedb_operator() {
   # Use the latest chart and image.
   helm upgrade --install greptimedb-operator greptime/greptimedb-operator \
     --create-namespace \
-    --set image.tag="$GREPTIMEDB_OPERATOR_IMAGE_TAG" \
+    --set image.tag=latest \
     -n "$DEFAULT_INSTALL_NAMESPACE"
 
   # Wait for greptimedb-operator to be ready.
@@ -80,7 +78,6 @@ function deploy_greptimedb_cluster() {
   helm upgrade --install "$cluster_name" greptime/greptimedb-cluster \
     --create-namespace \
     --set image.tag="$GREPTIMEDB_IMAGE_TAG" \
-    --set initializer.tag="$GREPTIMEDB_INITIALIZER_IMAGE_TAG" \
     --set meta.backendStorage.etcd.endpoints="etcd.$install_namespace:2379" \
     --set meta.backendStorage.etcd.storeKeyPrefix="$cluster_name" \
     -n "$install_namespace"
@@ -118,7 +115,6 @@ function deploy_greptimedb_cluster_with_s3_storage() {
   helm upgrade --install "$cluster_name" greptime/greptimedb-cluster -n "$install_namespace" \
     --create-namespace \
     --set image.tag="$GREPTIMEDB_IMAGE_TAG" \
-    --set initializer.tag="$GREPTIMEDB_INITIALIZER_IMAGE_TAG" \
     --set meta.backendStorage.etcd.endpoints="etcd.$install_namespace:2379" \
     --set meta.backendStorage.etcd.storeKeyPrefix="$cluster_name" \
     --set objectStorage.s3.bucket="$AWS_CI_TEST_BUCKET" \

.github/scripts/package-lock.json

@@ -1,507 +0,0 @@
-{
-  "name": "greptimedb-github-scripts",
-  "version": "1.0.0",
-  "lockfileVersion": 3,
-  "requires": true,
-  "packages": {
-    "": {
-      "name": "greptimedb-github-scripts",
-      "version": "1.0.0",
-      "dependencies": {
-        "@octokit/rest": "^21.0.0",
-        "axios": "^1.7.0"
-      }
-    },
-    "node_modules/@octokit/auth-token": {
-      "version": "5.1.2",
-      "resolved": "https://registry.npmjs.org/@octokit/auth-token/-/auth-token-5.1.2.tgz",
-      "integrity": "sha512-JcQDsBdg49Yky2w2ld20IHAlwr8d/d8N6NiOXbtuoPCqzbsiJgF633mVUw3x4mo0H5ypataQIX7SFu3yy44Mpw==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 18"
-      }
-    },
-    "node_modules/@octokit/core": {
-      "version": "6.1.6",
-      "resolved": "https://registry.npmjs.org/@octokit/core/-/core-6.1.6.tgz",
-      "integrity": "sha512-kIU8SLQkYWGp3pVKiYzA5OSaNF5EE03P/R8zEmmrG6XwOg5oBjXyQVVIauQ0dgau4zYhpZEhJrvIYt6oM+zZZA==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/auth-token": "^5.0.0",
-        "@octokit/graphql": "^8.2.2",
-        "@octokit/request": "^9.2.3",
-        "@octokit/request-error": "^6.1.8",
-        "@octokit/types": "^14.0.0",
-        "before-after-hook": "^3.0.2",
-        "universal-user-agent": "^7.0.0"
-      },
-      "engines": {
-        "node": ">= 18"
-      }
-    },
-    "node_modules/@octokit/endpoint": {
-      "version": "10.1.4",
-      "resolved": "https://registry.npmjs.org/@octokit/endpoint/-/endpoint-10.1.4.tgz",
-      "integrity": "sha512-OlYOlZIsfEVZm5HCSR8aSg02T2lbUWOsCQoPKfTXJwDzcHQBrVBGdGXb89dv2Kw2ToZaRtudp8O3ZIYoaOjKlA==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/types": "^14.0.0",
-        "universal-user-agent": "^7.0.2"
-      },
-      "engines": {
-        "node": ">= 18"
-      }
-    },
-    "node_modules/@octokit/graphql": {
-      "version": "8.2.2",
-      "resolved": "https://registry.npmjs.org/@octokit/graphql/-/graphql-8.2.2.tgz",
-      "integrity": "sha512-Yi8hcoqsrXGdt0yObxbebHXFOiUA+2v3n53epuOg1QUgOB6c4XzvisBNVXJSl8RYA5KrDuSL2yq9Qmqe5N0ryA==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/request": "^9.2.3",
-        "@octokit/types": "^14.0.0",
-        "universal-user-agent": "^7.0.0"
-      },
-      "engines": {
-        "node": ">= 18"
-      }
-    },
-    "node_modules/@octokit/openapi-types": {
-      "version": "25.1.0",
-      "resolved": "https://registry.npmjs.org/@octokit/openapi-types/-/openapi-types-25.1.0.tgz",
-      "integrity": "sha512-idsIggNXUKkk0+BExUn1dQ92sfysJrje03Q0bv0e+KPLrvyqZF8MnBpFz8UNfYDwB3Ie7Z0TByjWfzxt7vseaA==",
-      "license": "MIT"
-    },
-    "node_modules/@octokit/plugin-paginate-rest": {
-      "version": "11.6.0",
-      "resolved": "https://registry.npmjs.org/@octokit/plugin-paginate-rest/-/plugin-paginate-rest-11.6.0.tgz",
-      "integrity": "sha512-n5KPteiF7pWKgBIBJSk8qzoZWcUkza2O6A0za97pMGVrGfPdltxrfmfF5GucHYvHGZD8BdaZmmHGz5cX/3gdpw==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/types": "^13.10.0"
-      },
-      "engines": {
-        "node": ">= 18"
-      },
-      "peerDependencies": {
-        "@octokit/core": ">=6"
-      }
-    },
-    "node_modules/@octokit/plugin-paginate-rest/node_modules/@octokit/openapi-types": {
-      "version": "24.2.0",
-      "resolved": "https://registry.npmjs.org/@octokit/openapi-types/-/openapi-types-24.2.0.tgz",
-      "integrity": "sha512-9sIH3nSUttelJSXUrmGzl7QUBFul0/mB8HRYl3fOlgHbIWG+WnYDXU3v/2zMtAvuzZ/ed00Ei6on975FhBfzrg==",
-      "license": "MIT"
-    },
-    "node_modules/@octokit/plugin-paginate-rest/node_modules/@octokit/types": {
-      "version": "13.10.0",
-      "resolved": "https://registry.npmjs.org/@octokit/types/-/types-13.10.0.tgz",
-      "integrity": "sha512-ifLaO34EbbPj0Xgro4G5lP5asESjwHracYJvVaPIyXMuiuXLlhic3S47cBdTb+jfODkTE5YtGCLt3Ay3+J97sA==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/openapi-types": "^24.2.0"
-      }
-    },
-    "node_modules/@octokit/plugin-request-log": {
-      "version": "5.3.1",
-      "resolved": "https://registry.npmjs.org/@octokit/plugin-request-log/-/plugin-request-log-5.3.1.tgz",
-      "integrity": "sha512-n/lNeCtq+9ofhC15xzmJCNKP2BWTv8Ih2TTy+jatNCCq/gQP/V7rK3fjIfuz0pDWDALO/o/4QY4hyOF6TQQFUw==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 18"
-      },
-      "peerDependencies": {
-        "@octokit/core": ">=6"
-      }
-    },
-    "node_modules/@octokit/plugin-rest-endpoint-methods": {
-      "version": "13.5.0",
-      "resolved": "https://registry.npmjs.org/@octokit/plugin-rest-endpoint-methods/-/plugin-rest-endpoint-methods-13.5.0.tgz",
-      "integrity": "sha512-9Pas60Iv9ejO3WlAX3maE1+38c5nqbJXV5GrncEfkndIpZrJ/WPMRd2xYDcPPEt5yzpxcjw9fWNoPhsSGzqKqw==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/types": "^13.10.0"
-      },
-      "engines": {
-        "node": ">= 18"
-      },
-      "peerDependencies": {
-        "@octokit/core": ">=6"
-      }
-    },
-    "node_modules/@octokit/plugin-rest-endpoint-methods/node_modules/@octokit/openapi-types": {
-      "version": "24.2.0",
-      "resolved": "https://registry.npmjs.org/@octokit/openapi-types/-/openapi-types-24.2.0.tgz",
-      "integrity": "sha512-9sIH3nSUttelJSXUrmGzl7QUBFul0/mB8HRYl3fOlgHbIWG+WnYDXU3v/2zMtAvuzZ/ed00Ei6on975FhBfzrg==",
-      "license": "MIT"
-    },
-    "node_modules/@octokit/plugin-rest-endpoint-methods/node_modules/@octokit/types": {
-      "version": "13.10.0",
-      "resolved": "https://registry.npmjs.org/@octokit/types/-/types-13.10.0.tgz",
-      "integrity": "sha512-ifLaO34EbbPj0Xgro4G5lP5asESjwHracYJvVaPIyXMuiuXLlhic3S47cBdTb+jfODkTE5YtGCLt3Ay3+J97sA==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/openapi-types": "^24.2.0"
-      }
-    },
-    "node_modules/@octokit/request": {
-      "version": "9.2.4",
-      "resolved": "https://registry.npmjs.org/@octokit/request/-/request-9.2.4.tgz",
-      "integrity": "sha512-q8ybdytBmxa6KogWlNa818r0k1wlqzNC+yNkcQDECHvQo8Vmstrg18JwqJHdJdUiHD2sjlwBgSm9kHkOKe2iyA==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/endpoint": "^10.1.4",
-        "@octokit/request-error": "^6.1.8",
-        "@octokit/types": "^14.0.0",
-        "fast-content-type-parse": "^2.0.0",
-        "universal-user-agent": "^7.0.2"
-      },
-      "engines": {
-        "node": ">= 18"
-      }
-    },
-    "node_modules/@octokit/request-error": {
-      "version": "6.1.8",
-      "resolved": "https://registry.npmjs.org/@octokit/request-error/-/request-error-6.1.8.tgz",
-      "integrity": "sha512-WEi/R0Jmq+IJKydWlKDmryPcmdYSVjL3ekaiEL1L9eo1sUnqMJ+grqmC9cjk7CA7+b2/T397tO5d8YLOH3qYpQ==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/types": "^14.0.0"
-      },
-      "engines": {
-        "node": ">= 18"
-      }
-    },
-    "node_modules/@octokit/rest": {
-      "version": "21.1.1",
-      "resolved": "https://registry.npmjs.org/@octokit/rest/-/rest-21.1.1.tgz",
-      "integrity": "sha512-sTQV7va0IUVZcntzy1q3QqPm/r8rWtDCqpRAmb8eXXnKkjoQEtFe3Nt5GTVsHft+R6jJoHeSiVLcgcvhtue/rg==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/core": "^6.1.4",
-        "@octokit/plugin-paginate-rest": "^11.4.2",
-        "@octokit/plugin-request-log": "^5.3.1",
-        "@octokit/plugin-rest-endpoint-methods": "^13.3.0"
-      },
-      "engines": {
-        "node": ">= 18"
-      }
-    },
-    "node_modules/@octokit/types": {
-      "version": "14.1.0",
-      "resolved": "https://registry.npmjs.org/@octokit/types/-/types-14.1.0.tgz",
-      "integrity": "sha512-1y6DgTy8Jomcpu33N+p5w58l6xyt55Ar2I91RPiIA0xCJBXyUAhXCcmZaDWSANiha7R9a6qJJ2CRomGPZ6f46g==",
-      "license": "MIT",
-      "dependencies": {
-        "@octokit/openapi-types": "^25.1.0"
-      }
-    },
-    "node_modules/asynckit": {
-      "version": "0.4.0",
-      "resolved": "https://registry.npmjs.org/asynckit/-/asynckit-0.4.0.tgz",
-      "integrity": "sha512-Oei9OH4tRh0YqU3GxhX79dM/mwVgvbZJaSNaRk+bshkj0S5cfHcgYakreBjrHwatXKbz+IoIdYLxrKim2MjW0Q==",
-      "license": "MIT"
-    },
-    "node_modules/axios": {
-      "version": "1.12.2",
-      "resolved": "https://registry.npmjs.org/axios/-/axios-1.12.2.tgz",
-      "integrity": "sha512-vMJzPewAlRyOgxV2dU0Cuz2O8zzzx9VYtbJOaBgXFeLc4IV/Eg50n4LowmehOOR61S8ZMpc2K5Sa7g6A4jfkUw==",
-      "license": "MIT",
-      "dependencies": {
-        "follow-redirects": "^1.15.6",
-        "form-data": "^4.0.4",
-        "proxy-from-env": "^1.1.0"
-      }
-    },
-    "node_modules/before-after-hook": {
-      "version": "3.0.2",
-      "resolved": "https://registry.npmjs.org/before-after-hook/-/before-after-hook-3.0.2.tgz",
-      "integrity": "sha512-Nik3Sc0ncrMK4UUdXQmAnRtzmNQTAAXmXIopizwZ1W1t8QmfJj+zL4OA2I7XPTPW5z5TDqv4hRo/JzouDJnX3A==",
-      "license": "Apache-2.0"
-    },
-    "node_modules/call-bind-apply-helpers": {
-      "version": "1.0.2",
-      "resolved": "https://registry.npmjs.org/call-bind-apply-helpers/-/call-bind-apply-helpers-1.0.2.tgz",
-      "integrity": "sha512-Sp1ablJ0ivDkSzjcaJdxEunN5/XvksFJ2sMBFfq6x0ryhQV/2b/KwFe21cMpmHtPOSij8K99/wSfoEuTObmuMQ==",
-      "license": "MIT",
-      "dependencies": {
-        "es-errors": "^1.3.0",
-        "function-bind": "^1.1.2"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/combined-stream": {
-      "version": "1.0.8",
-      "resolved": "https://registry.npmjs.org/combined-stream/-/combined-stream-1.0.8.tgz",
-      "integrity": "sha512-FQN4MRfuJeHf7cBbBMJFXhKSDq+2kAArBlmRBvcvFE5BB1HZKXtSFASDhdlz9zOYwxh8lDdnvmMOe/+5cdoEdg==",
-      "license": "MIT",
-      "dependencies": {
-        "delayed-stream": "~1.0.0"
-      },
-      "engines": {
-        "node": ">= 0.8"
-      }
-    },
-    "node_modules/delayed-stream": {
-      "version": "1.0.0",
-      "resolved": "https://registry.npmjs.org/delayed-stream/-/delayed-stream-1.0.0.tgz",
-      "integrity": "sha512-ZySD7Nf91aLB0RxL4KGrKHBXl7Eds1DAmEdcoVawXnLD7SDhpNgtuII2aAkg7a7QS41jxPSZ17p4VdGnMHk3MQ==",
-      "license": "MIT",
-      "engines": {
-        "node": ">=0.4.0"
-      }
-    },
-    "node_modules/dunder-proto": {
-      "version": "1.0.1",
-      "resolved": "https://registry.npmjs.org/dunder-proto/-/dunder-proto-1.0.1.tgz",
-      "integrity": "sha512-KIN/nDJBQRcXw0MLVhZE9iQHmG68qAVIBg9CqmUYjmQIhgij9U5MFvrqkUL5FbtyyzZuOeOt0zdeRe4UY7ct+A==",
-      "license": "MIT",
-      "dependencies": {
-        "call-bind-apply-helpers": "^1.0.1",
-        "es-errors": "^1.3.0",
-        "gopd": "^1.2.0"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/es-define-property": {
-      "version": "1.0.1",
-      "resolved": "https://registry.npmjs.org/es-define-property/-/es-define-property-1.0.1.tgz",
-      "integrity": "sha512-e3nRfgfUZ4rNGL232gUgX06QNyyez04KdjFrF+LTRoOXmrOgFKDg4BCdsjW8EnT69eqdYGmRpJwiPVYNrCaW3g==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/es-errors": {
-      "version": "1.3.0",
-      "resolved": "https://registry.npmjs.org/es-errors/-/es-errors-1.3.0.tgz",
-      "integrity": "sha512-Zf5H2Kxt2xjTvbJvP2ZWLEICxA6j+hAmMzIlypy4xcBg1vKVnx89Wy0GbS+kf5cwCVFFzdCFh2XSCFNULS6csw==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/es-object-atoms": {
-      "version": "1.1.1",
-      "resolved": "https://registry.npmjs.org/es-object-atoms/-/es-object-atoms-1.1.1.tgz",
-      "integrity": "sha512-FGgH2h8zKNim9ljj7dankFPcICIK9Cp5bm+c2gQSYePhpaG5+esrLODihIorn+Pe6FGJzWhXQotPv73jTaldXA==",
-      "license": "MIT",
-      "dependencies": {
-        "es-errors": "^1.3.0"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/es-set-tostringtag": {
-      "version": "2.1.0",
-      "resolved": "https://registry.npmjs.org/es-set-tostringtag/-/es-set-tostringtag-2.1.0.tgz",
-      "integrity": "sha512-j6vWzfrGVfyXxge+O0x5sh6cvxAog0a/4Rdd2K36zCMV5eJ+/+tOAngRO8cODMNWbVRdVlmGZQL2YS3yR8bIUA==",
-      "license": "MIT",
-      "dependencies": {
-        "es-errors": "^1.3.0",
-        "get-intrinsic": "^1.2.6",
-        "has-tostringtag": "^1.0.2",
-        "hasown": "^2.0.2"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/fast-content-type-parse": {
-      "version": "2.0.1",
-      "resolved": "https://registry.npmjs.org/fast-content-type-parse/-/fast-content-type-parse-2.0.1.tgz",
-      "integrity": "sha512-nGqtvLrj5w0naR6tDPfB4cUmYCqouzyQiz6C5y/LtcDllJdrcc6WaWW6iXyIIOErTa/XRybj28aasdn4LkVk6Q==",
-      "funding": [
-        {
-          "type": "github",
-          "url": "https://github.com/sponsors/fastify"
-        },
-        {
-          "type": "opencollective",
-          "url": "https://opencollective.com/fastify"
-        }
-      ],
-      "license": "MIT"
-    },
-    "node_modules/follow-redirects": {
-      "version": "1.15.11",
-      "resolved": "https://registry.npmjs.org/follow-redirects/-/follow-redirects-1.15.11.tgz",
-      "integrity": "sha512-deG2P0JfjrTxl50XGCDyfI97ZGVCxIpfKYmfyrQ54n5FO/0gfIES8C/Psl6kWVDolizcaaxZJnTS0QSMxvnsBQ==",
-      "funding": [
-        {
-          "type": "individual",
-          "url": "https://github.com/sponsors/RubenVerborgh"
-        }
-      ],
-      "license": "MIT",
-      "engines": {
-        "node": ">=4.0"
-      },
-      "peerDependenciesMeta": {
-        "debug": {
-          "optional": true
-        }
-      }
-    },
-    "node_modules/form-data": {
-      "version": "4.0.4",
-      "resolved": "https://registry.npmjs.org/form-data/-/form-data-4.0.4.tgz",
-      "integrity": "sha512-KrGhL9Q4zjj0kiUt5OO4Mr/A/jlI2jDYs5eHBpYHPcBEVSiipAvn2Ko2HnPe20rmcuuvMHNdZFp+4IlGTMF0Ow==",
-      "license": "MIT",
-      "dependencies": {
-        "asynckit": "^0.4.0",
-        "combined-stream": "^1.0.8",
-        "es-set-tostringtag": "^2.1.0",
-        "hasown": "^2.0.2",
-        "mime-types": "^2.1.12"
-      },
-      "engines": {
-        "node": ">= 6"
-      }
-    },
-    "node_modules/function-bind": {
-      "version": "1.1.2",
-      "resolved": "https://registry.npmjs.org/function-bind/-/function-bind-1.1.2.tgz",
-      "integrity": "sha512-7XHNxH7qX9xG5mIwxkhumTox/MIRNcOgDrxWsMt2pAr23WHp6MrRlN7FBSFpCpr+oVO0F744iUgR82nJMfG2SA==",
-      "license": "MIT",
-      "funding": {
-        "url": "https://github.com/sponsors/ljharb"
-      }
-    },
-    "node_modules/get-intrinsic": {
-      "version": "1.3.0",
-      "resolved": "https://registry.npmjs.org/get-intrinsic/-/get-intrinsic-1.3.0.tgz",
-      "integrity": "sha512-9fSjSaos/fRIVIp+xSJlE6lfwhES7LNtKaCBIamHsjr2na1BiABJPo0mOjjz8GJDURarmCPGqaiVg5mfjb98CQ==",
-      "license": "MIT",
-      "dependencies": {
-        "call-bind-apply-helpers": "^1.0.2",
-        "es-define-property": "^1.0.1",
-        "es-errors": "^1.3.0",
-        "es-object-atoms": "^1.1.1",
-        "function-bind": "^1.1.2",
-        "get-proto": "^1.0.1",
-        "gopd": "^1.2.0",
-        "has-symbols": "^1.1.0",
-        "hasown": "^2.0.2",
-        "math-intrinsics": "^1.1.0"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      },
-      "funding": {
-        "url": "https://github.com/sponsors/ljharb"
-      }
-    },
-    "node_modules/get-proto": {
-      "version": "1.0.1",
-      "resolved": "https://registry.npmjs.org/get-proto/-/get-proto-1.0.1.tgz",
-      "integrity": "sha512-sTSfBjoXBp89JvIKIefqw7U2CCebsc74kiY6awiGogKtoSGbgjYE/G/+l9sF3MWFPNc9IcoOC4ODfKHfxFmp0g==",
-      "license": "MIT",
-      "dependencies": {
-        "dunder-proto": "^1.0.1",
-        "es-object-atoms": "^1.0.0"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/gopd": {
-      "version": "1.2.0",
-      "resolved": "https://registry.npmjs.org/gopd/-/gopd-1.2.0.tgz",
-      "integrity": "sha512-ZUKRh6/kUFoAiTAtTYPZJ3hw9wNxx+BIBOijnlG9PnrJsCcSjs1wyyD6vJpaYtgnzDrKYRSqf3OO6Rfa93xsRg==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 0.4"
-      },
-      "funding": {
-        "url": "https://github.com/sponsors/ljharb"
-      }
-    },
-    "node_modules/has-symbols": {
-      "version": "1.1.0",
-      "resolved": "https://registry.npmjs.org/has-symbols/-/has-symbols-1.1.0.tgz",
-      "integrity": "sha512-1cDNdwJ2Jaohmb3sg4OmKaMBwuC48sYni5HUw2DvsC8LjGTLK9h+eb1X6RyuOHe4hT0ULCW68iomhjUoKUqlPQ==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 0.4"
-      },
-      "funding": {
-        "url": "https://github.com/sponsors/ljharb"
-      }
-    },
-    "node_modules/has-tostringtag": {
-      "version": "1.0.2",
-      "resolved": "https://registry.npmjs.org/has-tostringtag/-/has-tostringtag-1.0.2.tgz",
-      "integrity": "sha512-NqADB8VjPFLM2V0VvHUewwwsw0ZWBaIdgo+ieHtK3hasLz4qeCRjYcqfB6AQrBggRKppKF8L52/VqdVsO47Dlw==",
-      "license": "MIT",
-      "dependencies": {
-        "has-symbols": "^1.0.3"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      },
-      "funding": {
-        "url": "https://github.com/sponsors/ljharb"
-      }
-    },
-    "node_modules/hasown": {
-      "version": "2.0.2",
-      "resolved": "https://registry.npmjs.org/hasown/-/hasown-2.0.2.tgz",
-      "integrity": "sha512-0hJU9SCPvmMzIBdZFqNPXWa6dqh7WdH0cII9y+CyS8rG3nL48Bclra9HmKhVVUHyPWNH5Y7xDwAB7bfgSjkUMQ==",
-      "license": "MIT",
-      "dependencies": {
-        "function-bind": "^1.1.2"
-      },
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/math-intrinsics": {
-      "version": "1.1.0",
-      "resolved": "https://registry.npmjs.org/math-intrinsics/-/math-intrinsics-1.1.0.tgz",
-      "integrity": "sha512-/IXtbwEk5HTPyEwyKX6hGkYXxM9nbj64B+ilVJnC/R6B0pH5G4V3b0pVbL7DBj4tkhBAppbQUlf6F6Xl9LHu1g==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 0.4"
-      }
-    },
-    "node_modules/mime-db": {
-      "version": "1.52.0",
-      "resolved": "https://registry.npmjs.org/mime-db/-/mime-db-1.52.0.tgz",
-      "integrity": "sha512-sPU4uV7dYlvtWJxwwxHD0PuihVNiE7TyAbQ5SWxDCB9mUYvOgroQOwYQQOKPJ8CIbE+1ETVlOoK1UC2nU3gYvg==",
-      "license": "MIT",
-      "engines": {
-        "node": ">= 0.6"
-      }
-    },
-    "node_modules/mime-types": {
-      "version": "2.1.35",
-      "resolved": "https://registry.npmjs.org/mime-types/-/mime-types-2.1.35.tgz",
-      "integrity": "sha512-ZDY+bPm5zTTF+YpCrAU9nK0UgICYPT0QtT1NZWFv4s++TNkcgVaT0g6+4R2uI4MjQjzysHB1zxuWL50hzaeXiw==",
-      "license": "MIT",
-      "dependencies": {
-        "mime-db": "1.52.0"
-      },
-      "engines": {
-        "node": ">= 0.6"
-      }
-    },
-    "node_modules/proxy-from-env": {
-      "version": "1.1.0",
-      "resolved": "https://registry.npmjs.org/proxy-from-env/-/proxy-from-env-1.1.0.tgz",
-      "integrity": "sha512-D+zkORCbA9f1tdWRK0RaCR3GPv50cMxcrz4X8k5LTSUD1Dkw47mKJEZQNunItRTkWwgtaUSo1RVFRIG9ZXiFYg==",
-      "license": "MIT"
-    },
-    "node_modules/universal-user-agent": {
-      "version": "7.0.3",
-      "resolved": "https://registry.npmjs.org/universal-user-agent/-/universal-user-agent-7.0.3.tgz",
-      "integrity": "sha512-TmnEAEAsBJVZM/AADELsK76llnwcf9vMKuPz8JflO1frO8Lchitr0fNaN9d+Ap0BjKtqWqd/J17qeDnXh8CL2A==",
-      "license": "ISC"
-    }
-  }
-}

.github/scripts/package.json

@@ -1,10 +0,0 @@
-{
-  "name": "greptimedb-github-scripts",
-  "version": "1.0.0",
-  "type": "module",
-  "description": "GitHub automation scripts for GreptimeDB",
-  "dependencies": {
-    "@octokit/rest": "^21.0.0",
-    "axios": "^1.7.0"
-  }
-}

.github/scripts/pr-review-reminder.js

@@ -1,152 +0,0 @@
-// Daily PR Review Reminder Script
-// Fetches open PRs from GreptimeDB repository and sends Slack notifications
-// to PR owners and assigned reviewers to keep review process moving.
-
-(async () => {
-  const { Octokit } = await import("@octokit/rest");
-  const { default: axios } = await import('axios');
-
-  // Configuration
-  const GITHUB_TOKEN = process.env.GITHUB_TOKEN;
-  const SLACK_WEBHOOK_URL = process.env.SLACK_PR_REVIEW_WEBHOOK_URL;
-  const REPO_OWNER = "GreptimeTeam";
-  const REPO_NAME = "greptimedb";
-  const GITHUB_TO_SLACK = JSON.parse(process.env.GITHUBID_SLACKID_MAPPING || '{}');
-
-  // Debug: Print environment variable status
-  console.log("=== Environment Variables Debug ===");
-  console.log(`GITHUB_TOKEN: ${GITHUB_TOKEN ? 'Set ✓' : 'NOT SET ✗'}`);
-  console.log(`SLACK_PR_REVIEW_WEBHOOK_URL: ${SLACK_WEBHOOK_URL ? 'Set ✓' : 'NOT SET ✗'}`);
-  console.log(`GITHUBID_SLACKID_MAPPING: ${process.env.GITHUBID_SLACKID_MAPPING ? `Set ✓ (${Object.keys(GITHUB_TO_SLACK).length} mappings)` : 'NOT SET ✗'}`);
-  console.log("===================================\n");
-
-  const octokit = new Octokit({
-    auth: GITHUB_TOKEN
-  });
-
-  // Fetch all open PRs from the repository
-  async function fetchOpenPRs() {
-    try {
-      const prs = await octokit.pulls.list({
-        owner: REPO_OWNER,
-        repo: REPO_NAME,
-        state: "open",
-        per_page: 100,
-        sort: "created",
-        direction: "asc"
-      });
-      return prs.data.filter((pr) => !pr.draft);
-    } catch (error) {
-      console.error("Error fetching PRs:", error);
-      return [];
-    }
-  }
-
-  // Convert GitHub username to Slack mention or fallback to GitHub username
-  function toSlackMention(githubUser) {
-    const slackUserId = GITHUB_TO_SLACK[githubUser];
-    return slackUserId ? `<@${slackUserId}>` : `@${githubUser}`;
-  }
-
-  // Calculate days since PR was opened
-  function getDaysOpen(createdAt) {
-    const created = new Date(createdAt);
-    const now = new Date();
-    const diffMs = now - created;
-    const days = Math.floor(diffMs / (1000 * 60 * 60 * 24));
-    return days;
-  }
-
-  // Build Slack notification message from PR list
-  function buildSlackMessage(prs) {
-    if (prs.length === 0) {
-      return "*🎉 Great job! No pending PRs for review.*";
-    }
-
-    // Separate PRs by age threshold (14 days)
-    const criticalPRs = [];
-    const recentPRs = [];
-
-    prs.forEach(pr => {
-      const daysOpen = getDaysOpen(pr.created_at);
-      if (daysOpen >= 14) {
-        criticalPRs.push(pr);
-      } else {
-        recentPRs.push(pr);
-      }
-    });
-
-    const lines = [
-      `*🔍 Daily PR Review Reminder 🔍*`,
-      `Found *${criticalPRs.length}* critical PR(s) (14+ days old)\n`
-    ];
-
-    // Show critical PRs (14+ days) in detail
-    if (criticalPRs.length > 0) {
-      criticalPRs.forEach((pr, index) => {
-        const owner = toSlackMention(pr.user.login);
-        const reviewers = pr.requested_reviewers || [];
-        const reviewerMentions = reviewers.map(r => toSlackMention(r.login)).join(", ");
-        const daysOpen = getDaysOpen(pr.created_at);
-
-        const prInfo = `${index + 1}. <${pr.html_url}|#${pr.number}: ${pr.title}>`;
-        const ageInfo = `   🔴 Opened *${daysOpen}* day(s) ago`;
-        const ownerInfo = `   👤 Owner: ${owner}`;
-        const reviewerInfo = reviewers.length > 0
-          ? `   👁️ Reviewers: ${reviewerMentions}`
-          : `   👁️ Reviewers: _Not assigned yet_`;
-
-        lines.push(prInfo);
-        lines.push(ageInfo);
-        lines.push(ownerInfo);
-        lines.push(reviewerInfo);
-        lines.push(""); // Empty line between PRs
-      });
-    }
-
-    lines.push("_Let's keep the code review process moving! 🚀_");
-
-    return lines.join("\n");
-  }
-
-  // Send notification to Slack webhook
-  async function sendSlackNotification(message) {
-    if (!SLACK_WEBHOOK_URL) {
-      console.log("⚠️  SLACK_PR_REVIEW_WEBHOOK_URL not configured. Message preview:");
-      console.log("=".repeat(60));
-      console.log(message);
-      console.log("=".repeat(60));
-      return;
-    }
-
-    try {
-      const response = await axios.post(SLACK_WEBHOOK_URL, {
-        text: message
-      });
-
-      if (response.status !== 200) {
-        throw new Error(`Slack API returned status ${response.status}`);
-      }
-      console.log("Slack notification sent successfully.");
-    } catch (error) {
-      console.error("Error sending Slack notification:", error);
-      throw error;
-    }
-  }
-
-  // Main execution flow
-  async function run() {
-    console.log(`Fetching open PRs from ${REPO_OWNER}/${REPO_NAME}...`);
-    const prs = await fetchOpenPRs();
-    console.log(`Found ${prs.length} open PR(s).`);
-
-    const message = buildSlackMessage(prs);
-    console.log("Sending Slack notification...");
-    await sendSlackNotification(message);
-  }
-
-  run().catch(error => {
-    console.error("Script execution failed:", error);
-    process.exit(1);
-  });
-})();

.github/workflows/develop.yml

@@ -632,7 +632,7 @@ jobs:
       - name: Unzip binaries
         run: tar -xvf ./bins.tar.gz
       - name: Run sqlness
-        run: RUST_BACKTRACE=1 ./bins/sqlness-runner bare ${{ matrix.mode.opts }} -c ./tests/cases --bins-dir ./bins --preserve-state
+        run: RUST_BACKTRACE=1 ./bins/sqlness-runner ${{ matrix.mode.opts }} -c ./tests/cases --bins-dir ./bins --preserve-state
       - name: Upload sqlness logs
         if: failure()
         uses: actions/upload-artifact@v4

.github/workflows/pr-review-reminder.yml

@@ -1,36 +0,0 @@
-name: PR Review Reminder
-
-on:
-  schedule:
-    # Run at 9:00 AM UTC+8 (01:00 AM UTC) on Monday, Wednesday, Friday
-    - cron: '0 1 * * 1,3,5'
-  workflow_dispatch:
-
-jobs:
-  pr-review-reminder:
-    name: Send PR Review Reminders
-    runs-on: ubuntu-latest
-    permissions:
-      contents: read
-      pull-requests: read
-    if: ${{ github.repository == 'GreptimeTeam/greptimedb' }}
-    steps:
-      - name: Checkout repository
-        uses: actions/checkout@v4
-
-      - name: Setup Node.js
-        uses: actions/setup-node@v4
-        with:
-          node-version: '20'
-
-      - name: Install dependencies
-        working-directory: .github/scripts
-        run: npm ci
-
-      - name: Run PR review reminder
-        working-directory: .github/scripts
-        run: node pr-review-reminder.js
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-          SLACK_PR_REVIEW_WEBHOOK_URL: ${{ vars.SLACK_PR_REVIEW_WEBHOOK_URL }}
-          GITHUBID_SLACKID_MAPPING: ${{ vars.GITHUBID_SLACKID_MAPPING }}

.github/workflows/semantic-pull-request.yml

@@ -1,7 +1,7 @@
 name: "Semantic Pull Request"
 
 on:
-  pull_request_target:
+  pull_request:
     types:
       - opened
       - reopened
@@ -12,9 +12,9 @@ concurrency:
   cancel-in-progress: true
 
 permissions:
-  contents: read
-  pull-requests: write
   issues: write
+  contents: write
+  pull-requests: write
 
 jobs:
   check:

Cargo.toml

@@ -61,7 +61,6 @@ members = [
     "src/promql",
     "src/puffin",
     "src/query",
-    "src/standalone",
     "src/servers",
     "src/session",
     "src/sql",
@@ -99,12 +98,12 @@ rust.unexpected_cfgs = { level = "warn", check-cfg = ['cfg(tokio_unstable)'] }
 # See for more detaiils: https://github.com/rust-lang/cargo/issues/11329
 ahash = { version = "0.8", features = ["compile-time-rng"] }
 aquamarine = "0.6"
-arrow = { version = "56.2", features = ["prettyprint"] }
+arrow = { version = "56.0", features = ["prettyprint"] }
-arrow-array = { version = "56.2", default-features = false, features = ["chrono-tz"] }
+arrow-array = { version = "56.0", default-features = false, features = ["chrono-tz"] }
-arrow-buffer = "56.2"
+arrow-buffer = "56.0"
-arrow-flight = "56.2"
+arrow-flight = "56.0"
-arrow-ipc = { version = "56.2", default-features = false, features = ["lz4", "zstd"] }
+arrow-ipc = { version = "56.0", default-features = false, features = ["lz4", "zstd"] }
-arrow-schema = { version = "56.2", features = ["serde"] }
+arrow-schema = { version = "56.0", features = ["serde"] }
 async-stream = "0.3"
 async-trait = "0.1"
 # Remember to update axum-extra, axum-macros when updating axum
@@ -121,21 +120,19 @@ chrono = { version = "0.4", features = ["serde"] }
 chrono-tz = "0.10.1"
 clap = { version = "4.4", features = ["derive"] }
 config = "0.13.0"
-const_format = "0.2"
 crossbeam-utils = "0.8"
 dashmap = "6.1"
-datafusion = "50"
+datafusion = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-common = "50"
+datafusion-common = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-expr = "50"
+datafusion-expr = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-functions = "50"
+datafusion-functions = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-functions-aggregate-common = "50"
+datafusion-functions-aggregate-common = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-optimizer = "50"
+datafusion-optimizer = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-orc = "0.5"
+datafusion-orc = { git = "https://github.com/GreptimeTeam/datafusion-orc", rev = "a0a5f902158f153119316eaeec868cff3fc8a99d" }
-datafusion-pg-catalog = "0.12.1"
+datafusion-physical-expr = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-physical-expr = "50"
+datafusion-physical-plan = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-physical-plan = "50"
+datafusion-sql = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-sql = "50"
+datafusion-substrait = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "7d5214512740b4dfb742b6b3d91ed9affcc2c9d0" }
-datafusion-substrait = "50"
 deadpool = "0.12"
 deadpool-postgres = "0.14"
 derive_builder = "0.20"
@@ -148,7 +145,7 @@ etcd-client = { git = "https://github.com/GreptimeTeam/etcd-client", rev = "f62d
 fst = "0.4.7"
 futures = "0.3"
 futures-util = "0.3"
-greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "14b9dc40bdc8288742b0cefc7bb024303b7429ef" }
+greptime-proto = { git = "https://github.com/GreptimeTeam/greptime-proto.git", rev = "f9836cf8aab30e672f640c6ef4c1cfd2cf9fbc36" }
 hex = "0.4"
 http = "1"
 humantime = "2.1"
@@ -177,11 +174,8 @@ opentelemetry-proto = { version = "0.30", features = [
     "logs",
 ] }
 ordered-float = { version = "4.3", features = ["serde"] }
-otel-arrow-rust = { git = "https://github.com/GreptimeTeam/otel-arrow", rev = "2d64b7c0fa95642028a8205b36fe9ea0b023ec59", features = [
-    "server",
-] }
 parking_lot = "0.12"
-parquet = { version = "56.2", default-features = false, features = ["arrow", "async", "object_store"] }
+parquet = { version = "56.0", default-features = false, features = ["arrow", "async", "object_store"] }
 paste = "1.0"
 pin-project = "1.0"
 pretty_assertions = "1.4.0"
@@ -192,7 +186,7 @@ prost-types = "0.13"
 raft-engine = { version = "0.4.1", default-features = false }
 rand = "0.9"
 ratelimit = "0.10"
-regex = "1.12"
+regex = "1.8"
 regex-automata = "0.4"
 reqwest = { version = "0.12", default-features = false, features = [
     "json",
@@ -208,7 +202,6 @@ rstest_reuse = "0.7"
 rust_decimal = "1.33"
 rustc-hash = "2.0"
 # It is worth noting that we should try to avoid using aws-lc-rs until it can be compiled on various platforms.
-hostname = "0.4.0"
 rustls = { version = "0.23.25", default-features = false }
 sea-query = "0.32"
 serde = { version = "1.0", features = ["derive"] }
@@ -218,7 +211,10 @@ simd-json = "0.15"
 similar-asserts = "1.6.0"
 smallvec = { version = "1", features = ["serde"] }
 snafu = "0.8"
-sqlparser = { version = "0.58.0", default-features = false, features = ["std", "visitor", "serde"] }
+sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "39e4fc94c3c741981f77e9d63b5ce8c02e0a27ea", features = [
+    "visitor",
+    "serde",
+] } # branch = "v0.55.x"
 sqlx = { version = "0.8", features = [
     "runtime-tokio-rustls",
     "mysql",
@@ -279,7 +275,6 @@ common-recordbatch = { path = "src/common/recordbatch" }
 common-runtime = { path = "src/common/runtime" }
 common-session = { path = "src/common/session" }
 common-sql = { path = "src/common/sql" }
-common-stat = { path = "src/common/stat" }
 common-telemetry = { path = "src/common/telemetry" }
 common-test-util = { path = "src/common/test-util" }
 common-time = { path = "src/common/time" }
@@ -301,6 +296,9 @@ mito-codec = { path = "src/mito-codec" }
 mito2 = { path = "src/mito2" }
 object-store = { path = "src/object-store" }
 operator = { path = "src/operator" }
+otel-arrow-rust = { git = "https://github.com/GreptimeTeam/otel-arrow", rev = "2d64b7c0fa95642028a8205b36fe9ea0b023ec59", features = [
+    "server",
+] }
 partition = { path = "src/partition" }
 pipeline = { path = "src/pipeline" }
 plugins = { path = "src/plugins" }
@@ -310,7 +308,7 @@ query = { path = "src/query" }
 servers = { path = "src/servers" }
 session = { path = "src/session" }
 sql = { path = "src/sql" }
-standalone = { path = "src/standalone" }
+stat = { path = "src/common/stat" }
 store-api = { path = "src/store-api" }
 substrait = { path = "src/common/substrait" }
 table = { path = "src/table" }
@@ -319,22 +317,6 @@ table = { path = "src/table" }
 git = "https://github.com/GreptimeTeam/greptime-meter.git"
 rev = "5618e779cf2bb4755b499c630fba4c35e91898cb"
 
-[patch.crates-io]
-datafusion = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-common = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-expr = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-functions = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-functions-aggregate-common = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-optimizer = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-physical-expr = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-physical-expr-common = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-physical-plan = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-datasource = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-sql = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-datafusion-substrait = { git = "https://github.com/GreptimeTeam/datafusion.git", rev = "fd4b2abcf3c3e43e94951bda452c9fd35243aab0" }
-sqlparser = { git = "https://github.com/GreptimeTeam/sqlparser-rs.git", rev = "4b519a5caa95472cc3988f5556813a583dd35af1" }                           # branch = "v0.58.x"
-bytes = { git = "https://github.com/discord9/bytes", rev = "1572ab22c3cbad0e9b6681d1f68eca4139322a2a" }
-
 [profile.release]
 debug = 1
 

Makefile

@@ -8,7 +8,7 @@ CARGO_BUILD_OPTS := --locked
 IMAGE_REGISTRY ?= docker.io
 IMAGE_NAMESPACE ?= greptime
 IMAGE_TAG ?= latest
-DEV_BUILDER_IMAGE_TAG ?= 2025-10-01-8fe17d43-20251011080129
+DEV_BUILDER_IMAGE_TAG ?= 2025-05-19-f55023f3-20250829091211
 BUILDX_MULTI_PLATFORM_BUILD ?= false
 BUILDX_BUILDER_NAME ?= gtbuilder
 BASE_IMAGE ?= ubuntu
@@ -169,7 +169,7 @@ nextest: ## Install nextest tools.
 
 .PHONY: sqlness-test
 sqlness-test: ## Run sqlness test.
-	cargo sqlness bare ${SQLNESS_OPTS}
+	cargo sqlness ${SQLNESS_OPTS}
 
 RUNS ?= 1
 FUZZ_TARGET ?= fuzz_alter_table

config/config.md

@@ -13,7 +13,6 @@
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
 | `default_timezone` | String | Unset | The default timezone of the server. |
-| `default_column_prefix` | String | Unset | The default column prefix for auto-created time index and value columns. |
 | `init_regions_in_background` | Bool | `false` | Initialize all regions in the background during the startup.<br/>By default, it provides services after all regions have been initialized. |
 | `init_regions_parallelism` | Integer | `16` | Parallelism of initializing regions. |
 | `max_concurrent_queries` | Integer | `0` | The maximum current queries allowed to be executed. Zero means unlimited. |
@@ -26,15 +25,12 @@
 | `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
 | `http.timeout` | String | `0s` | HTTP request timeout. Set to 0 to disable timeout. |
 | `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
-| `http.max_total_body_memory` | String | Unset | Maximum total memory for all concurrent HTTP request bodies.<br/>Set to 0 to disable the limit. Default: "0" (unlimited) |
 | `http.enable_cors` | Bool | `true` | HTTP CORS support, it's turned on by default<br/>This allows browser to access http APIs without CORS restrictions |
 | `http.cors_allowed_origins` | Array | Unset | Customize allowed origins for HTTP CORS. |
 | `http.prom_validation_mode` | String | `strict` | Whether to enable validation for Prometheus remote write requests.<br/>Available options:<br/>- strict: deny invalid UTF-8 strings (default).<br/>- lossy: allow invalid UTF-8 strings, replace invalid characters with REPLACEMENT_CHARACTER(U+FFFD).<br/>- unchecked: do not valid strings. |
 | `grpc` | -- | -- | The gRPC server options. |
 | `grpc.bind_addr` | String | `127.0.0.1:4001` | The address to bind the gRPC server. |
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
-| `grpc.max_total_message_memory` | String | Unset | Maximum total memory for all concurrent gRPC request messages.<br/>Set to 0 to disable the limit. Default: "0" (unlimited) |
-| `grpc.max_connection_age` | String | Unset | The maximum connection age for gRPC connection.<br/>The value can be a human-readable time string. For example: `10m` for ten minutes or `1h` for one hour.<br/>Refer to https://grpc.io/docs/guides/keepalive/ for more details. |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
 | `grpc.tls.cert_path` | String | Unset | Certificate file path. |
@@ -107,7 +103,7 @@
 | `storage` | -- | -- | The data storage options. |
 | `storage.data_home` | String | `./greptimedb_data` | The working home directory. |
 | `storage.type` | String | `File` | The storage type used to store the data.<br/>- `File`: the data is stored in the local file system.<br/>- `S3`: the data is stored in the S3 object storage.<br/>- `Gcs`: the data is stored in the Google Cloud Storage.<br/>- `Azblob`: the data is stored in the Azure Blob Storage.<br/>- `Oss`: the data is stored in the Aliyun OSS. |
-| `storage.enable_read_cache` | Bool | `true` | Whether to enable read cache. If not set, the read cache will be enabled by default when using object storage. |
+| `storage.enable_read_cache` | Bool | `true` | Whether to enable read cache. If not set, the read cache will be enabled by default. |
 | `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}`. An empty string means disabling. |
 | `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. If your disk space is sufficient, it is recommended to set it larger. |
 | `storage.bucket` | String | Unset | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
@@ -156,7 +152,6 @@
 | `region_engine.mito.max_concurrent_scan_files` | Integer | `384` | Maximum number of SST files to scan concurrently. |
 | `region_engine.mito.allow_stale_entries` | Bool | `false` | Whether to allow stale WAL entries read during replay. |
 | `region_engine.mito.min_compaction_interval` | String | `0m` | Minimum time interval between two compactions.<br/>To align with the old behavior, the default value is 0 (no restrictions). |
-| `region_engine.mito.default_experimental_flat_format` | Bool | `false` | Whether to enable experimental flat format as the default format. |
 | `region_engine.mito.index` | -- | -- | The options for index in Mito engine. |
 | `region_engine.mito.index.aux_path` | String | `""` | Auxiliary directory path for the index in filesystem, used to store intermediate files for<br/>creating the index and staging files for searching the index, defaults to `{data_home}/index_intermediate`.<br/>The default name for this directory is `index_intermediate` for backward compatibility.<br/><br/>This path contains two subdirectories:<br/>- `__intm`: for storing intermediate files used during creating index.<br/>- `staging`: for storing staging files used during searching index. |
 | `region_engine.mito.index.staging_size` | String | `2GB` | The max capacity of the staging directory. |
@@ -227,7 +222,6 @@
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
 | `default_timezone` | String | Unset | The default timezone of the server. |
-| `default_column_prefix` | String | Unset | The default column prefix for auto-created time index and value columns. |
 | `max_in_flight_write_bytes` | String | Unset | The maximum in-flight write bytes. |
 | `runtime` | -- | -- | The runtime options. |
 | `runtime.global_rt_size` | Integer | `8` | The number of threads to execute the runtime for global read operations. |
@@ -239,7 +233,6 @@
 | `http.addr` | String | `127.0.0.1:4000` | The address to bind the HTTP server. |
 | `http.timeout` | String | `0s` | HTTP request timeout. Set to 0 to disable timeout. |
 | `http.body_limit` | String | `64MB` | HTTP request body limit.<br/>The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.<br/>Set to 0 to disable limit. |
-| `http.max_total_body_memory` | String | Unset | Maximum total memory for all concurrent HTTP request bodies.<br/>Set to 0 to disable the limit. Default: "0" (unlimited) |
 | `http.enable_cors` | Bool | `true` | HTTP CORS support, it's turned on by default<br/>This allows browser to access http APIs without CORS restrictions |
 | `http.cors_allowed_origins` | Array | Unset | Customize allowed origins for HTTP CORS. |
 | `http.prom_validation_mode` | String | `strict` | Whether to enable validation for Prometheus remote write requests.<br/>Available options:<br/>- strict: deny invalid UTF-8 strings (default).<br/>- lossy: allow invalid UTF-8 strings, replace invalid characters with REPLACEMENT_CHARACTER(U+FFFD).<br/>- unchecked: do not valid strings. |
@@ -247,9 +240,7 @@
 | `grpc.bind_addr` | String | `127.0.0.1:4001` | The address to bind the gRPC server. |
 | `grpc.server_addr` | String | `127.0.0.1:4001` | The address advertised to the metasrv, and used for connections from outside the host.<br/>If left empty or unset, the server will automatically use the IP address of the first network interface<br/>on the host, with the same port number as the one specified in `grpc.bind_addr`. |
 | `grpc.runtime_size` | Integer | `8` | The number of server worker threads. |
-| `grpc.max_total_message_memory` | String | Unset | Maximum total memory for all concurrent gRPC request messages.<br/>Set to 0 to disable the limit. Default: "0" (unlimited) |
 | `grpc.flight_compression` | String | `arrow_ipc` | Compression mode for frontend side Arrow IPC service. Available options:<br/>- `none`: disable all compression<br/>- `transport`: only enable gRPC transport compression (zstd)<br/>- `arrow_ipc`: only enable Arrow IPC compression (lz4)<br/>- `all`: enable all compression.<br/>Default to `none` |
-| `grpc.max_connection_age` | String | Unset | The maximum connection age for gRPC connection.<br/>The value can be a human-readable time string. For example: `10m` for ten minutes or `1h` for one hour.<br/>Refer to https://grpc.io/docs/guides/keepalive/ for more details. |
 | `grpc.tls` | -- | -- | gRPC server TLS options, see `mysql.tls` section. |
 | `grpc.tls.mode` | String | `disable` | TLS mode. |
 | `grpc.tls.cert_path` | String | Unset | Certificate file path. |
@@ -387,9 +378,10 @@
 | `procedure.max_metadata_value_size` | String | `1500KiB` | Auto split large value<br/>GreptimeDB procedure uses etcd as the default metadata storage backend.<br/>The etcd the maximum size of any request is 1.5 MiB<br/>1500KiB = 1536KiB (1.5MiB) - 36KiB (reserved size of key)<br/>Comments out the `max_metadata_value_size`, for don't split large value (no limit). |
 | `procedure.max_running_procedures` | Integer | `128` | Max running procedures.<br/>The maximum number of procedures that can be running at the same time.<br/>If the number of running procedures exceeds this limit, the procedure will be rejected. |
 | `failure_detector` | -- | -- | -- |
-| `failure_detector.threshold` | Float | `8.0` | Maximum acceptable φ before the peer is treated as failed.<br/>Lower values react faster but yield more false positives. |
+| `failure_detector.threshold` | Float | `8.0` | The threshold value used by the failure detector to determine failure conditions. |
-| `failure_detector.min_std_deviation` | String | `100ms` | The minimum standard deviation of the heartbeat intervals.<br/>So tiny variations don’t make φ explode. Prevents hypersensitivity when heartbeat intervals barely vary. |
+| `failure_detector.min_std_deviation` | String | `100ms` | The minimum standard deviation of the heartbeat intervals, used to calculate acceptable variations. |
-| `failure_detector.acceptable_heartbeat_pause` | String | `10000ms` | The acceptable pause duration between heartbeats.<br/>Additional extra grace period to the learned mean interval before φ rises, absorbing temporary network hiccups or GC pauses. |
+| `failure_detector.acceptable_heartbeat_pause` | String | `10000ms` | The acceptable pause duration between heartbeats, used to determine if a heartbeat interval is acceptable. |
+| `failure_detector.first_heartbeat_estimate` | String | `1000ms` | The initial estimate of the heartbeat interval used by the failure detector. |
 | `datanode` | -- | -- | Datanode options. |
 | `datanode.client` | -- | -- | Datanode client options. |
 | `datanode.client.timeout` | String | `10s` | Operation timeout. |
@@ -442,7 +434,6 @@
 | Key | Type | Default | Descriptions |
 | --- | -----| ------- | ----------- |
 | `node_id` | Integer | Unset | The datanode identifier and should be unique in the cluster. |
-| `default_column_prefix` | String | Unset | The default column prefix for auto-created time index and value columns. |
 | `require_lease_before_startup` | Bool | `false` | Start services after regions have obtained leases.<br/>It will block the datanode start if it can't receive leases in the heartbeat from metasrv. |
 | `init_regions_in_background` | Bool | `false` | Initialize all regions in the background during the startup.<br/>By default, it provides services after all regions have been initialized. |
 | `init_regions_parallelism` | Integer | `16` | Parallelism of initializing regions. |
@@ -481,7 +472,7 @@
 | `meta_client.metadata_cache_ttl` | String | `10m` | TTL of the metadata cache. |
 | `meta_client.metadata_cache_tti` | String | `5m` | -- |
 | `wal` | -- | -- | The WAL options. |
-| `wal.provider` | String | `raft_engine` | The provider of the WAL.<br/>- `raft_engine`: the wal is stored in the local file system by raft-engine.<br/>- `kafka`: it's remote wal that data is stored in Kafka.<br/>- `noop`: it's a no-op WAL provider that does not store any WAL data.<br/>**Notes: any unflushed data will be lost when the datanode is shutdown.** |
+| `wal.provider` | String | `raft_engine` | The provider of the WAL.<br/>- `raft_engine`: the wal is stored in the local file system by raft-engine.<br/>- `kafka`: it's remote wal that data is stored in Kafka. |
 | `wal.dir` | String | Unset | The directory to store the WAL files.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.file_size` | String | `128MB` | The size of the WAL segment file.<br/>**It's only used when the provider is `raft_engine`**. |
 | `wal.purge_threshold` | String | `1GB` | The threshold of the WAL size to trigger a purge.<br/>**It's only used when the provider is `raft_engine`**. |
@@ -504,7 +495,7 @@
 | `storage.data_home` | String | `./greptimedb_data` | The working home directory. |
 | `storage.type` | String | `File` | The storage type used to store the data.<br/>- `File`: the data is stored in the local file system.<br/>- `S3`: the data is stored in the S3 object storage.<br/>- `Gcs`: the data is stored in the Google Cloud Storage.<br/>- `Azblob`: the data is stored in the Azure Blob Storage.<br/>- `Oss`: the data is stored in the Aliyun OSS. |
 | `storage.cache_path` | String | Unset | Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.<br/>A local file directory, defaults to `{data_home}`. An empty string means disabling. |
-| `storage.enable_read_cache` | Bool | `true` | Whether to enable read cache. If not set, the read cache will be enabled by default when using object storage. |
+| `storage.enable_read_cache` | Bool | `true` | Whether to enable read cache. If not set, the read cache will be enabled by default. |
 | `storage.cache_capacity` | String | Unset | The local file cache capacity in bytes. If your disk space is sufficient, it is recommended to set it larger. |
 | `storage.bucket` | String | Unset | The S3 bucket name.<br/>**It's only used when the storage type is `S3`, `Oss` and `Gcs`**. |
 | `storage.root` | String | Unset | The S3 data will be stored in the specified prefix, for example, `s3://${bucket}/${root}`.<br/>**It's only used when the storage type is `S3`, `Oss` and `Azblob`**. |
@@ -554,7 +545,6 @@
 | `region_engine.mito.max_concurrent_scan_files` | Integer | `384` | Maximum number of SST files to scan concurrently. |
 | `region_engine.mito.allow_stale_entries` | Bool | `false` | Whether to allow stale WAL entries read during replay. |
 | `region_engine.mito.min_compaction_interval` | String | `0m` | Minimum time interval between two compactions.<br/>To align with the old behavior, the default value is 0 (no restrictions). |
-| `region_engine.mito.default_experimental_flat_format` | Bool | `false` | Whether to enable experimental flat format as the default format. |
 | `region_engine.mito.index` | -- | -- | The options for index in Mito engine. |
 | `region_engine.mito.index.aux_path` | String | `""` | Auxiliary directory path for the index in filesystem, used to store intermediate files for<br/>creating the index and staging files for searching the index, defaults to `{data_home}/index_intermediate`.<br/>The default name for this directory is `index_intermediate` for backward compatibility.<br/><br/>This path contains two subdirectories:<br/>- `__intm`: for storing intermediate files used during creating index.<br/>- `staging`: for storing staging files used during searching index. |
 | `region_engine.mito.index.staging_size` | String | `2GB` | The max capacity of the staging directory. |

config/datanode.example.toml

@@ -2,10 +2,6 @@
 ## @toml2docs:none-default
 node_id = 42
 
-## The default column prefix for auto-created time index and value columns.
-## @toml2docs:none-default
-default_column_prefix = "greptime"
-
 ## Start services after regions have obtained leases.
 ## It will block the datanode start if it can't receive leases in the heartbeat from metasrv.
 require_lease_before_startup = false
@@ -122,7 +118,6 @@ metadata_cache_tti = "5m"
 ## The provider of the WAL.
 ## - `raft_engine`: the wal is stored in the local file system by raft-engine.
 ## - `kafka`: it's remote wal that data is stored in Kafka.
-## - `noop`: it's a no-op WAL provider that does not store any WAL data.<br/>**Notes: any unflushed data will be lost when the datanode is shutdown.**
 provider = "raft_engine"
 
 ## The directory to store the WAL files.
@@ -279,8 +274,8 @@ type = "File"
 ## @toml2docs:none-default
 #+ cache_path = ""
 
-## Whether to enable read cache. If not set, the read cache will be enabled by default when using object storage.
+## Whether to enable read cache. If not set, the read cache will be enabled by default.
-#+ enable_read_cache = true
+enable_read_cache = true
 
 ## The local file cache capacity in bytes. If your disk space is sufficient, it is recommended to set it larger.
 ## @toml2docs:none-default
@@ -505,9 +500,6 @@ allow_stale_entries = false
 ## To align with the old behavior, the default value is 0 (no restrictions).
 min_compaction_interval = "0m"
 
-## Whether to enable experimental flat format as the default format.
-default_experimental_flat_format = false
-
 ## The options for index in Mito engine.
 [region_engine.mito.index]
 

config/frontend.example.toml

@@ -2,10 +2,6 @@
 ## @toml2docs:none-default
 default_timezone = "UTC"
 
-## The default column prefix for auto-created time index and value columns.
-## @toml2docs:none-default
-default_column_prefix = "greptime"
-
 ## The maximum in-flight write bytes.
 ## @toml2docs:none-default
 #+ max_in_flight_write_bytes = "500MB"
@@ -35,10 +31,6 @@ timeout = "0s"
 ## The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.
 ## Set to 0 to disable limit.
 body_limit = "64MB"
-## Maximum total memory for all concurrent HTTP request bodies.
-## Set to 0 to disable the limit. Default: "0" (unlimited)
-## @toml2docs:none-default
-#+ max_total_body_memory = "1GB"
 ## HTTP CORS support, it's turned on by default
 ## This allows browser to access http APIs without CORS restrictions
 enable_cors = true
@@ -62,10 +54,6 @@ bind_addr = "127.0.0.1:4001"
 server_addr = "127.0.0.1:4001"
 ## The number of server worker threads.
 runtime_size = 8
-## Maximum total memory for all concurrent gRPC request messages.
-## Set to 0 to disable the limit. Default: "0" (unlimited)
-## @toml2docs:none-default
-#+ max_total_message_memory = "1GB"
 ## Compression mode for frontend side Arrow IPC service. Available options:
 ## - `none`: disable all compression
 ## - `transport`: only enable gRPC transport compression (zstd)
@@ -73,11 +61,6 @@ runtime_size = 8
 ## - `all`: enable all compression.
 ## Default to `none`
 flight_compression = "arrow_ipc"
-## The maximum connection age for gRPC connection.
-## The value can be a human-readable time string. For example: `10m` for ten minutes or `1h` for one hour.
-## Refer to https://grpc.io/docs/guides/keepalive/ for more details.
-## @toml2docs:none-default
-#+ max_connection_age = "10m"
 
 ## gRPC server TLS options, see `mysql.tls` section.
 [grpc.tls]

config/metasrv.example.toml

@@ -149,18 +149,20 @@ max_metadata_value_size = "1500KiB"
 max_running_procedures = 128
 
 # Failure detectors options.
-# GreptimeDB uses the Phi Accrual Failure Detector algorithm to detect datanode failures.
 [failure_detector]
-## Maximum acceptable φ before the peer is treated as failed.
+
-## Lower values react faster but yield more false positives.
+## The threshold value used by the failure detector to determine failure conditions.
 threshold = 8.0
-## The minimum standard deviation of the heartbeat intervals.
+
-## So tiny variations don’t make φ explode. Prevents hypersensitivity when heartbeat intervals barely vary.
+## The minimum standard deviation of the heartbeat intervals, used to calculate acceptable variations.
 min_std_deviation = "100ms"
-## The acceptable pause duration between heartbeats.
+
-## Additional extra grace period to the learned mean interval before φ rises, absorbing temporary network hiccups or GC pauses.
+## The acceptable pause duration between heartbeats, used to determine if a heartbeat interval is acceptable.
 acceptable_heartbeat_pause = "10000ms"
 
+## The initial estimate of the heartbeat interval used by the failure detector.
+first_heartbeat_estimate = "1000ms"
+
 ## Datanode options.
 [datanode]
 

config/standalone.example.toml

@@ -2,10 +2,6 @@
 ## @toml2docs:none-default
 default_timezone = "UTC"
 
-## The default column prefix for auto-created time index and value columns.
-## @toml2docs:none-default
-default_column_prefix = "greptime"
-
 ## Initialize all regions in the background during the startup.
 ## By default, it provides services after all regions have been initialized.
 init_regions_in_background = false
@@ -40,10 +36,6 @@ timeout = "0s"
 ## The following units are supported: `B`, `KB`, `KiB`, `MB`, `MiB`, `GB`, `GiB`, `TB`, `TiB`, `PB`, `PiB`.
 ## Set to 0 to disable limit.
 body_limit = "64MB"
-## Maximum total memory for all concurrent HTTP request bodies.
-## Set to 0 to disable the limit. Default: "0" (unlimited)
-## @toml2docs:none-default
-#+ max_total_body_memory = "1GB"
 ## HTTP CORS support, it's turned on by default
 ## This allows browser to access http APIs without CORS restrictions
 enable_cors = true
@@ -64,15 +56,6 @@ prom_validation_mode = "strict"
 bind_addr = "127.0.0.1:4001"
 ## The number of server worker threads.
 runtime_size = 8
-## Maximum total memory for all concurrent gRPC request messages.
-## Set to 0 to disable the limit. Default: "0" (unlimited)
-## @toml2docs:none-default
-#+ max_total_message_memory = "1GB"
-## The maximum connection age for gRPC connection.
-## The value can be a human-readable time string. For example: `10m` for ten minutes or `1h` for one hour.
-## Refer to https://grpc.io/docs/guides/keepalive/ for more details.
-## @toml2docs:none-default
-#+ max_connection_age = "10m"
 
 ## gRPC server TLS options, see `mysql.tls` section.
 [grpc.tls]
@@ -378,8 +361,8 @@ data_home = "./greptimedb_data"
 ## - `Oss`: the data is stored in the Aliyun OSS.
 type = "File"
 
-## Whether to enable read cache. If not set, the read cache will be enabled by default when using object storage.
+## Whether to enable read cache. If not set, the read cache will be enabled by default.
-#+ enable_read_cache = true
+enable_read_cache = true
 
 ## Read cache configuration for object storage such as 'S3' etc, it's configured by default when using object storage. It is recommended to configure it when using object storage for better performance.
 ## A local file directory, defaults to `{data_home}`. An empty string means disabling.
@@ -596,9 +579,6 @@ allow_stale_entries = false
 ## To align with the old behavior, the default value is 0 (no restrictions).
 min_compaction_interval = "0m"
 
-## Whether to enable experimental flat format as the default format.
-default_experimental_flat_format = false
-
 ## The options for index in Mito engine.
 [region_engine.mito.index]
 

docs/how-to/how-to-profile-memory.md

@@ -30,7 +30,22 @@ curl https://raw.githubusercontent.com/brendangregg/FlameGraph/master/flamegraph
 
 ## Profiling
 
-### Enable memory profiling for greptimedb binary
+### Configuration
+
+You can control heap profiling activation through configuration. Add the following to your configuration file:
+
+```toml
+[memory]
+# Whether to enable heap profiling activation during startup.
+# When enabled, heap profiling will be activated if the `MALLOC_CONF` environment variable
+# is set to "prof:true,prof_active:false". The official image adds this env variable.
+# Default is true.
+enable_heap_profiling = true
+```
+
+By default, if you set `MALLOC_CONF=prof:true,prof_active:false`, the database will enable profiling during startup. You can disable this behavior by setting `enable_heap_profiling = false` in the configuration.
+
+### Starting with environment variables
 
 Start GreptimeDB instance with environment variables:
 
@@ -42,22 +57,6 @@ MALLOC_CONF=prof:true ./target/debug/greptime standalone start
 _RJEM_MALLOC_CONF=prof:true ./target/debug/greptime standalone start
 ```
 
-### Memory profiling for greptimedb docker image
-
-We have memory profiling enabled and activated by default in our official docker
-image.
-
-This behavior is controlled by configuration `enable_heap_profiling`:
-
-```toml
-[memory]
-# Whether to enable heap profiling activation during startup.
-# Default is true.
-enable_heap_profiling = true
-```
-
-To disable memory profiling, set `enable_heap_profiling` to `false`.
-
 ### Memory profiling control
 
 You can control heap profiling activation using the new HTTP APIs:
@@ -71,15 +70,6 @@ curl -X POST localhost:4000/debug/prof/mem/activate
 
 # Deactivate heap profiling
 curl -X POST localhost:4000/debug/prof/mem/deactivate
-
-# Activate gdump feature that dumps memory profiling data every time virtual memory usage exceeds previous maximum value.
-curl -X POST localhost:4000/debug/prof/mem/gdump -d 'activate=true'
-
-# Deactivate gdump.
-curl -X POST localhost:4000/debug/prof/mem/gdump -d 'activate=false'
-
-# Retrieve current gdump status.
-curl -X GET localhost:4000/debug/prof/mem/gdump
 ```
 
 ### Dump memory profiling data
@@ -92,9 +82,6 @@ curl -X POST localhost:4000/debug/prof/mem > greptime.hprof
 curl -X POST "localhost:4000/debug/prof/mem?output=flamegraph" > greptime.svg
 # or output pprof format
 curl -X POST "localhost:4000/debug/prof/mem?output=proto" > greptime.pprof
-
-
-curl -X POST "localhost:4000/debug/prof/bytes" > greptime.svg
 ```
 
 You can periodically dump profiling data and compare them to find the delta memory usage.

docs/rfcs/2025-09-08-laminar-flow.md

@@ -1,463 +0,0 @@
----
-Feature Name: "laminar-flow"
-Tracking Issue: https://github.com/GreptimeTeam/greptimedb/issues/TBD
-Date: 2025-09-08
-Author: "discord9 <discord9@163.com>"
----
-
-# laminar Flow
-
-## Summary
-
-This RFC proposes a redesign of the flow architecture where flownode becomes a lightweight in-memory state management node with an embedded frontend for direct computation. This approach optimizes resource utilization and improves scalability by eliminating network hops while maintaining clear separation between coordination and computation tasks.
-
-## Motivation
-
-The current flow architecture has several limitations:
-
-1. **Resource Inefficiency**: Flownodes perform both state management and computation, leading to resource duplication and inefficient utilization.
-2. **Scalability Constraints**: Computation resources are tied to flownode instances, limiting horizontal scaling capabilities.
-3. **State Management Complexity**: Mixing computation with state management makes the system harder to maintain and debug.
-4. **Network Overhead**: Additional network hops between flownode and separate frontend nodes add latency.
-
-The laminar Flow architecture addresses these issues by:
-- Consolidating computation within flownode through embedded frontend
-- Eliminating network overhead by removing separate frontend node communication
-- Simplifying state management by focusing flownode on its core responsibility
-- Improving system scalability and maintainability
-
-## Details
-
-### Architecture Overview
-
-The laminar Flow architecture transforms flownode into a lightweight coordinator that maintains flow state with an embedded frontend for computation. The key components involved are:
-
-1. **Flownode**: Maintains in-memory state, coordinates computation, and includes an embedded frontend for query execution
-2. **Embedded Frontend**: Executes **incremental** computations within the flownode
-3. **Datanode**: Stores final results and source data
-
-```mermaid
-graph TB
-    subgraph "laminar Flow Architecture"
-        subgraph Flownode["Flownode (State Manager + Embedded Frontend)"]
-            StateMap["Flow State Map<br/>Map<Timestamp, (Map<Key, Value>, Sequence)>"]
-            Coordinator["Computation Coordinator"]
-            subgraph EmbeddedFrontend["Embedded Frontend"]
-                QueryEngine["Query Engine"]
-                AggrState["__aggr_state Executor"]
-            end
-        end
-        
-        subgraph Datanode["Datanode"]
-            Storage["Data Storage"]
-            Results["Result Tables"]
-        end
-    end
-    
-    Coordinator -->|Internal Query| EmbeddedFrontend
-    EmbeddedFrontend -->|Incremental States| Coordinator
-    Flownode -->|Incremental Results| Datanode
-    EmbeddedFrontend -.->|Read Data| Datanode
-```
-
-### Core Components
-
-#### 1. Flow State Management
-
-Flownode maintains a state map for each flow:
-
-```rust
-type FlowState = Map<Timestamp, (Map<Key, Value>, Sequence)>;
-```
-
-Where:
-- **Timestamp**: Time window identifier for aggregation groups
-- **Key**: Aggregation group expressions (`group_exprs`)
-- **Value**: Aggregation expressions results (`aggr_exprs`)
-- **Sequence**: Computation progress marker for incremental updates
-
-#### 2. Incremental Computation Process
-
-The computation process follows these steps:
-
-1. **Trigger Evaluation**: Flownode determines when to trigger computation based on:
-   - Time intervals (periodic updates)
-   - Data volume thresholds
-   - Sequence progress requirements
-
-2. **Query Execution**: Flownode executes `__aggr_state` queries using its embedded frontend with:
-   - Time window filters
-   - Sequence range constraints
-
-3. **State Update**: Flownode receives partial state results and updates its internal state:
-   - Merges new values with existing aggregation state
-   - Updates sequence markers to track progress
-   - Identifies changed time windows for result computation
-
-4. **Result Materialization**: Flownode computes final results using `__aggr_merge` operations:
-   - Processes only updated time windows(and time series) for efficiency
-   - Writes results back to datanode directly through its embedded frontend
-
-### Detailed Workflow
-
-#### Incremental State Query
-
-```sql
--- Example incremental state query executed by embedded frontend
-SELECT
-    __aggr_state(avg(value)) as state,
-    time_window,
-    group_key
-FROM source_table
-WHERE
-    timestamp >= :window_start
-    AND timestamp < :window_end
-    AND __sequence >= :last_sequence
-    AND __sequence < :current_sequence
-    -- sequence range is actually written in grpc header, but shown here for clarity
-GROUP BY time_window, group_key;
-```
-
-#### State Merge Process
-
-```mermaid
-sequenceDiagram
-    participant F as Flownode (Coordinator)
-    participant EF as Embedded Frontend (Lightweight)
-    participant DN as Datanode (Heavy Computation)
-    
-    F->>F: Evaluate trigger conditions
-    F->>EF: Execute __aggr_state query with sequence range
-    EF->>DN: Send query to datanode (Heavy scan & aggregation)
-    DN->>DN: Scan data and compute partial aggregation state (Heavy CPU/I/O)
-    DN->>EF: Return aggregated state results
-    EF->>F: Forward state results (Lightweight merge)
-    F->>F: Merge with existing state
-    F->>F: Update sequence markers (Lightweight)
-    F->>EF: Compute incremental results with __aggr_merge
-    EF->>DN: Write incremental results to datanode
-```
-
-### Refill Implementation and State Management
-
-#### Refill Process
-
-Refill is implemented as a straightforward `__aggr_state` query with time and sequence constraints:
-
-```sql
--- Refill query for flow state recovery
-SELECT 
-    __aggr_state(aggregation_functions) as state,
-    time_window,
-    group_keys
-FROM source_table 
-WHERE 
-    timestamp >= :refill_start_time 
-    AND timestamp < :refill_end_time
-    AND __sequence >= :start_sequence 
-    AND __sequence < :end_sequence
-    -- sequence range is actually written in grpc header, but shown here for clarity
-GROUP BY time_window, group_keys;
-```
-
-#### State Recovery Strategy
-
-1. **Recent Data (Stream Mode)**: For recent time windows, flownode refills state using incremental queries
-2. **Historical Data (Batch Mode)**: For older time windows, flownode triggers batch computation directly and no need to refill state
-3. **Hybrid Approach**: Combines stream and batch processing based on data age and availability
-
-#### Mirror Write Optimization
-
-Mirror writes are simplified to only transmit timestamps to flownode:
-
-```rust
-struct MirrorWrite {
-    timestamps: Vec<Timestamp>,
-    // Removed: actual data payload
-}
-```
-
-This optimization:
-- Eliminates network overhead by using embedded frontend
-- Enables flownode to track pending time windows efficiently
-- Allows flownode to decide processing mode (stream vs batch) based on timestamp age
-
-Another optimization could be just send dirty time windows range for each flow to flownode directly, no need to send timestamps one by one.
-
-### Query Optimization Strategies
-
-#### Sequence-Based Incremental Processing
-
-The core optimization relies on sequence-constrained queries:
-
-```sql
--- Optimized incremental query
-SELECT __aggr_state(expr) 
-FROM table 
-WHERE time_range AND sequence_range
-```
-
-Benefits:
-- **Reduced Scan Volume**: Only processes data since last computation
-- **Efficient Resource Usage**: Minimizes CPU and I/O overhead
-- **Predictable Performance**: Query cost scales with incremental data size
-
-#### Time Window Partitioning
-
-```mermaid
-graph LR
-    subgraph "Time Windows"
-        W1["Window 1<br/>09:00-09:05"]
-        W2["Window 2<br/>09:05-09:10"]
-        W3["Window 3<br/>09:10-09:15"]
-    end
-    
-    subgraph "Processing Strategy"
-        W1 --> Batch["Batch Mode<br/>(Old Data)"]
-        W2 --> Stream["Stream Mode<br/>(Recent Data)"]
-        W3 --> Stream2["Stream Mode<br/>(Current Data)"]
-    end
-```
-
-### Performance Characteristics
-
-#### Memory Usage
-
-- **Flownode**: O(active_time_windows × group_cardinality) for state storage
-- **Embedded Frontend**: O(query_batch_size) for temporary computation
-- **Overall**: Significantly reduced compared to current architecture
-
-#### Computation Distribution
-
-- **Direct Processing**: Queries processed directly within flownode's embedded frontend
-- **Fault Tolerance**: Simplified error handling with fewer distributed components
-- **Scalability**: Computation capacity scales with flownode instances
-
-#### Network Optimization
-
-- **Reduced Payload**: Mirror writes only contain timestamps
-- **Efficient Queries**: Sequence constraints minimize data transfer
-- **Result Caching**: State results cached in flownode memory
-
-### Sequential Read Implementation for Incremental Queries
-
-#### Sequence Management
-
-Flow maintains two critical sequences to track incremental query progress for each region:
-
-- **`memtable_last_seq`**: Tracks the latest sequence number read from the memtable
-- **`sst_last_seq`**: Tracks the latest sequence number read from SST files
-
-These sequences enable precise incremental data processing by defining the exact range of data to query in subsequent iterations.
-
-#### Query Protocol
-
-When executing incremental queries, flownode provides both sequence parameters to datanode:
-
-```rust
-struct GrpcHeader {
-    ...
-    // Sequence tracking for incremental reads
-    memtable_last_seq: HashMap<RegionId, SequenceNumber>,
-    sst_last_seqs: HashMap<RegionId, SequenceNumber>,
-}
-```
-
-The datanode processes these parameters to return only the data within the specified sequence ranges, ensuring efficient incremental processing.
-
-#### Sequence Invalidation and Refill Mechanism
-
-A critical challenge occurs when data referenced by `memtable_last_seq` gets flushed from memory to disk. Since SST files only maintain a single maximum sequence number for the entire file (rather than per-record sequence tracking), precise incremental queries become impossible for the affected time ranges.
-
-**Detection of Invalidation:**
-```rust
-// When memtable_last_seq data has been flushed to SST
-if memtable_last_seq_flushed_to_disk {
-    // Incremental query is no longer feasible
-    // Need to trigger refill for affected time ranges
-}
-```
-
-**Refill Process:**
-1. **Identify Affected Time Range**: Query the time range corresponding to the flushed `memtable_last_seq` data
-2. **Full Recomputation**: Execute a complete aggregation query for the affected time windows
-3. **State Replacement**: Replace the existing flow state for these time ranges with newly computed values
-4. **Sequence Update**: Update `memtable_last_seq` to the current latest sequence, while `sst_last_seq` continues normal incremental updates
-
-```sql
--- Refill query when memtable data has been flushed
-SELECT
-    __aggr_state(aggregation_functions) as state,
-    time_window,
-    group_keys
-FROM source_table
-WHERE
-    timestamp >= :affected_time_start
-    AND timestamp < :affected_time_end
-    -- Full scan required since sequence precision is lost in SST
-GROUP BY time_window, group_keys;
-```
-
-#### Datanode Implementation Requirements
-
-Datanode must implement enhanced query processing capabilities to support sequence-based incremental reads:
-
-**Input Processing:**
-- Accept `memtable_last_seq` and `sst_last_seq` parameters in query requests
-- Filter data based on sequence ranges across both memtable and SST storage layers
-
-**Output Enhancement:**
-```rust
-struct OutputMeta {
-    pub plan: Option<Arc<dyn ExecutionPlan>>,
-    pub cost: OutputCost,
-    pub sequence_info: HashMap<RegionId, SequenceInfo>, // New field for sequence tracking per regions involved in the query
-}
-
-struct SequenceInfo {
-    // Sequence tracking for next iteration
-    max_memtable_seq: SequenceNumber,  // Highest sequence from memtable in this result
-    max_sst_seq: SequenceNumber,       // Highest sequence from SST in this result
-}
-```
-
-**Sequence Tracking Logic:**
-datanode already impl `max_sst_seq` in leader range read, can reuse similar logic for `max_memtable_seq`.
-
-#### Sequence Update Strategy
-
-**Normal Incremental Updates:**
-- Update both `memtable_last_seq` and `sst_last_seq` after successful query execution
-- Use returned `max_memtable_seq` and `max_sst_seq` values for next iteration
-
-**Refill Scenario:**
-- Reset `memtable_last_seq` to current maximum after refill completion
-- Continue normal `sst_last_seq` updates based on successful query responses
-- Maintain separate tracking to detect future flush events
-
-#### Performance Considerations
-
-**Sequence Range Optimization:**
-- Minimize sequence range spans to reduce scan overhead
-- Batch multiple small incremental updates when beneficial
-- Balance between query frequency and processing efficiency
-
-**Memory Management:**
-- Monitor memtable flush frequency to predict refill requirements
-- Implement adaptive query scheduling based on flush patterns
-- Optimize state storage to handle frequent updates efficiently
-
-This sequential read implementation ensures reliable incremental processing while gracefully handling the complexities of storage architecture, maintaining both correctness and performance in the face of background compaction and flush operations.
-
-## Implementation Plan
-
-### Phase 1: Core Infrastructure
-
-1. **State Management**: Implement in-memory state map in flownode
-2. **Query Interface**: Integrate `__aggr_state` query interface in embedded frontend(Already done in previous query pushdown optimizer work)
-3. **Basic Coordination**: Implement query dispatch and result collection
-4. **Sequence Tracking**: Implement sequence-based incremental processing(Can use similar interface which leader range read use)
-
-After phase 1, the system should support basic flow operations with incremental updates.
-
-### Phase 2: Optimization Features
-
-1. **Refill Logic**: Develop state recovery mechanisms
-2. **Mirror Write Optimization**: Simplify mirror write protocol
-
-### Phase 3: Advanced Features
-
-1. **Load Balancing**: Implement intelligent resource allocation for partitioned flow(Flow distributed executed on multiple flownodes)
-2. **Fault Tolerance**: Add retry mechanisms and error handling
-3. **Performance Tuning**: Optimize query batching and state management
-
-## Drawbacks
-
-### Reduced Network Communication
-
-- **Eliminated Hops**: Direct communication between flownode and datanode through embedded frontend
-- **Reduced Latency**: No separate frontend node communication overhead
-- **Simplified Network Topology**: Fewer network dependencies and failure points
-
-### Complexity in Error Handling
-
-- **Distributed Failures**: Need to handle failures across multiple components
-- **State Consistency**: Ensuring state consistency during partial failures
-- **Recovery Complexity**: More complex recovery procedures
-
-### Datanode Resource Requirements
-
-- **Computation Load**: Datanode handles the heavy computational workload for flow queries
-- **Query Interference**: Flow queries may impact regular query performance on datanode
-- **Resource Contention**: Need careful resource management and isolation on datanode
-
-## Alternatives
-
-### Alternative 1: Enhanced Current Architecture
-
-Keep computation in flownode but optimize through:
-- Better resource management
-- Improved query optimization
-- Enhanced state persistence
-
-**Pros:**
-- Simpler architecture
-- Fewer network hops
-- Easier debugging
-
-**Cons:**
-- Limited scalability
-- Resource inefficiency
-- Harder to optimize computation distribution
-
-### Alternative 2: Embedded Computation
-
-Embed lightweight computation engines within flownode:
-
-**Pros:**
-- Reduced network communication
-- Better performance for simple queries
-- Simpler deployment
-
-**Cons:**
-- Limited scalability
-- Resource constraints
-- Harder to leverage existing frontend optimizations
-
-## Future Work
-
-### Advanced Query Optimization
-
-- **Parallel Processing**: Enable parallel execution of flow queries
-- **Query Caching**: Cache frequently executed query patterns
-
-### Enhanced State Management
-
-- **State Compression**: Implement efficient state serialization
-- **Distributed State**: Support state distribution across multiple flownodes
-- **State Persistence**: Add optional state persistence for durability
-
-### Monitoring and Observability
-
-- **Performance Metrics**: Track query execution times and resource usage
-- **State Visualization**: Provide tools for state inspection and debugging
-- **Health Monitoring**: Monitor system health and performance characteristics
-
-### Integration Improvements
-
-- **Embedded Frontend Optimization**: Optimize embedded frontend query planning and execution
-- **Datanode Optimization**: Optimize result writing from flownode
-- **Metasrv Coordination**: Enhanced metadata management and coordination
-
-## Conclusion
-
-The laminar Flow architecture represents a significant improvement over the current flow system by separating state management from computation execution. This design enables better resource utilization, improved scalability, and simplified maintenance while maintaining the core functionality of continuous aggregation.
-
-The key benefits include:
-
-1. **Improved Scalability**: Computation can scale independently of state management
-2. **Better Resource Utilization**: Eliminates network overhead and leverages embedded frontend infrastructure
-3. **Simplified Architecture**: Clear separation of concerns between components
-4. **Enhanced Performance**: Sequence-based incremental processing reduces computational overhead
-
-While the architecture introduces some complexity in terms of distributed coordination and error handling, the benefits significantly outweigh the drawbacks, making it a compelling evolution of the flow system.

flake.lock

@@ -8,11 +8,11 @@
         "rust-analyzer-src": "rust-analyzer-src"
       },
       "locked": {
-        "lastModified": 1760078406,
+        "lastModified": 1745735608,
-        "narHash": "sha256-JeJK0ZA845PtkCHkfo4KjeI1mYrsr2s3cxBYKhF4BoE=",
+        "narHash": "sha256-L0jzm815XBFfF2wCFmR+M1CF+beIEFj6SxlqVKF59Ec=",
         "owner": "nix-community",
         "repo": "fenix",
-        "rev": "351277c60d104944122ee389cdf581c5ce2c6732",
+        "rev": "c39a78eba6ed2a022cc3218db90d485077101496",
         "type": "github"
       },
       "original": {
@@ -41,11 +41,11 @@
     },
     "nixpkgs": {
       "locked": {
-        "lastModified": 1759994382,
+        "lastModified": 1748162331,
-        "narHash": "sha256-wSK+3UkalDZRVHGCRikZ//CyZUJWDJkBDTQX1+G77Ow=",
+        "narHash": "sha256-rqc2RKYTxP3tbjA+PB3VMRQNnjesrT0pEofXQTrMsS8=",
         "owner": "NixOS",
         "repo": "nixpkgs",
-        "rev": "5da4a26309e796daa7ffca72df93dbe53b8164c7",
+        "rev": "7c43f080a7f28b2774f3b3f43234ca11661bf334",
         "type": "github"
       },
       "original": {
@@ -65,11 +65,11 @@
     "rust-analyzer-src": {
       "flake": false,
       "locked": {
-        "lastModified": 1760014945,
+        "lastModified": 1745694049,
-        "narHash": "sha256-ySdl7F9+oeWNHVrg3QL/brazqmJvYFEdpGnF3pyoDH8=",
+        "narHash": "sha256-fxvRYH/tS7hGQeg9zCVh5RBcSWT+JGJet7RA8Ss+rC0=",
         "owner": "rust-lang",
         "repo": "rust-analyzer",
-        "rev": "90d2e1ce4dfe7dc49250a8b88a0f08ffdb9cb23f",
+        "rev": "d8887c0758bbd2d5f752d5bd405d4491e90e7ed6",
         "type": "github"
       },
       "original": {

flake.nix

@@ -19,7 +19,7 @@
         lib = nixpkgs.lib;
         rustToolchain = fenix.packages.${system}.fromToolchainName {
           name = (lib.importTOML ./rust-toolchain.toml).toolchain.channel;
-          sha256 = "sha256-GCGEXGZeJySLND0KU5TdtTrqFV76TF3UdvAHSUegSsk=";
+          sha256 = "sha256-tJJr8oqX3YD+ohhPK7jlt/7kvKBnBqJVjYtoFr520d4=";
         };
       in
       {

grafana/dashboards/metrics/cluster/dashboard.json

@@ -1411,7 +1411,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{instance=~\"$datanode\"})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-datanode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1528,7 +1528,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{instance=~\"$datanode\"})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-datanode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1643,7 +1643,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{instance=~\"$frontend\"})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-frontend\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1760,7 +1760,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{instance=~\"$frontend\"})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-frontend\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1875,7 +1875,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{instance=~\"$metasrv\"})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-metasrv\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1992,7 +1992,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{instance=~\"$metasrv\"})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-metasrv\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -2107,7 +2107,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{instance=~\"$flownode\"})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-flownode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -2224,7 +2224,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{instance=~\"$flownode\"})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-flownode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",

grafana/dashboards/metrics/cluster/dashboard.md

@@ -21,14 +21,14 @@
 # Resources
 | Title | Query | Type | Description | Datasource | Unit | Legend Format |
 | --- | --- | --- | --- | --- | --- | --- |
-| Datanode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$datanode"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{instance=~"$datanode"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{instance}}]-[{{ pod }}]` |
+| Datanode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$datanode"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-datanode"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{instance}}]-[{{ pod }}]` |
-| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$datanode"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{instance=~"$datanode"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$datanode"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-datanode"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
-| Frontend Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$frontend"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{instance=~"$frontend"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Frontend Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$frontend"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-frontend"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
-| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$frontend"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{instance=~"$frontend"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
+| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$frontend"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-frontend"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
-| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$metasrv"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{instance=~"$metasrv"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
+| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$metasrv"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-metasrv"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
-| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$metasrv"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{instance=~"$metasrv"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$metasrv"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-metasrv"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
-| Flownode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$flownode"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{instance=~"$flownode"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode Memory per Instance | `sum(process_resident_memory_bytes{instance=~"$flownode"}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-flownode"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
-| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$flownode"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{instance=~"$flownode"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{instance=~"$flownode"}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-flownode"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
 # Frontend Requests
 | Title | Query | Type | Description | Datasource | Unit | Legend Format |
 | --- | --- | --- | --- | --- | --- | --- |

grafana/dashboards/metrics/cluster/dashboard.yaml

@@ -187,7 +187,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{instance}}]-[{{ pod }}]'
-            - expr: max(greptime_memory_limit_in_bytes{instance=~"$datanode"})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-datanode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -202,7 +202,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_cpu_limit_in_millicores{instance=~"$datanode"})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-datanode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -217,7 +217,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_memory_limit_in_bytes{instance=~"$frontend"})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-frontend"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -232,7 +232,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]-cpu'
-            - expr: max(greptime_cpu_limit_in_millicores{instance=~"$frontend"})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-frontend"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -247,7 +247,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]-resident'
-            - expr: max(greptime_memory_limit_in_bytes{instance=~"$metasrv"})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-metasrv"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -262,7 +262,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_cpu_limit_in_millicores{instance=~"$metasrv"})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-metasrv"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -277,7 +277,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_memory_limit_in_bytes{instance=~"$flownode"})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-flownode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -292,7 +292,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_cpu_limit_in_millicores{instance=~"$flownode"})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-flownode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}

grafana/dashboards/metrics/standalone/dashboard.json

@@ -1411,7 +1411,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-datanode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1528,7 +1528,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-datanode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1643,7 +1643,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-frontend\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1760,7 +1760,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-frontend\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1875,7 +1875,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-metasrv\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -1992,7 +1992,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-metasrv\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -2107,7 +2107,7 @@
                 "uid": "${metrics}"
               },
               "editorMode": "code",
-              "expr": "max(greptime_memory_limit_in_bytes{})",
+              "expr": "max(greptime_memory_limit_in_bytes{app=\"greptime-flownode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",
@@ -2224,7 +2224,7 @@
               },
               "editorMode": "code",
               "exemplar": false,
-              "expr": "max(greptime_cpu_limit_in_millicores{})",
+              "expr": "max(greptime_cpu_limit_in_millicores{app=\"greptime-flownode\"})",
               "hide": false,
               "instant": false,
               "legendFormat": "limit",

grafana/dashboards/metrics/standalone/dashboard.md

@@ -21,14 +21,14 @@
 # Resources
 | Title | Query | Type | Description | Datasource | Unit | Legend Format |
 | --- | --- | --- | --- | --- | --- | --- |
-| Datanode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{instance}}]-[{{ pod }}]` |
+| Datanode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-datanode"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{instance}}]-[{{ pod }}]` |
-| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Datanode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-datanode"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
-| Frontend Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Frontend Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-frontend"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
-| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
+| Frontend CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-frontend"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]-cpu` |
-| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
+| Metasrv Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-metasrv"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]-resident` |
-| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Metasrv CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-metasrv"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
-| Flownode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode Memory per Instance | `sum(process_resident_memory_bytes{}) by (instance, pod)`<br/>`max(greptime_memory_limit_in_bytes{app="greptime-flownode"})` | `timeseries` | Current memory usage by instance | `prometheus` | `bytes` | `[{{ instance }}]-[{{ pod }}]` |
-| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
+| Flownode CPU Usage per Instance | `sum(rate(process_cpu_seconds_total{}[$__rate_interval]) * 1000) by (instance, pod)`<br/>`max(greptime_cpu_limit_in_millicores{app="greptime-flownode"})` | `timeseries` | Current cpu usage by instance | `prometheus` | `none` | `[{{ instance }}]-[{{ pod }}]` |
 # Frontend Requests
 | Title | Query | Type | Description | Datasource | Unit | Legend Format |
 | --- | --- | --- | --- | --- | --- | --- |

grafana/dashboards/metrics/standalone/dashboard.yaml

@@ -187,7 +187,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{instance}}]-[{{ pod }}]'
-            - expr: max(greptime_memory_limit_in_bytes{})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-datanode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -202,7 +202,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_cpu_limit_in_millicores{})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-datanode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -217,7 +217,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_memory_limit_in_bytes{})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-frontend"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -232,7 +232,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]-cpu'
-            - expr: max(greptime_cpu_limit_in_millicores{})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-frontend"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -247,7 +247,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]-resident'
-            - expr: max(greptime_memory_limit_in_bytes{})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-metasrv"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -262,7 +262,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_cpu_limit_in_millicores{})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-metasrv"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -277,7 +277,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_memory_limit_in_bytes{})
+            - expr: max(greptime_memory_limit_in_bytes{app="greptime-flownode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}
@@ -292,7 +292,7 @@ groups:
                 type: prometheus
                 uid: ${metrics}
               legendFormat: '[{{ instance }}]-[{{ pod }}]'
-            - expr: max(greptime_cpu_limit_in_millicores{})
+            - expr: max(greptime_cpu_limit_in_millicores{app="greptime-flownode"})
               datasource:
                 type: prometheus
                 uid: ${metrics}

rust-toolchain.toml

@@ -1,2 +1,2 @@
 [toolchain]
-channel = "nightly-2025-10-01"
+channel = "nightly-2025-05-19"

scripts/generate_certs.sh

@@ -1,41 +0,0 @@
-#!/usr/bin/env bash
-set -euo pipefail
-
-CERT_DIR="${1:-$(dirname "$0")/../tests-integration/fixtures/certs}"
-DAYS="${2:-365}"
-
-mkdir -p "${CERT_DIR}"
-cd "${CERT_DIR}"
-
-echo "Generating CA certificate..."
-openssl req -new -x509 -days "${DAYS}" -nodes -text \
-  -out root.crt -keyout root.key \
-  -subj "/CN=GreptimeDBRootCA"
-
-
-echo "Generating server certificate..."
-openssl req -new -nodes -text \
-  -out server.csr -keyout server.key \
-  -subj "/CN=greptime"
-
-openssl x509 -req -in server.csr -text -days "${DAYS}" \
-  -CA root.crt -CAkey root.key -CAcreateserial \
-  -out server.crt \
-  -extensions v3_req -extfile <(printf "[v3_req]\nsubjectAltName=DNS:localhost,IP:127.0.0.1")
-
-echo "Generating client certificate..."
-# Make sure the client certificate is for the greptimedb user
-openssl req -new -nodes -text \
-  -out client.csr -keyout client.key \
-  -subj "/CN=greptimedb"
-
-openssl x509 -req -in client.csr -CA root.crt -CAkey root.key -CAcreateserial \
--out client.crt -days 365 -extensions v3_req -extfile <(printf "[v3_req]\nsubjectAltName=DNS:localhost")
-
-rm -f *.csr
-
-echo "TLS certificates generated successfully in ${CERT_DIR}"
-
-chmod 644 root.key
-chmod 644 client.key
-chmod 644 server.key

src/api/src/lib.rs

@@ -12,6 +12,8 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+#![feature(let_chains)]
+
 pub mod error;
 pub mod helper;
 

src/api/src/v1/column_def.rs

@@ -16,8 +16,8 @@ use std::collections::HashMap;
 
 use datatypes::schema::{
     COMMENT_KEY, ColumnDefaultConstraint, ColumnSchema, FULLTEXT_KEY, FulltextAnalyzer,
-    FulltextBackend, FulltextOptions, INVERTED_INDEX_KEY, JSON_STRUCTURE_SETTINGS_KEY,
+    FulltextBackend, FulltextOptions, INVERTED_INDEX_KEY, SKIPPING_INDEX_KEY, SkippingIndexOptions,
-    SKIPPING_INDEX_KEY, SkippingIndexOptions, SkippingIndexType,
+    SkippingIndexType,
 };
 use greptime_proto::v1::{
     Analyzer, FulltextBackend as PbFulltextBackend, SkippingIndexType as PbSkippingIndexType,
@@ -37,10 +37,8 @@ const SKIPPING_INDEX_GRPC_KEY: &str = "skipping_index";
 
 /// Tries to construct a `ColumnSchema` from the given  `ColumnDef`.
 pub fn try_as_column_schema(column_def: &ColumnDef) -> Result<ColumnSchema> {
-    let data_type = ColumnDataTypeWrapper::try_new(
+    let data_type =
-        column_def.data_type,
+        ColumnDataTypeWrapper::try_new(column_def.data_type, column_def.datatype_extension)?;
-        column_def.datatype_extension.clone(),
-    )?;
 
     let constraint = if column_def.default_constraint.is_empty() {
         None
@@ -68,9 +66,6 @@ pub fn try_as_column_schema(column_def: &ColumnDef) -> Result<ColumnSchema> {
         if let Some(skipping_index) = options.options.get(SKIPPING_INDEX_GRPC_KEY) {
             metadata.insert(SKIPPING_INDEX_KEY.to_string(), skipping_index.to_owned());
         }
-        if let Some(settings) = options.options.get(JSON_STRUCTURE_SETTINGS_KEY) {
-            metadata.insert(JSON_STRUCTURE_SETTINGS_KEY.to_string(), settings.clone());
-        }
     }
 
     ColumnSchema::new(&column_def.name, data_type.into(), column_def.is_nullable)
@@ -142,11 +137,6 @@ pub fn options_from_column_schema(column_schema: &ColumnSchema) -> Option<Column
             .options
             .insert(SKIPPING_INDEX_GRPC_KEY.to_string(), skipping_index.clone());
     }
-    if let Some(settings) = column_schema.metadata().get(JSON_STRUCTURE_SETTINGS_KEY) {
-        options
-            .options
-            .insert(JSON_STRUCTURE_SETTINGS_KEY.to_string(), settings.clone());
-    }
 
     (!options.options.is_empty()).then_some(options)
 }

src/auth/src/common.rs

@@ -35,7 +35,7 @@ pub fn userinfo_by_name(username: Option<String>) -> UserInfoRef {
     DefaultUserInfo::with_name(username.unwrap_or_else(|| DEFAULT_USERNAME.to_string()))
 }
 
-pub fn user_provider_from_option(opt: &str) -> Result<UserProviderRef> {
+pub fn user_provider_from_option(opt: &String) -> Result<UserProviderRef> {
     let (name, content) = opt.split_once(':').with_context(|| InvalidConfigSnafu {
         value: opt.to_string(),
         msg: "UserProviderOption must be in format `<option>:<value>`",
@@ -57,7 +57,7 @@ pub fn user_provider_from_option(opt: &str) -> Result<UserProviderRef> {
     }
 }
 
-pub fn static_user_provider_from_option(opt: &str) -> Result<StaticUserProvider> {
+pub fn static_user_provider_from_option(opt: &String) -> Result<StaticUserProvider> {
     let (name, content) = opt.split_once(':').with_context(|| InvalidConfigSnafu {
         value: opt.to_string(),
         msg: "UserProviderOption must be in format `<option>:<value>`",

src/auth/src/lib.rs

@@ -25,7 +25,7 @@ pub use common::{
     HashedPassword, Identity, Password, auth_mysql, static_user_provider_from_option,
     user_provider_from_option, userinfo_by_name,
 };
-pub use permission::{DefaultPermissionChecker, PermissionChecker, PermissionReq, PermissionResp};
+pub use permission::{PermissionChecker, PermissionReq, PermissionResp};
 pub use user_info::UserInfo;
 pub use user_provider::UserProvider;
 pub use user_provider::static_user_provider::StaticUserProvider;

src/auth/src/permission.rs

@@ -13,15 +13,12 @@
 // limitations under the License.
 
 use std::fmt::Debug;
-use std::sync::Arc;
 
 use api::v1::greptime_request::Request;
-use common_telemetry::debug;
 use sql::statements::statement::Statement;
 
 use crate::error::{PermissionDeniedSnafu, Result};
-use crate::user_info::DefaultUserInfo;
+use crate::{PermissionCheckerRef, UserInfoRef};
-use crate::{PermissionCheckerRef, UserInfo, UserInfoRef};
 
 #[derive(Debug, Clone)]
 pub enum PermissionReq<'a> {
@@ -38,32 +35,6 @@ pub enum PermissionReq<'a> {
     BulkInsert,
 }
 
-impl<'a> PermissionReq<'a> {
-    /// Returns true if the permission request is for read operations.
-    pub fn is_readonly(&self) -> bool {
-        match self {
-            PermissionReq::GrpcRequest(Request::Query(_))
-            | PermissionReq::PromQuery
-            | PermissionReq::LogQuery
-            | PermissionReq::PromStoreRead => true,
-            PermissionReq::SqlStatement(stmt) => stmt.is_readonly(),
-
-            PermissionReq::GrpcRequest(_)
-            | PermissionReq::Opentsdb
-            | PermissionReq::LineProtocol
-            | PermissionReq::PromStoreWrite
-            | PermissionReq::Otlp
-            | PermissionReq::LogWrite
-            | PermissionReq::BulkInsert => false,
-        }
-    }
-
-    /// Returns true if the permission request is for write operations.
-    pub fn is_write(&self) -> bool {
-        !self.is_readonly()
-    }
-}
-
 #[derive(Debug)]
 pub enum PermissionResp {
     Allow,
@@ -94,106 +65,3 @@ impl PermissionChecker for Option<&PermissionCheckerRef> {
         }
     }
 }
-
-/// The default permission checker implementation.
-/// It checks the permission mode of [DefaultUserInfo].
-pub struct DefaultPermissionChecker;
-
-impl DefaultPermissionChecker {
-    /// Returns a new [PermissionCheckerRef] instance.
-    pub fn arc() -> PermissionCheckerRef {
-        Arc::new(DefaultPermissionChecker)
-    }
-}
-
-impl PermissionChecker for DefaultPermissionChecker {
-    fn check_permission(
-        &self,
-        user_info: UserInfoRef,
-        req: PermissionReq,
-    ) -> Result<PermissionResp> {
-        if let Some(default_user) = user_info.as_any().downcast_ref::<DefaultUserInfo>() {
-            let permission_mode = default_user.permission_mode();
-
-            if req.is_readonly() && !permission_mode.can_read() {
-                debug!(
-                    "Permission denied: read operation not allowed, user = {}, permission = {}",
-                    default_user.username(),
-                    permission_mode.as_str()
-                );
-                return Ok(PermissionResp::Reject);
-            }
-
-            if req.is_write() && !permission_mode.can_write() {
-                debug!(
-                    "Permission denied: write operation not allowed, user = {}, permission = {}",
-                    default_user.username(),
-                    permission_mode.as_str()
-                );
-                return Ok(PermissionResp::Reject);
-            }
-        }
-
-        // default allow all
-        Ok(PermissionResp::Allow)
-    }
-}
-#[cfg(test)]
-mod tests {
-    use super::*;
-    use crate::user_info::PermissionMode;
-
-    #[test]
-    fn test_default_permission_checker_allow_all_operations() {
-        let checker = DefaultPermissionChecker;
-        let user_info =
-            DefaultUserInfo::with_name_and_permission("test_user", PermissionMode::ReadWrite);
-
-        let read_req = PermissionReq::PromQuery;
-        let write_req = PermissionReq::PromStoreWrite;
-
-        let read_result = checker
-            .check_permission(user_info.clone(), read_req)
-            .unwrap();
-        let write_result = checker.check_permission(user_info, write_req).unwrap();
-
-        assert!(matches!(read_result, PermissionResp::Allow));
-        assert!(matches!(write_result, PermissionResp::Allow));
-    }
-
-    #[test]
-    fn test_default_permission_checker_readonly_user() {
-        let checker = DefaultPermissionChecker;
-        let user_info =
-            DefaultUserInfo::with_name_and_permission("readonly_user", PermissionMode::ReadOnly);
-
-        let read_req = PermissionReq::PromQuery;
-        let write_req = PermissionReq::PromStoreWrite;
-
-        let read_result = checker
-            .check_permission(user_info.clone(), read_req)
-            .unwrap();
-        let write_result = checker.check_permission(user_info, write_req).unwrap();
-
-        assert!(matches!(read_result, PermissionResp::Allow));
-        assert!(matches!(write_result, PermissionResp::Reject));
-    }
-
-    #[test]
-    fn test_default_permission_checker_writeonly_user() {
-        let checker = DefaultPermissionChecker;
-        let user_info =
-            DefaultUserInfo::with_name_and_permission("writeonly_user", PermissionMode::WriteOnly);
-
-        let read_req = PermissionReq::LogQuery;
-        let write_req = PermissionReq::LogWrite;
-
-        let read_result = checker
-            .check_permission(user_info.clone(), read_req)
-            .unwrap();
-        let write_result = checker.check_permission(user_info, write_req).unwrap();
-
-        assert!(matches!(read_result, PermissionResp::Reject));
-        assert!(matches!(write_result, PermissionResp::Allow));
-    }
-}

src/auth/src/user_info.rs

@@ -23,86 +23,17 @@ pub trait UserInfo: Debug + Sync + Send {
     fn username(&self) -> &str;
 }
 
-/// The user permission mode
-#[derive(Debug, Clone, Copy, PartialEq, Eq, Default)]
-pub enum PermissionMode {
-    #[default]
-    ReadWrite,
-    ReadOnly,
-    WriteOnly,
-}
-
-impl PermissionMode {
-    /// Parse permission mode from string.
-    /// Supported values are:
-    /// - "rw", "readwrite", "read_write" => ReadWrite
-    /// - "ro", "readonly", "read_only" => ReadOnly
-    /// - "wo", "writeonly", "write_only" => WriteOnly
-    ///     Returns None if the input string is not a valid permission mode.
-    pub fn from_str(s: &str) -> Self {
-        match s.to_lowercase().as_str() {
-            "readwrite" | "read_write" | "rw" => PermissionMode::ReadWrite,
-            "readonly" | "read_only" | "ro" => PermissionMode::ReadOnly,
-            "writeonly" | "write_only" | "wo" => PermissionMode::WriteOnly,
-            _ => PermissionMode::ReadWrite,
-        }
-    }
-
-    /// Convert permission mode to string.
-    /// - ReadWrite => "rw"
-    /// - ReadOnly => "ro"
-    /// - WriteOnly => "wo"
-    ///     The returned string is a static string slice.
-    pub fn as_str(&self) -> &'static str {
-        match self {
-            PermissionMode::ReadWrite => "rw",
-            PermissionMode::ReadOnly => "ro",
-            PermissionMode::WriteOnly => "wo",
-        }
-    }
-
-    /// Returns true if the permission mode allows read operations.
-    pub fn can_read(&self) -> bool {
-        matches!(self, PermissionMode::ReadWrite | PermissionMode::ReadOnly)
-    }
-
-    /// Returns true if the permission mode allows write operations.
-    pub fn can_write(&self) -> bool {
-        matches!(self, PermissionMode::ReadWrite | PermissionMode::WriteOnly)
-    }
-}
-
-impl std::fmt::Display for PermissionMode {
-    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
-        write!(f, "{}", self.as_str())
-    }
-}
-
 #[derive(Debug)]
 pub(crate) struct DefaultUserInfo {
     username: String,
-    permission_mode: PermissionMode,
 }
 
 impl DefaultUserInfo {
     pub(crate) fn with_name(username: impl Into<String>) -> UserInfoRef {
-        Self::with_name_and_permission(username, PermissionMode::default())
-    }
-
-    /// Create a UserInfo with specified permission mode.
-    pub(crate) fn with_name_and_permission(
-        username: impl Into<String>,
-        permission_mode: PermissionMode,
-    ) -> UserInfoRef {
         Arc::new(Self {
             username: username.into(),
-            permission_mode,
         })
     }
-
-    pub(crate) fn permission_mode(&self) -> &PermissionMode {
-        &self.permission_mode
-    }
 }
 
 impl UserInfo for DefaultUserInfo {
@@ -114,120 +45,3 @@ impl UserInfo for DefaultUserInfo {
         self.username.as_str()
     }
 }
-#[cfg(test)]
-mod tests {
-    use super::*;
-
-    #[test]
-    fn test_permission_mode_from_str() {
-        // Test ReadWrite variants
-        assert_eq!(
-            PermissionMode::from_str("readwrite"),
-            PermissionMode::ReadWrite
-        );
-        assert_eq!(
-            PermissionMode::from_str("read_write"),
-            PermissionMode::ReadWrite
-        );
-        assert_eq!(PermissionMode::from_str("rw"), PermissionMode::ReadWrite);
-        assert_eq!(
-            PermissionMode::from_str("ReadWrite"),
-            PermissionMode::ReadWrite
-        );
-        assert_eq!(PermissionMode::from_str("RW"), PermissionMode::ReadWrite);
-
-        // Test ReadOnly variants
-        assert_eq!(
-            PermissionMode::from_str("readonly"),
-            PermissionMode::ReadOnly
-        );
-        assert_eq!(
-            PermissionMode::from_str("read_only"),
-            PermissionMode::ReadOnly
-        );
-        assert_eq!(PermissionMode::from_str("ro"), PermissionMode::ReadOnly);
-        assert_eq!(
-            PermissionMode::from_str("ReadOnly"),
-            PermissionMode::ReadOnly
-        );
-        assert_eq!(PermissionMode::from_str("RO"), PermissionMode::ReadOnly);
-
-        // Test WriteOnly variants
-        assert_eq!(
-            PermissionMode::from_str("writeonly"),
-            PermissionMode::WriteOnly
-        );
-        assert_eq!(
-            PermissionMode::from_str("write_only"),
-            PermissionMode::WriteOnly
-        );
-        assert_eq!(PermissionMode::from_str("wo"), PermissionMode::WriteOnly);
-        assert_eq!(
-            PermissionMode::from_str("WriteOnly"),
-            PermissionMode::WriteOnly
-        );
-        assert_eq!(PermissionMode::from_str("WO"), PermissionMode::WriteOnly);
-
-        // Test invalid inputs default to ReadWrite
-        assert_eq!(
-            PermissionMode::from_str("invalid"),
-            PermissionMode::ReadWrite
-        );
-        assert_eq!(PermissionMode::from_str(""), PermissionMode::ReadWrite);
-        assert_eq!(PermissionMode::from_str("xyz"), PermissionMode::ReadWrite);
-    }
-
-    #[test]
-    fn test_permission_mode_as_str() {
-        assert_eq!(PermissionMode::ReadWrite.as_str(), "rw");
-        assert_eq!(PermissionMode::ReadOnly.as_str(), "ro");
-        assert_eq!(PermissionMode::WriteOnly.as_str(), "wo");
-    }
-
-    #[test]
-    fn test_permission_mode_default() {
-        assert_eq!(PermissionMode::default(), PermissionMode::ReadWrite);
-    }
-
-    #[test]
-    fn test_permission_mode_round_trip() {
-        let modes = [
-            PermissionMode::ReadWrite,
-            PermissionMode::ReadOnly,
-            PermissionMode::WriteOnly,
-        ];
-
-        for mode in modes {
-            let str_repr = mode.as_str();
-            let parsed = PermissionMode::from_str(str_repr);
-            assert_eq!(mode, parsed);
-        }
-    }
-
-    #[test]
-    fn test_default_user_info_with_name() {
-        let user_info = DefaultUserInfo::with_name("test_user");
-        assert_eq!(user_info.username(), "test_user");
-    }
-
-    #[test]
-    fn test_default_user_info_with_name_and_permission() {
-        let user_info =
-            DefaultUserInfo::with_name_and_permission("test_user", PermissionMode::ReadOnly);
-        assert_eq!(user_info.username(), "test_user");
-
-        // Cast to DefaultUserInfo to access permission_mode
-        let default_user = user_info
-            .as_any()
-            .downcast_ref::<DefaultUserInfo>()
-            .unwrap();
-        assert_eq!(default_user.permission_mode, PermissionMode::ReadOnly);
-    }
-
-    #[test]
-    fn test_user_info_as_any() {
-        let user_info = DefaultUserInfo::with_name("test_user");
-        let any_ref = user_info.as_any();
-        assert!(any_ref.downcast_ref::<DefaultUserInfo>().is_some());
-    }
-}

src/auth/src/user_provider.rs

@@ -29,7 +29,7 @@ use crate::error::{
     IllegalParamSnafu, InvalidConfigSnafu, IoSnafu, Result, UnsupportedPasswordTypeSnafu,
     UserNotFoundSnafu, UserPasswordMismatchSnafu,
 };
-use crate::user_info::{DefaultUserInfo, PermissionMode};
+use crate::user_info::DefaultUserInfo;
 use crate::{UserInfoRef, auth_mysql};
 
 #[async_trait::async_trait]
@@ -64,19 +64,11 @@ pub trait UserProvider: Send + Sync {
     }
 }
 
-/// Type alias for user info map
+fn load_credential_from_file(filepath: &str) -> Result<Option<HashMap<String, Vec<u8>>>> {
-/// Key is username, value is (password, permission_mode)
-pub type UserInfoMap = HashMap<String, (Vec<u8>, PermissionMode)>;
-
-fn load_credential_from_file(filepath: &str) -> Result<UserInfoMap> {
     // check valid path
     let path = Path::new(filepath);
     if !path.exists() {
-        return InvalidConfigSnafu {
+        return Ok(None);
-            value: filepath.to_string(),
-            msg: "UserProvider file must exist",
-        }
-        .fail();
     }
 
     ensure!(
@@ -91,19 +83,13 @@ fn load_credential_from_file(filepath: &str) -> Result<UserInfoMap> {
         .lines()
         .map_while(std::result::Result::ok)
         .filter_map(|line| {
-            // The line format is:
+            if let Some((k, v)) = line.split_once('=') {
-            // - `username=password` - Basic user with default permissions
+                Some((k.to_string(), v.as_bytes().to_vec()))
-            // - `username:permission_mode=password` - User with specific permission mode
+            } else {
-            // - Lines starting with '#' are treated as comments and ignored
+                None
-            // - Empty lines are ignored
-            let line = line.trim();
-            if line.is_empty() || line.starts_with('#') {
-                return None;
             }
-
-            parse_credential_line(line)
         })
-        .collect::<HashMap<String, _>>();
+        .collect::<HashMap<String, Vec<u8>>>();
 
     ensure!(
         !credential.is_empty(),
@@ -113,31 +99,11 @@ fn load_credential_from_file(filepath: &str) -> Result<UserInfoMap> {
         }
     );
 
-    Ok(credential)
+    Ok(Some(credential))
-}
-
-/// Parse a line of credential in the format of `username=password` or `username:permission_mode=password`
-pub(crate) fn parse_credential_line(line: &str) -> Option<(String, (Vec<u8>, PermissionMode))> {
-    let parts = line.split('=').collect::<Vec<&str>>();
-    if parts.len() != 2 {
-        return None;
-    }
-
-    let (username_part, password) = (parts[0], parts[1]);
-    let (username, permission_mode) = if let Some((user, perm)) = username_part.split_once(':') {
-        (user, PermissionMode::from_str(perm))
-    } else {
-        (username_part, PermissionMode::default())
-    };
-
-    Some((
-        username.to_string(),
-        (password.as_bytes().to_vec(), permission_mode),
-    ))
 }
 
 fn authenticate_with_credential(
-    users: &UserInfoMap,
+    users: &HashMap<String, Vec<u8>>,
     input_id: Identity<'_>,
     input_pwd: Password<'_>,
 ) -> Result<UserInfoRef> {
@@ -149,7 +115,7 @@ fn authenticate_with_credential(
                     msg: "blank username"
                 }
             );
-            let (save_pwd, permission_mode) = users.get(username).context(UserNotFoundSnafu {
+            let save_pwd = users.get(username).context(UserNotFoundSnafu {
                 username: username.to_string(),
             })?;
 
@@ -162,10 +128,7 @@ fn authenticate_with_credential(
                         }
                     );
                     if save_pwd == pwd.expose_secret().as_bytes() {
-                        Ok(DefaultUserInfo::with_name_and_permission(
+                        Ok(DefaultUserInfo::with_name(username))
-                            username,
-                            *permission_mode,
-                        ))
                     } else {
                         UserPasswordMismatchSnafu {
                             username: username.to_string(),
@@ -174,9 +137,8 @@ fn authenticate_with_credential(
                     }
                 }
                 Password::MysqlNativePassword(auth_data, salt) => {
-                    auth_mysql(auth_data, salt, username, save_pwd).map(|_| {
+                    auth_mysql(auth_data, salt, username, save_pwd)
-                        DefaultUserInfo::with_name_and_permission(username, *permission_mode)
+                        .map(|_| DefaultUserInfo::with_name(username))
-                    })
                 }
                 Password::PgMD5(_, _) => UnsupportedPasswordTypeSnafu {
                     password_type: "pg_md5",
@@ -186,108 +148,3 @@ fn authenticate_with_credential(
         }
     }
 }
-#[cfg(test)]
-mod tests {
-    use super::*;
-
-    #[test]
-    fn test_parse_credential_line() {
-        // Basic username=password format
-        let result = parse_credential_line("admin=password123");
-        assert_eq!(
-            result,
-            Some((
-                "admin".to_string(),
-                ("password123".as_bytes().to_vec(), PermissionMode::default())
-            ))
-        );
-
-        // Username with permission mode
-        let result = parse_credential_line("user:ReadOnly=secret");
-        assert_eq!(
-            result,
-            Some((
-                "user".to_string(),
-                ("secret".as_bytes().to_vec(), PermissionMode::ReadOnly)
-            ))
-        );
-        let result = parse_credential_line("user:ro=secret");
-        assert_eq!(
-            result,
-            Some((
-                "user".to_string(),
-                ("secret".as_bytes().to_vec(), PermissionMode::ReadOnly)
-            ))
-        );
-        // Username with WriteOnly permission mode
-        let result = parse_credential_line("writer:WriteOnly=mypass");
-        assert_eq!(
-            result,
-            Some((
-                "writer".to_string(),
-                ("mypass".as_bytes().to_vec(), PermissionMode::WriteOnly)
-            ))
-        );
-
-        // Username with 'wo' as WriteOnly permission shorthand
-        let result = parse_credential_line("writer:wo=mypass");
-        assert_eq!(
-            result,
-            Some((
-                "writer".to_string(),
-                ("mypass".as_bytes().to_vec(), PermissionMode::WriteOnly)
-            ))
-        );
-
-        // Username with complex password containing special characters
-        let result = parse_credential_line("admin:rw=p@ssw0rd!123");
-        assert_eq!(
-            result,
-            Some((
-                "admin".to_string(),
-                (
-                    "p@ssw0rd!123".as_bytes().to_vec(),
-                    PermissionMode::ReadWrite
-                )
-            ))
-        );
-
-        // Username with spaces should be preserved
-        let result = parse_credential_line("user name:WriteOnly=password");
-        assert_eq!(
-            result,
-            Some((
-                "user name".to_string(),
-                ("password".as_bytes().to_vec(), PermissionMode::WriteOnly)
-            ))
-        );
-
-        // Invalid format - no equals sign
-        let result = parse_credential_line("invalid_line");
-        assert_eq!(result, None);
-
-        // Invalid format - multiple equals signs
-        let result = parse_credential_line("user=pass=word");
-        assert_eq!(result, None);
-
-        // Empty password
-        let result = parse_credential_line("user=");
-        assert_eq!(
-            result,
-            Some((
-                "user".to_string(),
-                ("".as_bytes().to_vec(), PermissionMode::default())
-            ))
-        );
-
-        // Empty username
-        let result = parse_credential_line("=password");
-        assert_eq!(
-            result,
-            Some((
-                "".to_string(),
-                ("password".as_bytes().to_vec(), PermissionMode::default())
-            ))
-        );
-    }
-}

src/auth/src/user_provider/static_user_provider.rs

@@ -12,19 +12,19 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use std::collections::HashMap;
+
 use async_trait::async_trait;
 use snafu::{OptionExt, ResultExt};
 
 use crate::error::{FromUtf8Snafu, InvalidConfigSnafu, Result};
-use crate::user_provider::{
+use crate::user_provider::{authenticate_with_credential, load_credential_from_file};
-    UserInfoMap, authenticate_with_credential, load_credential_from_file, parse_credential_line,
-};
 use crate::{Identity, Password, UserInfoRef, UserProvider};
 
 pub(crate) const STATIC_USER_PROVIDER: &str = "static_user_provider";
 
 pub struct StaticUserProvider {
-    users: UserInfoMap,
+    users: HashMap<String, Vec<u8>>,
 }
 
 impl StaticUserProvider {
@@ -35,18 +35,23 @@ impl StaticUserProvider {
         })?;
         match mode {
             "file" => {
-                let users = load_credential_from_file(content)?;
+                let users = load_credential_from_file(content)?
+                    .context(InvalidConfigSnafu {
+                        value: content.to_string(),
+                        msg: "StaticFileUserProvider must be a valid file path",
+                    })?;
                 Ok(StaticUserProvider { users })
             }
             "cmd" => content
                 .split(',')
                 .map(|kv| {
-                   parse_credential_line(kv).context(InvalidConfigSnafu {
+                    let (k, v) = kv.split_once('=').context(InvalidConfigSnafu {
                         value: kv.to_string(),
                         msg: "StaticUserProviderOption cmd values must be in format `user=pwd[,user=pwd]`",
-                    })
+                    })?;
+                    Ok((k.to_string(), v.as_bytes().to_vec()))
                 })
-                .collect::<Result<UserInfoMap>>()
+                .collect::<Result<HashMap<String, Vec<u8>>>>()
                 .map(|users| StaticUserProvider { users }),
             _ => InvalidConfigSnafu {
                 value: mode.to_string(),
@@ -64,7 +69,7 @@ impl StaticUserProvider {
             msg: "Expect at least one pair of username and password",
         })?;
         let username = kv.0;
-        let pwd = String::from_utf8(kv.1.0.clone()).context(FromUtf8Snafu)?;
+        let pwd = String::from_utf8(kv.1.clone()).context(FromUtf8Snafu)?;
         Ok((username.clone(), pwd))
     }
 }

src/auth/src/user_provider/watch_file_user_provider.rs

@@ -12,6 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use std::collections::HashMap;
 use std::path::Path;
 use std::sync::mpsc::channel;
 use std::sync::{Arc, Mutex};
@@ -22,17 +23,17 @@ use notify::{EventKind, RecursiveMode, Watcher};
 use snafu::{ResultExt, ensure};
 
 use crate::error::{FileWatchSnafu, InvalidConfigSnafu, Result};
-use crate::user_provider::{UserInfoMap, authenticate_with_credential, load_credential_from_file};
+use crate::user_info::DefaultUserInfo;
+use crate::user_provider::{authenticate_with_credential, load_credential_from_file};
 use crate::{Identity, Password, UserInfoRef, UserProvider};
 
 pub(crate) const WATCH_FILE_USER_PROVIDER: &str = "watch_file_user_provider";
 
-type WatchedCredentialRef = Arc<Mutex<UserInfoMap>>;
+type WatchedCredentialRef = Arc<Mutex<Option<HashMap<String, Vec<u8>>>>>;
 
 /// A user provider that reads user credential from a file and watches the file for changes.
 ///
-/// Both empty file and non-existent file are invalid and will cause initialization to fail.
+/// Empty file is invalid; but file not exist means every user can be authenticated.
-#[derive(Debug)]
 pub(crate) struct WatchFileUserProvider {
     users: WatchedCredentialRef,
 }
@@ -107,7 +108,16 @@ impl UserProvider for WatchFileUserProvider {
 
     async fn authenticate(&self, id: Identity<'_>, password: Password<'_>) -> Result<UserInfoRef> {
         let users = self.users.lock().expect("users credential must be valid");
-        authenticate_with_credential(&users, id, password)
+        if let Some(users) = users.as_ref() {
+            authenticate_with_credential(users, id, password)
+        } else {
+            match id {
+                Identity::UserId(id, _) => {
+                    warn!(id, "User provider file not exist, allow all users");
+                    Ok(DefaultUserInfo::with_name(id))
+                }
+            }
+        }
     }
 
     async fn authorize(&self, _: &str, _: &str, _: &UserInfoRef) -> Result<()> {
@@ -168,21 +178,6 @@ pub mod test {
         }
     }
 
-    #[tokio::test]
-    async fn test_file_provider_initialization_with_missing_file() {
-        common_telemetry::init_default_ut_logging();
-
-        let dir = create_temp_dir("test_missing_file");
-        let file_path = format!("{}/non_existent_file", dir.path().to_str().unwrap());
-
-        // Try to create provider with non-existent file should fail
-        let result = WatchFileUserProvider::new(file_path.as_str());
-        assert!(result.is_err());
-
-        let error = result.unwrap_err();
-        assert!(error.to_string().contains("UserProvider file must exist"));
-    }
-
     #[tokio::test]
     async fn test_file_provider() {
         common_telemetry::init_default_ut_logging();
@@ -207,10 +202,9 @@ pub mod test {
 
         // remove the tmp file
         assert!(std::fs::remove_file(&file_path).is_ok());
-        // When file is deleted during runtime, keep the last known good credentials
+        test_authenticate(&provider, "root", "123456", true, Some(timeout)).await;
         test_authenticate(&provider, "root", "654321", true, Some(timeout)).await;
-        test_authenticate(&provider, "root", "123456", false, Some(timeout)).await;
+        test_authenticate(&provider, "admin", "654321", true, Some(timeout)).await;
-        test_authenticate(&provider, "admin", "654321", false, Some(timeout)).await;
 
         // recreate the tmp file
         assert!(std::fs::write(&file_path, "root=123456\n").is_ok());

src/catalog/Cargo.toml

@@ -35,7 +35,6 @@ common-version.workspace = true
 common-workload.workspace = true
 dashmap.workspace = true
 datafusion.workspace = true
-datafusion-pg-catalog.workspace = true
 datatypes.workspace = true
 futures.workspace = true
 futures-util.workspace = true
@@ -49,6 +48,7 @@ paste.workspace = true
 prometheus.workspace = true
 promql-parser.workspace = true
 rand.workspace = true
+rustc-hash.workspace = true
 serde.workspace = true
 serde_json.workspace = true
 session.workspace = true

src/catalog/src/kvbackend/builder.rs

@@ -29,7 +29,6 @@ use crate::information_schema::{InformationExtensionRef, InformationSchemaProvid
 use crate::kvbackend::KvBackendCatalogManager;
 use crate::kvbackend::manager::{CATALOG_CACHE_MAX_CAPACITY, SystemCatalog};
 use crate::process_manager::ProcessManagerRef;
-use crate::system_schema::numbers_table_provider::NumbersTableProvider;
 use crate::system_schema::pg_catalog::PGCatalogProvider;
 
 pub struct KvBackendCatalogManagerBuilder {
@@ -120,7 +119,6 @@ impl KvBackendCatalogManagerBuilder {
                     DEFAULT_CATALOG_NAME.to_string(),
                     me.clone(),
                 )),
-                numbers_table_provider: NumbersTableProvider,
                 backend,
                 process_manager,
                 #[cfg(feature = "enterprise")]

src/catalog/src/kvbackend/manager.rs

@@ -18,7 +18,8 @@ use std::sync::{Arc, Weak};
 
 use async_stream::try_stream;
 use common_catalog::consts::{
-    DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, INFORMATION_SCHEMA_NAME, PG_CATALOG_NAME,
+    DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, INFORMATION_SCHEMA_NAME, NUMBERS_TABLE_ID,
+    PG_CATALOG_NAME,
 };
 use common_error::ext::BoxedError;
 use common_meta::cache::{
@@ -44,6 +45,7 @@ use table::TableRef;
 use table::dist_table::DistTable;
 use table::metadata::{TableId, TableInfoRef};
 use table::table::PartitionRules;
+use table::table::numbers::{NUMBERS_TABLE_NAME, NumbersTable};
 use table::table_name::TableName;
 use tokio::sync::Semaphore;
 use tokio_stream::wrappers::ReceiverStream;
@@ -59,7 +61,6 @@ use crate::information_schema::{InformationExtensionRef, InformationSchemaProvid
 use crate::kvbackend::TableCacheRef;
 use crate::process_manager::ProcessManagerRef;
 use crate::system_schema::SystemSchemaProvider;
-use crate::system_schema::numbers_table_provider::NumbersTableProvider;
 use crate::system_schema::pg_catalog::PGCatalogProvider;
 
 /// Access all existing catalog, schema and tables.
@@ -554,7 +555,6 @@ pub(super) struct SystemCatalog {
     // system_schema_provider for default catalog
     pub(super) information_schema_provider: Arc<InformationSchemaProvider>,
     pub(super) pg_catalog_provider: Arc<PGCatalogProvider>,
-    pub(super) numbers_table_provider: NumbersTableProvider,
     pub(super) backend: KvBackendRef,
     pub(super) process_manager: Option<ProcessManagerRef>,
     #[cfg(feature = "enterprise")]
@@ -584,7 +584,9 @@ impl SystemCatalog {
             PG_CATALOG_NAME if channel == Channel::Postgres => {
                 self.pg_catalog_provider.table_names()
             }
-            DEFAULT_SCHEMA_NAME => self.numbers_table_provider.table_names(),
+            DEFAULT_SCHEMA_NAME => {
+                vec![NUMBERS_TABLE_NAME.to_string()]
+            }
             _ => vec![],
         }
     }
@@ -602,7 +604,7 @@ impl SystemCatalog {
         if schema == INFORMATION_SCHEMA_NAME {
             self.information_schema_provider.table(table).is_some()
         } else if schema == DEFAULT_SCHEMA_NAME {
-            self.numbers_table_provider.table_exists(table)
+            table == NUMBERS_TABLE_NAME
         } else if schema == PG_CATALOG_NAME && channel == Channel::Postgres {
             self.pg_catalog_provider.table(table).is_some()
         } else {
@@ -647,8 +649,8 @@ impl SystemCatalog {
                     });
                 pg_catalog_provider.table(table_name)
             }
-        } else if schema == DEFAULT_SCHEMA_NAME {
+        } else if schema == DEFAULT_SCHEMA_NAME && table_name == NUMBERS_TABLE_NAME {
-            self.numbers_table_provider.table(table_name)
+            Some(NumbersTable::table(NUMBERS_TABLE_ID))
         } else {
             None
         }

src/catalog/src/lib.rs

@@ -14,6 +14,7 @@
 
 #![feature(assert_matches)]
 #![feature(try_blocks)]
+#![feature(let_chains)]
 
 use std::any::Any;
 use std::fmt::{Debug, Formatter};

src/catalog/src/memory/manager.rs

@@ -392,15 +392,15 @@ impl MemoryCatalogManager {
         if !manager.schema_exist_sync(catalog, schema).unwrap() {
             manager
                 .register_schema_sync(RegisterSchemaRequest {
-                    catalog: catalog.clone(),
+                    catalog: catalog.to_string(),
-                    schema: schema.clone(),
+                    schema: schema.to_string(),
                 })
                 .unwrap();
         }
 
         let request = RegisterTableRequest {
-            catalog: catalog.clone(),
+            catalog: catalog.to_string(),
-            schema: schema.clone(),
+            schema: schema.to_string(),
             table_name: table.table_info().name.clone(),
             table_id: table.table_info().ident.table_id,
             table,

src/catalog/src/process_manager.rs

@@ -56,21 +56,14 @@ pub struct ProcessManager {
 #[derive(Debug, Clone)]
 pub enum QueryStatement {
     Sql(Statement),
-    // The optional string is the alias of the PromQL query.
+    Promql(EvalStmt),
-    Promql(EvalStmt, Option<String>),
 }
 
 impl Display for QueryStatement {
     fn fmt(&self, f: &mut Formatter<'_>) -> std::fmt::Result {
         match self {
             QueryStatement::Sql(stmt) => write!(f, "{}", stmt),
-            QueryStatement::Promql(eval_stmt, alias) => {
+            QueryStatement::Promql(eval_stmt) => write!(f, "{}", eval_stmt),
-                if let Some(alias) = alias {
-                    write!(f, "{} AS {}", eval_stmt, alias)
-                } else {
-                    write!(f, "{}", eval_stmt)
-                }
-            }
         }
     }
 }
@@ -345,9 +338,9 @@ impl SlowQueryTimer {
         };
 
         match &self.stmt {
-            QueryStatement::Promql(stmt, _alias) => {
+            QueryStatement::Promql(stmt) => {
                 slow_query_event.is_promql = true;
-                slow_query_event.query = self.stmt.to_string();
+                slow_query_event.query = stmt.expr.to_string();
                 slow_query_event.promql_step = Some(stmt.interval.as_millis() as u64);
 
                 let start = stmt

src/catalog/src/system_schema.rs

@@ -14,7 +14,6 @@
 
 pub mod information_schema;
 mod memory_table;
-pub mod numbers_table_provider;
 pub mod pg_catalog;
 pub mod predicate;
 mod utils;
@@ -138,24 +137,21 @@ impl DataSource for SystemTableDataSource {
         &self,
         request: ScanRequest,
     ) -> std::result::Result<SendableRecordBatchStream, BoxedError> {
-        let projected_schema = match &request.projection {
+        let projection = request.projection.clone();
+        let projected_schema = match &projection {
             Some(projection) => self.try_project(projection)?,
             None => self.table.schema(),
         };
 
-        let projection = request.projection.clone();
         let stream = self
             .table
             .to_stream(request)
             .map_err(BoxedError::new)
             .context(TablesRecordBatchSnafu)
             .map_err(BoxedError::new)?
-            .map(move |batch| match (&projection, batch) {
+            .map(move |batch| match &projection {
-                // Some tables (e.g., inspect tables) already honor projection in their inner stream;
+                Some(p) => batch.and_then(|b| b.try_project(p)),
-                // others ignore it and return full rows. We will only apply projection here if the
+                None => batch,
-                // inner batch width doesn't match the projection size.
-                (Some(p), Ok(b)) if b.num_columns() != p.len() => b.try_project(p),
-                (_, res) => res,
             });
 
         let stream = RecordBatchStreamWrapper {

src/catalog/src/system_schema/information_schema.rs

@@ -24,7 +24,6 @@ pub mod region_peers;
 mod region_statistics;
 mod runtime_metrics;
 pub mod schemata;
-mod ssts;
 mod table_constraints;
 mod table_names;
 pub mod tables;
@@ -48,7 +47,7 @@ use datatypes::schema::SchemaRef;
 use lazy_static::lazy_static;
 use paste::paste;
 use process_list::InformationSchemaProcessList;
-use store_api::sst_entry::{ManifestSstEntry, PuffinIndexMetaEntry, StorageSstEntry};
+use store_api::sst_entry::{ManifestSstEntry, StorageSstEntry};
 use store_api::storage::{ScanRequest, TableId};
 use table::TableRef;
 use table::metadata::TableType;
@@ -67,9 +66,6 @@ use crate::system_schema::information_schema::partitions::InformationSchemaParti
 use crate::system_schema::information_schema::region_peers::InformationSchemaRegionPeers;
 use crate::system_schema::information_schema::runtime_metrics::InformationSchemaMetrics;
 use crate::system_schema::information_schema::schemata::InformationSchemaSchemata;
-use crate::system_schema::information_schema::ssts::{
-    InformationSchemaSstsIndexMeta, InformationSchemaSstsManifest, InformationSchemaSstsStorage,
-};
 use crate::system_schema::information_schema::table_constraints::InformationSchemaTableConstraints;
 use crate::system_schema::information_schema::tables::InformationSchemaTables;
 use crate::system_schema::memory_table::MemoryTable;
@@ -257,15 +253,6 @@ impl SystemSchemaProviderInner for InformationSchemaProvider {
                 .process_manager
                 .as_ref()
                 .map(|p| Arc::new(InformationSchemaProcessList::new(p.clone())) as _),
-            SSTS_MANIFEST => Some(Arc::new(InformationSchemaSstsManifest::new(
-                self.catalog_manager.clone(),
-            )) as _),
-            SSTS_STORAGE => Some(Arc::new(InformationSchemaSstsStorage::new(
-                self.catalog_manager.clone(),
-            )) as _),
-            SSTS_INDEX_META => Some(Arc::new(InformationSchemaSstsIndexMeta::new(
-                self.catalog_manager.clone(),
-            )) as _),
             _ => None,
         }
     }
@@ -337,18 +324,6 @@ impl InformationSchemaProvider {
                 REGION_STATISTICS.to_string(),
                 self.build_table(REGION_STATISTICS).unwrap(),
             );
-            tables.insert(
-                SSTS_MANIFEST.to_string(),
-                self.build_table(SSTS_MANIFEST).unwrap(),
-            );
-            tables.insert(
-                SSTS_STORAGE.to_string(),
-                self.build_table(SSTS_STORAGE).unwrap(),
-            );
-            tables.insert(
-                SSTS_INDEX_META.to_string(),
-                self.build_table(SSTS_INDEX_META).unwrap(),
-            );
         }
 
         tables.insert(TABLES.to_string(), self.build_table(TABLES).unwrap());
@@ -369,7 +344,7 @@ impl InformationSchemaProvider {
         }
         #[cfg(feature = "enterprise")]
         for name in self.extra_table_factories.keys() {
-            tables.insert(name.clone(), self.build_table(name).expect(name));
+            tables.insert(name.to_string(), self.build_table(name).expect(name));
         }
         // Add memory tables
         for name in MEMORY_TABLES.iter() {
@@ -463,8 +438,6 @@ pub enum DatanodeInspectKind {
     SstManifest,
     /// List SST entries discovered in storage layer
     SstStorage,
-    /// List index metadata collected from manifest
-    SstIndexMeta,
 }
 
 impl DatanodeInspectRequest {
@@ -473,7 +446,6 @@ impl DatanodeInspectRequest {
         match self.kind {
             DatanodeInspectKind::SstManifest => ManifestSstEntry::build_plan(self.scan),
             DatanodeInspectKind::SstStorage => StorageSstEntry::build_plan(self.scan),
-            DatanodeInspectKind::SstIndexMeta => PuffinIndexMetaEntry::build_plan(self.scan),
         }
     }
 }

src/catalog/src/system_schema/information_schema/cluster_info.rs

@@ -33,6 +33,7 @@ use datatypes::timestamp::TimestampMillisecond;
 use datatypes::value::Value;
 use datatypes::vectors::{
     Int64VectorBuilder, StringVectorBuilder, TimestampMillisecondVectorBuilder,
+    UInt32VectorBuilder, UInt64VectorBuilder,
 };
 use serde::Serialize;
 use snafu::ResultExt;
@@ -49,11 +50,8 @@ const PEER_TYPE_METASRV: &str = "METASRV";
 const PEER_ID: &str = "peer_id";
 const PEER_TYPE: &str = "peer_type";
 const PEER_ADDR: &str = "peer_addr";
-const PEER_HOSTNAME: &str = "peer_hostname";
+const CPUS: &str = "cpus";
-const TOTAL_CPU_MILLICORES: &str = "total_cpu_millicores";
+const MEMORY_BYTES: &str = "memory_bytes";
-const TOTAL_MEMORY_BYTES: &str = "total_memory_bytes";
-const CPU_USAGE_MILLICORES: &str = "cpu_usage_millicores";
-const MEMORY_USAGE_BYTES: &str = "memory_usage_bytes";
 const VERSION: &str = "version";
 const GIT_COMMIT: &str = "git_commit";
 const START_TIME: &str = "start_time";
@@ -68,11 +66,8 @@ const INIT_CAPACITY: usize = 42;
 /// - `peer_id`: the peer server id.
 /// - `peer_type`: the peer type, such as `datanode`, `frontend`, `metasrv` etc.
 /// - `peer_addr`: the peer gRPC address.
-/// - `peer_hostname`: the hostname of the peer.
+/// - `cpus`: the number of CPUs of the peer.
-/// - `total_cpu_millicores`: the total CPU millicores of the peer.
+/// - `memory_bytes`: the memory bytes of the peer.
-/// - `total_memory_bytes`: the total memory bytes of the peer.
-/// - `cpu_usage_millicores`: the CPU usage millicores of the peer.
-/// - `memory_usage_bytes`: the memory usage bytes of the peer.
 /// - `version`: the build package version of the peer.
 /// - `git_commit`: the build git commit hash of the peer.
 /// - `start_time`: the starting time of the peer.
@@ -99,27 +94,8 @@ impl InformationSchemaClusterInfo {
             ColumnSchema::new(PEER_ID, ConcreteDataType::int64_datatype(), false),
             ColumnSchema::new(PEER_TYPE, ConcreteDataType::string_datatype(), false),
             ColumnSchema::new(PEER_ADDR, ConcreteDataType::string_datatype(), true),
-            ColumnSchema::new(PEER_HOSTNAME, ConcreteDataType::string_datatype(), true),
+            ColumnSchema::new(CPUS, ConcreteDataType::uint32_datatype(), false),
-            ColumnSchema::new(
+            ColumnSchema::new(MEMORY_BYTES, ConcreteDataType::uint64_datatype(), false),
-                TOTAL_CPU_MILLICORES,
-                ConcreteDataType::int64_datatype(),
-                false,
-            ),
-            ColumnSchema::new(
-                TOTAL_MEMORY_BYTES,
-                ConcreteDataType::int64_datatype(),
-                false,
-            ),
-            ColumnSchema::new(
-                CPU_USAGE_MILLICORES,
-                ConcreteDataType::int64_datatype(),
-                false,
-            ),
-            ColumnSchema::new(
-                MEMORY_USAGE_BYTES,
-                ConcreteDataType::int64_datatype(),
-                false,
-            ),
             ColumnSchema::new(VERSION, ConcreteDataType::string_datatype(), false),
             ColumnSchema::new(GIT_COMMIT, ConcreteDataType::string_datatype(), false),
             ColumnSchema::new(
@@ -179,11 +155,8 @@ struct InformationSchemaClusterInfoBuilder {
     peer_ids: Int64VectorBuilder,
     peer_types: StringVectorBuilder,
     peer_addrs: StringVectorBuilder,
-    peer_hostnames: StringVectorBuilder,
+    cpus: UInt32VectorBuilder,
-    total_cpu_millicores: Int64VectorBuilder,
+    memory_bytes: UInt64VectorBuilder,
-    total_memory_bytes: Int64VectorBuilder,
-    cpu_usage_millicores: Int64VectorBuilder,
-    memory_usage_bytes: Int64VectorBuilder,
     versions: StringVectorBuilder,
     git_commits: StringVectorBuilder,
     start_times: TimestampMillisecondVectorBuilder,
@@ -200,11 +173,8 @@ impl InformationSchemaClusterInfoBuilder {
             peer_ids: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
             peer_types: StringVectorBuilder::with_capacity(INIT_CAPACITY),
             peer_addrs: StringVectorBuilder::with_capacity(INIT_CAPACITY),
-            peer_hostnames: StringVectorBuilder::with_capacity(INIT_CAPACITY),
+            cpus: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
-            total_cpu_millicores: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
+            memory_bytes: UInt64VectorBuilder::with_capacity(INIT_CAPACITY),
-            total_memory_bytes: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
-            cpu_usage_millicores: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
-            memory_usage_bytes: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
             versions: StringVectorBuilder::with_capacity(INIT_CAPACITY),
             git_commits: StringVectorBuilder::with_capacity(INIT_CAPACITY),
             start_times: TimestampMillisecondVectorBuilder::with_capacity(INIT_CAPACITY),
@@ -233,7 +203,6 @@ impl InformationSchemaClusterInfoBuilder {
             (PEER_ID, &Value::from(peer_id)),
             (PEER_TYPE, &Value::from(peer_type)),
             (PEER_ADDR, &Value::from(node_info.peer.addr.as_str())),
-            (PEER_HOSTNAME, &Value::from(node_info.hostname.as_str())),
             (VERSION, &Value::from(node_info.version.as_str())),
             (GIT_COMMIT, &Value::from(node_info.git_commit.as_str())),
         ];
@@ -245,7 +214,6 @@ impl InformationSchemaClusterInfoBuilder {
         self.peer_ids.push(Some(peer_id));
         self.peer_types.push(Some(peer_type));
         self.peer_addrs.push(Some(&node_info.peer.addr));
-        self.peer_hostnames.push(Some(&node_info.hostname));
         self.versions.push(Some(&node_info.version));
         self.git_commits.push(Some(&node_info.git_commit));
         if node_info.start_time_ms > 0 {
@@ -260,14 +228,8 @@ impl InformationSchemaClusterInfoBuilder {
             self.start_times.push(None);
             self.uptimes.push(None);
         }
-        self.total_cpu_millicores
+        self.cpus.push(Some(node_info.cpus));
-            .push(Some(node_info.total_cpu_millicores));
+        self.memory_bytes.push(Some(node_info.memory_bytes));
-        self.total_memory_bytes
-            .push(Some(node_info.total_memory_bytes));
-        self.cpu_usage_millicores
-            .push(Some(node_info.cpu_usage_millicores));
-        self.memory_usage_bytes
-            .push(Some(node_info.memory_usage_bytes));
 
         if node_info.last_activity_ts > 0 {
             self.active_times.push(Some(
@@ -291,11 +253,8 @@ impl InformationSchemaClusterInfoBuilder {
             Arc::new(self.peer_ids.finish()),
             Arc::new(self.peer_types.finish()),
             Arc::new(self.peer_addrs.finish()),
-            Arc::new(self.peer_hostnames.finish()),
+            Arc::new(self.cpus.finish()),
-            Arc::new(self.total_cpu_millicores.finish()),
+            Arc::new(self.memory_bytes.finish()),
-            Arc::new(self.total_memory_bytes.finish()),
-            Arc::new(self.cpu_usage_millicores.finish()),
-            Arc::new(self.memory_usage_bytes.finish()),
             Arc::new(self.versions.finish()),
             Arc::new(self.git_commits.finish()),
             Arc::new(self.start_times.finish()),

src/catalog/src/system_schema/information_schema/flows.rs

@@ -254,9 +254,9 @@ impl InformationSchemaFlowsBuilder {
                 .await
                 .map_err(BoxedError::new)
                 .context(InternalSnafu)?
-                .with_context(|| FlowInfoNotFoundSnafu {
+                .context(FlowInfoNotFoundSnafu {
-                    catalog_name: catalog_name.clone(),
+                    catalog_name: catalog_name.to_string(),
-                    flow_name: flow_name.clone(),
+                    flow_name: flow_name.to_string(),
                 })?;
             self.add_flow(&predicates, flow_id.flow_id(), flow_info, &flow_stat)
                 .await?;
@@ -273,11 +273,11 @@ impl InformationSchemaFlowsBuilder {
         flow_stat: &Option<FlowStat>,
     ) -> Result<()> {
         let row = [
-            (FLOW_NAME, &Value::from(flow_info.flow_name().clone())),
+            (FLOW_NAME, &Value::from(flow_info.flow_name().to_string())),
             (FLOW_ID, &Value::from(flow_id)),
             (
                 TABLE_CATALOG,
-                &Value::from(flow_info.catalog_name().clone()),
+                &Value::from(flow_info.catalog_name().to_string()),
             ),
         ];
         if !predicates.eval(&row) {

src/catalog/src/system_schema/information_schema/partitions.rs

@@ -26,11 +26,12 @@ use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatch
 use datafusion::physical_plan::streaming::PartitionStream as DfPartitionStream;
 use datatypes::prelude::{ConcreteDataType, ScalarVectorBuilder, VectorRef};
 use datatypes::schema::{ColumnSchema, Schema, SchemaRef};
-use datatypes::timestamp::TimestampSecond;
+use datatypes::timestamp::TimestampMicrosecond;
 use datatypes::value::Value;
 use datatypes::vectors::{
     ConstantVector, Int64Vector, Int64VectorBuilder, MutableVector, StringVector,
-    StringVectorBuilder, TimestampSecondVector, TimestampSecondVectorBuilder, UInt64VectorBuilder,
+    StringVectorBuilder, TimestampMicrosecondVector, TimestampMicrosecondVectorBuilder,
+    UInt64VectorBuilder,
 };
 use futures::{StreamExt, TryStreamExt};
 use partition::manager::PartitionInfo;
@@ -128,17 +129,17 @@ impl InformationSchemaPartitions {
             ColumnSchema::new("data_free", ConcreteDataType::int64_datatype(), true),
             ColumnSchema::new(
                 "create_time",
-                ConcreteDataType::timestamp_second_datatype(),
+                ConcreteDataType::timestamp_microsecond_datatype(),
                 true,
             ),
             ColumnSchema::new(
                 "update_time",
-                ConcreteDataType::timestamp_second_datatype(),
+                ConcreteDataType::timestamp_microsecond_datatype(),
                 true,
             ),
             ColumnSchema::new(
                 "check_time",
-                ConcreteDataType::timestamp_second_datatype(),
+                ConcreteDataType::timestamp_microsecond_datatype(),
                 true,
             ),
             ColumnSchema::new("checksum", ConcreteDataType::int64_datatype(), true),
@@ -211,7 +212,7 @@ struct InformationSchemaPartitionsBuilder {
     partition_names: StringVectorBuilder,
     partition_ordinal_positions: Int64VectorBuilder,
     partition_expressions: StringVectorBuilder,
-    create_times: TimestampSecondVectorBuilder,
+    create_times: TimestampMicrosecondVectorBuilder,
     partition_ids: UInt64VectorBuilder,
 }
 
@@ -231,7 +232,7 @@ impl InformationSchemaPartitionsBuilder {
             partition_names: StringVectorBuilder::with_capacity(INIT_CAPACITY),
             partition_ordinal_positions: Int64VectorBuilder::with_capacity(INIT_CAPACITY),
             partition_expressions: StringVectorBuilder::with_capacity(INIT_CAPACITY),
-            create_times: TimestampSecondVectorBuilder::with_capacity(INIT_CAPACITY),
+            create_times: TimestampMicrosecondVectorBuilder::with_capacity(INIT_CAPACITY),
             partition_ids: UInt64VectorBuilder::with_capacity(INIT_CAPACITY),
         }
     }
@@ -330,8 +331,8 @@ impl InformationSchemaPartitionsBuilder {
                 .push(Some((index + 1) as i64));
             let expression = partition.partition_expr.as_ref().map(|e| e.to_string());
             self.partition_expressions.push(expression.as_deref());
-            self.create_times.push(Some(TimestampSecond::from(
+            self.create_times.push(Some(TimestampMicrosecond::from(
-                table_info.meta.created_on.timestamp(),
+                table_info.meta.created_on.timestamp_millis(),
             )));
             self.partition_ids.push(Some(partition.id.as_u64()));
         }
@@ -348,8 +349,8 @@ impl InformationSchemaPartitionsBuilder {
             Arc::new(Int64Vector::from(vec![None])),
             rows_num,
         ));
-        let null_timestamp_second_vector = Arc::new(ConstantVector::new(
+        let null_timestampmicrosecond_vector = Arc::new(ConstantVector::new(
-            Arc::new(TimestampSecondVector::from(vec![None])),
+            Arc::new(TimestampMicrosecondVector::from(vec![None])),
             rows_num,
         ));
         let partition_methods = Arc::new(ConstantVector::new(
@@ -379,8 +380,8 @@ impl InformationSchemaPartitionsBuilder {
             null_i64_vector.clone(),
             Arc::new(self.create_times.finish()),
             // TODO(dennis): supports update_time
-            null_timestamp_second_vector.clone(),
+            null_timestampmicrosecond_vector.clone(),
-            null_timestamp_second_vector,
+            null_timestampmicrosecond_vector,
             null_i64_vector,
             null_string_vector.clone(),
             null_string_vector.clone(),

src/catalog/src/system_schema/information_schema/process_list.rs

@@ -135,7 +135,7 @@ async fn make_process_list(
 
     for process in queries {
         let display_id = DisplayProcessId {
-            server_addr: process.frontend.clone(),
+            server_addr: process.frontend.to_string(),
             id: process.id,
         }
         .to_string();

src/catalog/src/system_schema/information_schema/region_peers.rs

@@ -199,7 +199,10 @@ impl InformationSchemaRegionPeersBuilder {
                     if table_info.table_type == TableType::Temporary {
                         Ok(None)
                     } else {
-                        Ok(Some((table_info.ident.table_id, table_info.name.clone())))
+                        Ok(Some((
+                            table_info.ident.table_id,
+                            table_info.name.to_string(),
+                        )))
                     }
                 });
 

src/catalog/src/system_schema/information_schema/ssts.rs

@@ -1,199 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use std::sync::{Arc, Weak};
-
-use common_catalog::consts::{
-    INFORMATION_SCHEMA_SSTS_INDEX_META_TABLE_ID, INFORMATION_SCHEMA_SSTS_MANIFEST_TABLE_ID,
-    INFORMATION_SCHEMA_SSTS_STORAGE_TABLE_ID,
-};
-use common_error::ext::BoxedError;
-use common_recordbatch::SendableRecordBatchStream;
-use common_recordbatch::adapter::AsyncRecordBatchStreamAdapter;
-use datatypes::schema::SchemaRef;
-use snafu::ResultExt;
-use store_api::sst_entry::{ManifestSstEntry, PuffinIndexMetaEntry, StorageSstEntry};
-use store_api::storage::{ScanRequest, TableId};
-
-use crate::CatalogManager;
-use crate::error::{ProjectSchemaSnafu, Result};
-use crate::information_schema::{
-    DatanodeInspectKind, DatanodeInspectRequest, InformationTable, SSTS_INDEX_META, SSTS_MANIFEST,
-    SSTS_STORAGE,
-};
-use crate::system_schema::utils;
-
-/// Information schema table for sst manifest.
-pub struct InformationSchemaSstsManifest {
-    schema: SchemaRef,
-    catalog_manager: Weak<dyn CatalogManager>,
-}
-
-impl InformationSchemaSstsManifest {
-    pub(super) fn new(catalog_manager: Weak<dyn CatalogManager>) -> Self {
-        Self {
-            schema: ManifestSstEntry::schema(),
-            catalog_manager,
-        }
-    }
-}
-
-impl InformationTable for InformationSchemaSstsManifest {
-    fn table_id(&self) -> TableId {
-        INFORMATION_SCHEMA_SSTS_MANIFEST_TABLE_ID
-    }
-
-    fn table_name(&self) -> &'static str {
-        SSTS_MANIFEST
-    }
-
-    fn schema(&self) -> SchemaRef {
-        self.schema.clone()
-    }
-
-    fn to_stream(&self, request: ScanRequest) -> Result<SendableRecordBatchStream> {
-        let schema = if let Some(p) = &request.projection {
-            Arc::new(self.schema.try_project(p).context(ProjectSchemaSnafu)?)
-        } else {
-            self.schema.clone()
-        };
-        let info_ext = utils::information_extension(&self.catalog_manager)?;
-        let req = DatanodeInspectRequest {
-            kind: DatanodeInspectKind::SstManifest,
-            scan: request,
-        };
-
-        let future = async move {
-            info_ext
-                .inspect_datanode(req)
-                .await
-                .map_err(BoxedError::new)
-                .context(common_recordbatch::error::ExternalSnafu)
-        };
-        Ok(Box::pin(AsyncRecordBatchStreamAdapter::new(
-            schema,
-            Box::pin(future),
-        )))
-    }
-}
-
-/// Information schema table for sst storage.
-pub struct InformationSchemaSstsStorage {
-    schema: SchemaRef,
-    catalog_manager: Weak<dyn CatalogManager>,
-}
-
-impl InformationSchemaSstsStorage {
-    pub(super) fn new(catalog_manager: Weak<dyn CatalogManager>) -> Self {
-        Self {
-            schema: StorageSstEntry::schema(),
-            catalog_manager,
-        }
-    }
-}
-
-impl InformationTable for InformationSchemaSstsStorage {
-    fn table_id(&self) -> TableId {
-        INFORMATION_SCHEMA_SSTS_STORAGE_TABLE_ID
-    }
-
-    fn table_name(&self) -> &'static str {
-        SSTS_STORAGE
-    }
-
-    fn schema(&self) -> SchemaRef {
-        self.schema.clone()
-    }
-
-    fn to_stream(&self, request: ScanRequest) -> Result<SendableRecordBatchStream> {
-        let schema = if let Some(p) = &request.projection {
-            Arc::new(self.schema.try_project(p).context(ProjectSchemaSnafu)?)
-        } else {
-            self.schema.clone()
-        };
-
-        let info_ext = utils::information_extension(&self.catalog_manager)?;
-        let req = DatanodeInspectRequest {
-            kind: DatanodeInspectKind::SstStorage,
-            scan: request,
-        };
-
-        let future = async move {
-            info_ext
-                .inspect_datanode(req)
-                .await
-                .map_err(BoxedError::new)
-                .context(common_recordbatch::error::ExternalSnafu)
-        };
-        Ok(Box::pin(AsyncRecordBatchStreamAdapter::new(
-            schema,
-            Box::pin(future),
-        )))
-    }
-}
-
-/// Information schema table for index metadata.
-pub struct InformationSchemaSstsIndexMeta {
-    schema: SchemaRef,
-    catalog_manager: Weak<dyn CatalogManager>,
-}
-
-impl InformationSchemaSstsIndexMeta {
-    pub(super) fn new(catalog_manager: Weak<dyn CatalogManager>) -> Self {
-        Self {
-            schema: PuffinIndexMetaEntry::schema(),
-            catalog_manager,
-        }
-    }
-}
-
-impl InformationTable for InformationSchemaSstsIndexMeta {
-    fn table_id(&self) -> TableId {
-        INFORMATION_SCHEMA_SSTS_INDEX_META_TABLE_ID
-    }
-
-    fn table_name(&self) -> &'static str {
-        SSTS_INDEX_META
-    }
-
-    fn schema(&self) -> SchemaRef {
-        self.schema.clone()
-    }
-
-    fn to_stream(&self, request: ScanRequest) -> Result<SendableRecordBatchStream> {
-        let schema = if let Some(p) = &request.projection {
-            Arc::new(self.schema.try_project(p).context(ProjectSchemaSnafu)?)
-        } else {
-            self.schema.clone()
-        };
-
-        let info_ext = utils::information_extension(&self.catalog_manager)?;
-        let req = DatanodeInspectRequest {
-            kind: DatanodeInspectKind::SstIndexMeta,
-            scan: request,
-        };
-
-        let future = async move {
-            info_ext
-                .inspect_datanode(req)
-                .await
-                .map_err(BoxedError::new)
-                .context(common_recordbatch::error::ExternalSnafu)
-        };
-        Ok(Box::pin(AsyncRecordBatchStreamAdapter::new(
-            schema,
-            Box::pin(future),
-        )))
-    }
-}

src/catalog/src/system_schema/information_schema/table_names.rs

@@ -48,6 +48,3 @@ pub const FLOWS: &str = "flows";
 pub const PROCEDURE_INFO: &str = "procedure_info";
 pub const REGION_STATISTICS: &str = "region_statistics";
 pub const PROCESS_LIST: &str = "process_list";
-pub const SSTS_MANIFEST: &str = "ssts_manifest";
-pub const SSTS_STORAGE: &str = "ssts_storage";
-pub const SSTS_INDEX_META: &str = "ssts_index_meta";

src/catalog/src/system_schema/information_schema/tables.rs

@@ -371,8 +371,7 @@ impl InformationSchemaTablesBuilder {
         self.auto_increment.push(Some(0));
         self.row_format.push(Some("Fixed"));
         self.table_collation.push(Some("utf8_bin"));
-        self.update_time
+        self.update_time.push(None);
-            .push(Some(table_info.meta.updated_on.timestamp().into()));
         self.check_time.push(None);
         // use mariadb default table version number here
         self.version.push(Some(11));

src/catalog/src/system_schema/numbers_table_provider.rs

@@ -1,59 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-#[cfg(any(test, feature = "testing", debug_assertions))]
-use common_catalog::consts::NUMBERS_TABLE_ID;
-use table::TableRef;
-#[cfg(any(test, feature = "testing", debug_assertions))]
-use table::table::numbers::NUMBERS_TABLE_NAME;
-#[cfg(any(test, feature = "testing", debug_assertions))]
-use table::table::numbers::NumbersTable;
-
-// NumbersTableProvider is a dedicated provider for feature-gating the numbers table.
-#[derive(Clone)]
-pub struct NumbersTableProvider;
-
-#[cfg(any(test, feature = "testing", debug_assertions))]
-impl NumbersTableProvider {
-    pub(crate) fn table_exists(&self, name: &str) -> bool {
-        name == NUMBERS_TABLE_NAME
-    }
-
-    pub(crate) fn table_names(&self) -> Vec<String> {
-        vec![NUMBERS_TABLE_NAME.to_string()]
-    }
-
-    pub(crate) fn table(&self, name: &str) -> Option<TableRef> {
-        if name == NUMBERS_TABLE_NAME {
-            Some(NumbersTable::table(NUMBERS_TABLE_ID))
-        } else {
-            None
-        }
-    }
-}
-
-#[cfg(not(any(test, feature = "testing", debug_assertions)))]
-impl NumbersTableProvider {
-    pub(crate) fn table_exists(&self, _name: &str) -> bool {
-        false
-    }
-
-    pub(crate) fn table_names(&self) -> Vec<String> {
-        vec![]
-    }
-
-    pub(crate) fn table(&self, _name: &str) -> Option<TableRef> {
-        None
-    }
-}

src/catalog/src/system_schema/pg_catalog.rs

@@ -12,42 +12,53 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+mod pg_catalog_memory_table;
+mod pg_class;
+mod pg_database;
+mod pg_namespace;
+mod table_names;
+
 use std::collections::HashMap;
-use std::sync::{Arc, Weak};
+use std::sync::{Arc, LazyLock, Weak};
 
-use arrow_schema::SchemaRef;
+use common_catalog::consts::{self, DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, PG_CATALOG_NAME};
-use async_trait::async_trait;
+use datatypes::schema::ColumnSchema;
-use common_catalog::consts::{DEFAULT_CATALOG_NAME, PG_CATALOG_NAME, PG_CATALOG_TABLE_ID_START};
+use lazy_static::lazy_static;
-use common_error::ext::BoxedError;
+use paste::paste;
-use common_recordbatch::SendableRecordBatchStream;
+use pg_catalog_memory_table::get_schema_columns;
-use common_recordbatch::adapter::RecordBatchStreamAdapter;
+use pg_class::PGClass;
-use common_telemetry::warn;
+use pg_database::PGDatabase;
-use datafusion::datasource::TableType;
+use pg_namespace::PGNamespace;
-use datafusion::error::DataFusionError;
+use session::context::{Channel, QueryContext};
-use datafusion::execution::TaskContext;
-use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
-use datafusion_pg_catalog::pg_catalog::catalog_info::CatalogInfo;
-use datafusion_pg_catalog::pg_catalog::context::EmptyContextProvider;
-use datafusion_pg_catalog::pg_catalog::{
-    PG_CATALOG_TABLES, PgCatalogSchemaProvider, PgCatalogStaticTables, PgCatalogTable,
-};
-use snafu::ResultExt;
-use store_api::storage::ScanRequest;
 use table::TableRef;
-use table::metadata::TableId;
+pub use table_names::*;
 
+use self::pg_namespace::oid_map::{PGNamespaceOidMap, PGNamespaceOidMapRef};
 use crate::CatalogManager;
-use crate::error::{InternalSnafu, ProjectSchemaSnafu, Result};
+use crate::system_schema::memory_table::MemoryTable;
-use crate::system_schema::{
+use crate::system_schema::utils::tables::u32_column;
-    SystemSchemaProvider, SystemSchemaProviderInner, SystemTable, SystemTableRef,
+use crate::system_schema::{SystemSchemaProvider, SystemSchemaProviderInner, SystemTableRef};
-};
+
+lazy_static! {
+    static ref MEMORY_TABLES: &'static [&'static str] = &[table_names::PG_TYPE];
+}
+
+/// The column name for the OID column.
+/// The OID column is a unique identifier of type u32 for each object in the database.
+const OID_COLUMN_NAME: &str = "oid";
+
+fn oid_column() -> ColumnSchema {
+    u32_column(OID_COLUMN_NAME)
+}
 
 /// [`PGCatalogProvider`] is the provider for a schema named `pg_catalog`, it is not a catalog.
 pub struct PGCatalogProvider {
     catalog_name: String,
-    inner: PgCatalogSchemaProvider<CatalogManagerWrapper, EmptyContextProvider>,
+    catalog_manager: Weak<dyn CatalogManager>,
     tables: HashMap<String, TableRef>,
-    table_ids: HashMap<&'static str, u32>,
+
+    // Workaround to store mapping of schema_name to a numeric id
+    namespace_oid_map: PGNamespaceOidMapRef,
 }
 
 impl SystemSchemaProvider for PGCatalogProvider {
@@ -58,34 +69,30 @@ impl SystemSchemaProvider for PGCatalogProvider {
     }
 }
 
+// TODO(j0hn50n133): Not sure whether to avoid duplication with `information_schema` or not.
+macro_rules! setup_memory_table {
+    ($name: expr) => {
+        paste! {
+            {
+                let (schema, columns) = get_schema_columns($name);
+                Some(Arc::new(MemoryTable::new(
+                    consts::[<PG_CATALOG_ $name  _TABLE_ID>],
+                    $name,
+                    schema,
+                    columns
+                )) as _)
+            }
+        }
+    };
+}
+
 impl PGCatalogProvider {
     pub fn new(catalog_name: String, catalog_manager: Weak<dyn CatalogManager>) -> Self {
-        // safe to expect/unwrap because it contains only schema read, this can
-        // be ensured by sqlness tests
-        let static_tables =
-            PgCatalogStaticTables::try_new().expect("Failed to initialize static tables");
-        let inner = PgCatalogSchemaProvider::try_new(
-            CatalogManagerWrapper {
-                catalog_name: catalog_name.clone(),
-                catalog_manager,
-            },
-            Arc::new(static_tables),
-            EmptyContextProvider,
-        )
-        .expect("Failed to initialize PgCatalogSchemaProvider");
-
-        let mut table_ids = HashMap::new();
-        let mut table_id = PG_CATALOG_TABLE_ID_START;
-        for name in PG_CATALOG_TABLES {
-            table_ids.insert(*name, table_id);
-            table_id += 1;
-        }
-
         let mut provider = Self {
             catalog_name,
-            inner,
+            catalog_manager,
             tables: HashMap::new(),
-            table_ids,
+            namespace_oid_map: Arc::new(PGNamespaceOidMap::new()),
         };
         provider.build_tables();
         provider
@@ -95,13 +102,23 @@ impl PGCatalogProvider {
         // SECURITY NOTE:
         // Must follow the same security rules as [`InformationSchemaProvider::build_tables`].
         let mut tables = HashMap::new();
+        // TODO(J0HN50N133): modeling the table_name as a enum type to get rid of expect/unwrap here
         // It's safe to unwrap here because we are sure that the constants have been handle correctly inside system_table.
-        for name in PG_CATALOG_TABLES {
+        for name in MEMORY_TABLES.iter() {
-            if let Some(table) = self.build_table(name) {
+            tables.insert(name.to_string(), self.build_table(name).expect(name));
-                tables.insert(name.to_string(), table);
-            }
         }
-
+        tables.insert(
+            PG_NAMESPACE.to_string(),
+            self.build_table(PG_NAMESPACE).expect(PG_NAMESPACE),
+        );
+        tables.insert(
+            PG_CLASS.to_string(),
+            self.build_table(PG_CLASS).expect(PG_NAMESPACE),
+        );
+        tables.insert(
+            PG_DATABASE.to_string(),
+            self.build_table(PG_DATABASE).expect(PG_DATABASE),
+        );
         self.tables = tables;
     }
 }
@@ -112,26 +129,24 @@ impl SystemSchemaProviderInner for PGCatalogProvider {
     }
 
     fn system_table(&self, name: &str) -> Option<SystemTableRef> {
-        if let Some((table_name, table_id)) = self.table_ids.get_key_value(name) {
+        match name {
-            let table = self.inner.build_table_by_name(name).expect(name);
+            table_names::PG_TYPE => setup_memory_table!(PG_TYPE),
-
+            table_names::PG_NAMESPACE => Some(Arc::new(PGNamespace::new(
-            if let Some(table) = table {
+                self.catalog_name.clone(),
-                if let Ok(system_table) = DFTableProviderAsSystemTable::try_new(
+                self.catalog_manager.clone(),
-                    *table_id,
+                self.namespace_oid_map.clone(),
-                    table_name,
+            ))),
-                    table::metadata::TableType::Temporary,
+            table_names::PG_CLASS => Some(Arc::new(PGClass::new(
-                    table,
+                self.catalog_name.clone(),
-                ) {
+                self.catalog_manager.clone(),
-                    Some(Arc::new(system_table))
+                self.namespace_oid_map.clone(),
-                } else {
+            ))),
-                    warn!("failed to create pg_catalog system table {}", name);
+            table_names::PG_DATABASE => Some(Arc::new(PGDatabase::new(
-                    None
+                self.catalog_name.clone(),
-                }
+                self.catalog_manager.clone(),
-            } else {
+                self.namespace_oid_map.clone(),
-                None
+            ))),
-            }
+            _ => None,
-        } else {
-            None
         }
     }
 
@@ -140,177 +155,11 @@ impl SystemSchemaProviderInner for PGCatalogProvider {
     }
 }
 
-#[derive(Clone)]
+/// Provide query context to call the [`CatalogManager`]'s method.
-pub struct CatalogManagerWrapper {
+static PG_QUERY_CTX: LazyLock<QueryContext> = LazyLock::new(|| {
-    catalog_name: String,
+    QueryContext::with_channel(DEFAULT_CATALOG_NAME, DEFAULT_SCHEMA_NAME, Channel::Postgres)
-    catalog_manager: Weak<dyn CatalogManager>,
+});
-}
+
-
+fn query_ctx() -> Option<&'static QueryContext> {
-impl CatalogManagerWrapper {
+    Some(&PG_QUERY_CTX)
-    fn catalog_manager(&self) -> std::result::Result<Arc<dyn CatalogManager>, DataFusionError> {
-        self.catalog_manager.upgrade().ok_or_else(|| {
-            DataFusionError::Internal("Failed to access catalog manager".to_string())
-        })
-    }
-}
-
-impl std::fmt::Debug for CatalogManagerWrapper {
-    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
-        f.debug_struct("CatalogManagerWrapper").finish()
-    }
-}
-
-#[async_trait]
-impl CatalogInfo for CatalogManagerWrapper {
-    async fn catalog_names(&self) -> std::result::Result<Vec<String>, DataFusionError> {
-        if self.catalog_name == DEFAULT_CATALOG_NAME {
-            CatalogManager::catalog_names(self.catalog_manager()?.as_ref())
-                .await
-                .map_err(|e| DataFusionError::External(Box::new(e)))
-        } else {
-            Ok(vec![self.catalog_name.clone()])
-        }
-    }
-
-    async fn schema_names(
-        &self,
-        catalog_name: &str,
-    ) -> std::result::Result<Option<Vec<String>>, DataFusionError> {
-        self.catalog_manager()?
-            .schema_names(catalog_name, None)
-            .await
-            .map(Some)
-            .map_err(|e| DataFusionError::External(Box::new(e)))
-    }
-
-    async fn table_names(
-        &self,
-        catalog_name: &str,
-        schema_name: &str,
-    ) -> std::result::Result<Option<Vec<String>>, DataFusionError> {
-        self.catalog_manager()?
-            .table_names(catalog_name, schema_name, None)
-            .await
-            .map(Some)
-            .map_err(|e| DataFusionError::External(Box::new(e)))
-    }
-
-    async fn table_schema(
-        &self,
-        catalog_name: &str,
-        schema_name: &str,
-        table_name: &str,
-    ) -> std::result::Result<Option<SchemaRef>, DataFusionError> {
-        let table = self
-            .catalog_manager()?
-            .table(catalog_name, schema_name, table_name, None)
-            .await
-            .map_err(|e| DataFusionError::External(Box::new(e)))?;
-
-        Ok(table.map(|t| t.schema().arrow_schema().clone()))
-    }
-
-    async fn table_type(
-        &self,
-        catalog_name: &str,
-        schema_name: &str,
-        table_name: &str,
-    ) -> std::result::Result<Option<TableType>, DataFusionError> {
-        let table = self
-            .catalog_manager()?
-            .table(catalog_name, schema_name, table_name, None)
-            .await
-            .map_err(|e| DataFusionError::External(Box::new(e)))?;
-
-        Ok(table.map(|t| t.table_type().into()))
-    }
-}
-
-struct DFTableProviderAsSystemTable {
-    pub table_id: TableId,
-    pub table_name: &'static str,
-    pub table_type: table::metadata::TableType,
-    pub schema: Arc<datatypes::schema::Schema>,
-    pub table_provider: PgCatalogTable,
-}
-
-impl DFTableProviderAsSystemTable {
-    pub fn try_new(
-        table_id: TableId,
-        table_name: &'static str,
-        table_type: table::metadata::TableType,
-        table_provider: PgCatalogTable,
-    ) -> Result<Self> {
-        let arrow_schema = table_provider.schema();
-        let schema = Arc::new(arrow_schema.try_into().context(ProjectSchemaSnafu)?);
-        Ok(Self {
-            table_id,
-            table_name,
-            table_type,
-            schema,
-            table_provider,
-        })
-    }
-}
-
-impl SystemTable for DFTableProviderAsSystemTable {
-    fn table_id(&self) -> TableId {
-        self.table_id
-    }
-
-    fn table_name(&self) -> &'static str {
-        self.table_name
-    }
-
-    fn schema(&self) -> Arc<datatypes::schema::Schema> {
-        self.schema.clone()
-    }
-
-    fn table_type(&self) -> table::metadata::TableType {
-        self.table_type
-    }
-
-    fn to_stream(&self, _request: ScanRequest) -> Result<SendableRecordBatchStream> {
-        match &self.table_provider {
-            PgCatalogTable::Static(table) => {
-                let schema = self.schema.arrow_schema().clone();
-                let data = table
-                    .data()
-                    .iter()
-                    .map(|rb| Ok(rb.clone()))
-                    .collect::<Vec<_>>();
-                let stream = Box::pin(DfRecordBatchStreamAdapter::new(
-                    schema,
-                    futures::stream::iter(data),
-                ));
-                Ok(Box::pin(
-                    RecordBatchStreamAdapter::try_new(stream)
-                        .map_err(BoxedError::new)
-                        .context(InternalSnafu)?,
-                ))
-            }
-
-            PgCatalogTable::Dynamic(table) => {
-                let stream = table.execute(Arc::new(TaskContext::default()));
-                Ok(Box::pin(
-                    RecordBatchStreamAdapter::try_new(stream)
-                        .map_err(BoxedError::new)
-                        .context(InternalSnafu)?,
-                ))
-            }
-
-            PgCatalogTable::Empty(_) => {
-                let schema = self.schema.arrow_schema().clone();
-                let stream = Box::pin(DfRecordBatchStreamAdapter::new(
-                    schema,
-                    futures::stream::iter(vec![]),
-                ));
-                Ok(Box::pin(
-                    RecordBatchStreamAdapter::try_new(stream)
-                        .map_err(BoxedError::new)
-                        .context(InternalSnafu)?,
-                ))
-            }
-        }
-    }
 }

src/catalog/src/system_schema/pg_catalog/pg_catalog_memory_table.rs

@@ -0,0 +1,69 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::sync::Arc;
+
+use datatypes::schema::{ColumnSchema, Schema, SchemaRef};
+use datatypes::vectors::{Int16Vector, StringVector, UInt32Vector, VectorRef};
+
+use crate::memory_table_cols;
+use crate::system_schema::pg_catalog::oid_column;
+use crate::system_schema::pg_catalog::table_names::PG_TYPE;
+use crate::system_schema::utils::tables::{i16_column, string_column};
+
+fn pg_type_schema_columns() -> (Vec<ColumnSchema>, Vec<VectorRef>) {
+    // TODO(j0hn50n133): acquire this information from `DataType` instead of hardcoding it to avoid regression.
+    memory_table_cols!(
+        [oid, typname, typlen],
+        [
+            (1, "String", -1),
+            (2, "Binary", -1),
+            (3, "Int8", 1),
+            (4, "Int16", 2),
+            (5, "Int32", 4),
+            (6, "Int64", 8),
+            (7, "UInt8", 1),
+            (8, "UInt16", 2),
+            (9, "UInt32", 4),
+            (10, "UInt64", 8),
+            (11, "Float32", 4),
+            (12, "Float64", 8),
+            (13, "Decimal", 16),
+            (14, "Date", 4),
+            (15, "DateTime", 8),
+            (16, "Timestamp", 8),
+            (17, "Time", 8),
+            (18, "Duration", 8),
+            (19, "Interval", 16),
+            (20, "List", -1),
+        ]
+    );
+    (
+        // not quiet identical with pg, we only follow the definition in pg
+        vec![oid_column(), string_column("typname"), i16_column("typlen")],
+        vec![
+            Arc::new(UInt32Vector::from_vec(oid)), // oid
+            Arc::new(StringVector::from(typname)),
+            Arc::new(Int16Vector::from_vec(typlen)), // typlen in bytes
+        ],
+    )
+}
+
+pub(super) fn get_schema_columns(table_name: &str) -> (SchemaRef, Vec<VectorRef>) {
+    let (column_schemas, columns): (_, Vec<VectorRef>) = match table_name {
+        PG_TYPE => pg_type_schema_columns(),
+        _ => unreachable!("Unknown table in pg_catalog: {}", table_name),
+    };
+    (Arc::new(Schema::new(column_schemas)), columns)
+}

src/catalog/src/system_schema/pg_catalog/pg_class.rs

@@ -0,0 +1,276 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::fmt;
+use std::sync::{Arc, Weak};
+
+use arrow_schema::SchemaRef as ArrowSchemaRef;
+use common_catalog::consts::PG_CATALOG_PG_CLASS_TABLE_ID;
+use common_error::ext::BoxedError;
+use common_recordbatch::adapter::RecordBatchStreamAdapter;
+use common_recordbatch::{DfSendableRecordBatchStream, RecordBatch};
+use datafusion::execution::TaskContext;
+use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
+use datafusion::physical_plan::streaming::PartitionStream as DfPartitionStream;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::schema::{Schema, SchemaRef};
+use datatypes::value::Value;
+use datatypes::vectors::{StringVectorBuilder, UInt32VectorBuilder, VectorRef};
+use futures::TryStreamExt;
+use snafu::{OptionExt, ResultExt};
+use store_api::storage::ScanRequest;
+use table::metadata::TableType;
+
+use crate::CatalogManager;
+use crate::error::{
+    CreateRecordBatchSnafu, InternalSnafu, Result, UpgradeWeakCatalogManagerRefSnafu,
+};
+use crate::information_schema::Predicates;
+use crate::system_schema::SystemTable;
+use crate::system_schema::pg_catalog::pg_namespace::oid_map::PGNamespaceOidMapRef;
+use crate::system_schema::pg_catalog::{OID_COLUMN_NAME, PG_CLASS, query_ctx};
+use crate::system_schema::utils::tables::{string_column, u32_column};
+
+// === column name ===
+pub const RELNAME: &str = "relname";
+pub const RELNAMESPACE: &str = "relnamespace";
+pub const RELKIND: &str = "relkind";
+pub const RELOWNER: &str = "relowner";
+
+// === enum value of relkind ===
+pub const RELKIND_TABLE: &str = "r";
+pub const RELKIND_VIEW: &str = "v";
+
+/// The initial capacity of the vector builders.
+const INIT_CAPACITY: usize = 42;
+/// The dummy owner id for the namespace.
+const DUMMY_OWNER_ID: u32 = 0;
+
+/// The `pg_catalog.pg_class` table implementation.
+pub(super) struct PGClass {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+
+    // Workaround to convert schema_name to a numeric id
+    namespace_oid_map: PGNamespaceOidMapRef,
+}
+
+impl PGClass {
+    pub(super) fn new(
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        namespace_oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema: Self::schema(),
+            catalog_name,
+            catalog_manager,
+            namespace_oid_map,
+        }
+    }
+
+    fn schema() -> SchemaRef {
+        Arc::new(Schema::new(vec![
+            u32_column(OID_COLUMN_NAME),
+            string_column(RELNAME),
+            u32_column(RELNAMESPACE),
+            string_column(RELKIND),
+            u32_column(RELOWNER),
+        ]))
+    }
+
+    fn builder(&self) -> PGClassBuilder {
+        PGClassBuilder::new(
+            self.schema.clone(),
+            self.catalog_name.clone(),
+            self.catalog_manager.clone(),
+            self.namespace_oid_map.clone(),
+        )
+    }
+}
+
+impl fmt::Debug for PGClass {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("PGClass")
+            .field("schema", &self.schema)
+            .field("catalog_name", &self.catalog_name)
+            .finish()
+    }
+}
+
+impl SystemTable for PGClass {
+    fn table_id(&self) -> table::metadata::TableId {
+        PG_CATALOG_PG_CLASS_TABLE_ID
+    }
+
+    fn table_name(&self) -> &'static str {
+        PG_CLASS
+    }
+
+    fn schema(&self) -> SchemaRef {
+        self.schema.clone()
+    }
+
+    fn to_stream(
+        &self,
+        request: ScanRequest,
+    ) -> Result<common_recordbatch::SendableRecordBatchStream> {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        let stream = Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_class(Some(request))
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ));
+        Ok(Box::pin(
+            RecordBatchStreamAdapter::try_new(stream)
+                .map_err(BoxedError::new)
+                .context(InternalSnafu)?,
+        ))
+    }
+}
+
+impl DfPartitionStream for PGClass {
+    fn schema(&self) -> &ArrowSchemaRef {
+        self.schema.arrow_schema()
+    }
+
+    fn execute(&self, _: Arc<TaskContext>) -> DfSendableRecordBatchStream {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_class(None)
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ))
+    }
+}
+
+/// Builds the `pg_catalog.pg_class` table row by row
+/// TODO(J0HN50N133): `relowner` is always the [`DUMMY_OWNER_ID`] because we don't have users.
+/// Once we have user system, make it the actual owner of the table.
+struct PGClassBuilder {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+    namespace_oid_map: PGNamespaceOidMapRef,
+
+    oid: UInt32VectorBuilder,
+    relname: StringVectorBuilder,
+    relnamespace: UInt32VectorBuilder,
+    relkind: StringVectorBuilder,
+    relowner: UInt32VectorBuilder,
+}
+
+impl PGClassBuilder {
+    fn new(
+        schema: SchemaRef,
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        namespace_oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema,
+            catalog_name,
+            catalog_manager,
+            namespace_oid_map,
+
+            oid: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
+            relname: StringVectorBuilder::with_capacity(INIT_CAPACITY),
+            relnamespace: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
+            relkind: StringVectorBuilder::with_capacity(INIT_CAPACITY),
+            relowner: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
+        }
+    }
+
+    async fn make_class(&mut self, request: Option<ScanRequest>) -> Result<RecordBatch> {
+        let catalog_name = self.catalog_name.clone();
+        let catalog_manager = self
+            .catalog_manager
+            .upgrade()
+            .context(UpgradeWeakCatalogManagerRefSnafu)?;
+        let predicates = Predicates::from_scan_request(&request);
+        for schema_name in catalog_manager
+            .schema_names(&catalog_name, query_ctx())
+            .await?
+        {
+            let mut stream = catalog_manager.tables(&catalog_name, &schema_name, query_ctx());
+            while let Some(table) = stream.try_next().await? {
+                let table_info = table.table_info();
+                self.add_class(
+                    &predicates,
+                    table_info.table_id(),
+                    &schema_name,
+                    &table_info.name,
+                    if table_info.table_type == TableType::View {
+                        RELKIND_VIEW
+                    } else {
+                        RELKIND_TABLE
+                    },
+                );
+            }
+        }
+        self.finish()
+    }
+
+    fn add_class(
+        &mut self,
+        predicates: &Predicates,
+        oid: u32,
+        schema: &str,
+        table: &str,
+        kind: &str,
+    ) {
+        let namespace_oid = self.namespace_oid_map.get_oid(schema);
+        let row = [
+            (OID_COLUMN_NAME, &Value::from(oid)),
+            (RELNAMESPACE, &Value::from(schema)),
+            (RELNAME, &Value::from(table)),
+            (RELKIND, &Value::from(kind)),
+            (RELOWNER, &Value::from(DUMMY_OWNER_ID)),
+        ];
+
+        if !predicates.eval(&row) {
+            return;
+        }
+
+        self.oid.push(Some(oid));
+        self.relnamespace.push(Some(namespace_oid));
+        self.relname.push(Some(table));
+        self.relkind.push(Some(kind));
+        self.relowner.push(Some(DUMMY_OWNER_ID));
+    }
+
+    fn finish(&mut self) -> Result<RecordBatch> {
+        let columns: Vec<VectorRef> = vec![
+            Arc::new(self.oid.finish()),
+            Arc::new(self.relname.finish()),
+            Arc::new(self.relnamespace.finish()),
+            Arc::new(self.relkind.finish()),
+            Arc::new(self.relowner.finish()),
+        ];
+        RecordBatch::new(self.schema.clone(), columns).context(CreateRecordBatchSnafu)
+    }
+}

src/catalog/src/system_schema/pg_catalog/pg_database.rs

@@ -0,0 +1,223 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::sync::{Arc, Weak};
+
+use arrow_schema::SchemaRef as ArrowSchemaRef;
+use common_catalog::consts::PG_CATALOG_PG_DATABASE_TABLE_ID;
+use common_error::ext::BoxedError;
+use common_recordbatch::adapter::RecordBatchStreamAdapter;
+use common_recordbatch::{DfSendableRecordBatchStream, RecordBatch};
+use datafusion::execution::TaskContext;
+use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
+use datafusion::physical_plan::streaming::PartitionStream as DfPartitionStream;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::schema::{Schema, SchemaRef};
+use datatypes::value::Value;
+use datatypes::vectors::{StringVectorBuilder, UInt32VectorBuilder, VectorRef};
+use snafu::{OptionExt, ResultExt};
+use store_api::storage::ScanRequest;
+
+use crate::CatalogManager;
+use crate::error::{
+    CreateRecordBatchSnafu, InternalSnafu, Result, UpgradeWeakCatalogManagerRefSnafu,
+};
+use crate::information_schema::Predicates;
+use crate::system_schema::SystemTable;
+use crate::system_schema::pg_catalog::pg_namespace::oid_map::PGNamespaceOidMapRef;
+use crate::system_schema::pg_catalog::{OID_COLUMN_NAME, PG_DATABASE, query_ctx};
+use crate::system_schema::utils::tables::{string_column, u32_column};
+
+// === column name ===
+pub const DATNAME: &str = "datname";
+
+/// The initial capacity of the vector builders.
+const INIT_CAPACITY: usize = 42;
+
+/// The `pg_catalog.database` table implementation.
+pub(super) struct PGDatabase {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+
+    // Workaround to convert schema_name to a numeric id
+    namespace_oid_map: PGNamespaceOidMapRef,
+}
+
+impl std::fmt::Debug for PGDatabase {
+    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
+        f.debug_struct("PGDatabase")
+            .field("schema", &self.schema)
+            .field("catalog_name", &self.catalog_name)
+            .finish()
+    }
+}
+
+impl PGDatabase {
+    pub(super) fn new(
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        namespace_oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema: Self::schema(),
+            catalog_name,
+            catalog_manager,
+            namespace_oid_map,
+        }
+    }
+
+    fn schema() -> SchemaRef {
+        Arc::new(Schema::new(vec![
+            u32_column(OID_COLUMN_NAME),
+            string_column(DATNAME),
+        ]))
+    }
+
+    fn builder(&self) -> PGCDatabaseBuilder {
+        PGCDatabaseBuilder::new(
+            self.schema.clone(),
+            self.catalog_name.clone(),
+            self.catalog_manager.clone(),
+            self.namespace_oid_map.clone(),
+        )
+    }
+}
+
+impl DfPartitionStream for PGDatabase {
+    fn schema(&self) -> &ArrowSchemaRef {
+        self.schema.arrow_schema()
+    }
+
+    fn execute(&self, _: Arc<TaskContext>) -> DfSendableRecordBatchStream {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_database(None)
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ))
+    }
+}
+
+impl SystemTable for PGDatabase {
+    fn table_id(&self) -> table::metadata::TableId {
+        PG_CATALOG_PG_DATABASE_TABLE_ID
+    }
+
+    fn table_name(&self) -> &'static str {
+        PG_DATABASE
+    }
+
+    fn schema(&self) -> SchemaRef {
+        self.schema.clone()
+    }
+
+    fn to_stream(
+        &self,
+        request: ScanRequest,
+    ) -> Result<common_recordbatch::SendableRecordBatchStream> {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        let stream = Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_database(Some(request))
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ));
+        Ok(Box::pin(
+            RecordBatchStreamAdapter::try_new(stream)
+                .map_err(BoxedError::new)
+                .context(InternalSnafu)?,
+        ))
+    }
+}
+
+/// Builds the `pg_catalog.pg_database` table row by row
+/// `oid` use schema name as a workaround since we don't have numeric schema id.
+/// `nspname` is the schema name.
+struct PGCDatabaseBuilder {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+    namespace_oid_map: PGNamespaceOidMapRef,
+
+    oid: UInt32VectorBuilder,
+    datname: StringVectorBuilder,
+}
+
+impl PGCDatabaseBuilder {
+    fn new(
+        schema: SchemaRef,
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        namespace_oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema,
+            catalog_name,
+            catalog_manager,
+            namespace_oid_map,
+
+            oid: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
+            datname: StringVectorBuilder::with_capacity(INIT_CAPACITY),
+        }
+    }
+
+    async fn make_database(&mut self, request: Option<ScanRequest>) -> Result<RecordBatch> {
+        let catalog_name = self.catalog_name.clone();
+        let catalog_manager = self
+            .catalog_manager
+            .upgrade()
+            .context(UpgradeWeakCatalogManagerRefSnafu)?;
+        let predicates = Predicates::from_scan_request(&request);
+        for schema_name in catalog_manager
+            .schema_names(&catalog_name, query_ctx())
+            .await?
+        {
+            self.add_database(&predicates, &schema_name);
+        }
+        self.finish()
+    }
+
+    fn add_database(&mut self, predicates: &Predicates, schema_name: &str) {
+        let oid = self.namespace_oid_map.get_oid(schema_name);
+        let row: [(&str, &Value); 2] = [
+            (OID_COLUMN_NAME, &Value::from(oid)),
+            (DATNAME, &Value::from(schema_name)),
+        ];
+
+        if !predicates.eval(&row) {
+            return;
+        }
+
+        self.oid.push(Some(oid));
+        self.datname.push(Some(schema_name));
+    }
+
+    fn finish(&mut self) -> Result<RecordBatch> {
+        let columns: Vec<VectorRef> =
+            vec![Arc::new(self.oid.finish()), Arc::new(self.datname.finish())];
+        RecordBatch::new(self.schema.clone(), columns).context(CreateRecordBatchSnafu)
+    }
+}

src/catalog/src/system_schema/pg_catalog/pg_namespace.rs

@@ -0,0 +1,221 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+//! The `pg_catalog.pg_namespace` table implementation.
+//! namespace is a schema in greptime
+
+pub(super) mod oid_map;
+
+use std::fmt;
+use std::sync::{Arc, Weak};
+
+use arrow_schema::SchemaRef as ArrowSchemaRef;
+use common_catalog::consts::PG_CATALOG_PG_NAMESPACE_TABLE_ID;
+use common_error::ext::BoxedError;
+use common_recordbatch::adapter::RecordBatchStreamAdapter;
+use common_recordbatch::{DfSendableRecordBatchStream, RecordBatch, SendableRecordBatchStream};
+use datafusion::execution::TaskContext;
+use datafusion::physical_plan::stream::RecordBatchStreamAdapter as DfRecordBatchStreamAdapter;
+use datafusion::physical_plan::streaming::PartitionStream as DfPartitionStream;
+use datatypes::scalars::ScalarVectorBuilder;
+use datatypes::schema::{Schema, SchemaRef};
+use datatypes::value::Value;
+use datatypes::vectors::{StringVectorBuilder, UInt32VectorBuilder, VectorRef};
+use snafu::{OptionExt, ResultExt};
+use store_api::storage::ScanRequest;
+
+use crate::CatalogManager;
+use crate::error::{
+    CreateRecordBatchSnafu, InternalSnafu, Result, UpgradeWeakCatalogManagerRefSnafu,
+};
+use crate::information_schema::Predicates;
+use crate::system_schema::SystemTable;
+use crate::system_schema::pg_catalog::{
+    OID_COLUMN_NAME, PG_NAMESPACE, PGNamespaceOidMapRef, query_ctx,
+};
+use crate::system_schema::utils::tables::{string_column, u32_column};
+
+const NSPNAME: &str = "nspname";
+const INIT_CAPACITY: usize = 42;
+
+pub(super) struct PGNamespace {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+
+    // Workaround to convert schema_name to a numeric id
+    oid_map: PGNamespaceOidMapRef,
+}
+
+impl PGNamespace {
+    pub(super) fn new(
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema: Self::schema(),
+            catalog_name,
+            catalog_manager,
+            oid_map,
+        }
+    }
+
+    fn schema() -> SchemaRef {
+        Arc::new(Schema::new(vec![
+            // TODO(J0HN50N133): we do not have a numeric schema id, use schema name as a workaround. Use a proper schema id once we have it.
+            u32_column(OID_COLUMN_NAME),
+            string_column(NSPNAME),
+        ]))
+    }
+
+    fn builder(&self) -> PGNamespaceBuilder {
+        PGNamespaceBuilder::new(
+            self.schema.clone(),
+            self.catalog_name.clone(),
+            self.catalog_manager.clone(),
+            self.oid_map.clone(),
+        )
+    }
+}
+
+impl fmt::Debug for PGNamespace {
+    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
+        f.debug_struct("PGNamespace")
+            .field("schema", &self.schema)
+            .field("catalog_name", &self.catalog_name)
+            .finish()
+    }
+}
+
+impl SystemTable for PGNamespace {
+    fn schema(&self) -> SchemaRef {
+        self.schema.clone()
+    }
+
+    fn table_id(&self) -> table::metadata::TableId {
+        PG_CATALOG_PG_NAMESPACE_TABLE_ID
+    }
+
+    fn table_name(&self) -> &'static str {
+        PG_NAMESPACE
+    }
+
+    fn to_stream(&self, request: ScanRequest) -> Result<SendableRecordBatchStream> {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        let stream = Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_namespace(Some(request))
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ));
+        Ok(Box::pin(
+            RecordBatchStreamAdapter::try_new(stream)
+                .map_err(BoxedError::new)
+                .context(InternalSnafu)?,
+        ))
+    }
+}
+
+impl DfPartitionStream for PGNamespace {
+    fn schema(&self) -> &ArrowSchemaRef {
+        self.schema.arrow_schema()
+    }
+
+    fn execute(&self, _: Arc<TaskContext>) -> DfSendableRecordBatchStream {
+        let schema = self.schema.arrow_schema().clone();
+        let mut builder = self.builder();
+        Box::pin(DfRecordBatchStreamAdapter::new(
+            schema,
+            futures::stream::once(async move {
+                builder
+                    .make_namespace(None)
+                    .await
+                    .map(|x| x.into_df_record_batch())
+                    .map_err(Into::into)
+            }),
+        ))
+    }
+}
+
+/// Builds the `pg_catalog.pg_namespace` table row by row
+/// `oid` use schema name as a workaround since we don't have numeric schema id.
+/// `nspname` is the schema name.
+struct PGNamespaceBuilder {
+    schema: SchemaRef,
+    catalog_name: String,
+    catalog_manager: Weak<dyn CatalogManager>,
+    namespace_oid_map: PGNamespaceOidMapRef,
+
+    oid: UInt32VectorBuilder,
+    nspname: StringVectorBuilder,
+}
+
+impl PGNamespaceBuilder {
+    fn new(
+        schema: SchemaRef,
+        catalog_name: String,
+        catalog_manager: Weak<dyn CatalogManager>,
+        namespace_oid_map: PGNamespaceOidMapRef,
+    ) -> Self {
+        Self {
+            schema,
+            catalog_name,
+            catalog_manager,
+            namespace_oid_map,
+            oid: UInt32VectorBuilder::with_capacity(INIT_CAPACITY),
+            nspname: StringVectorBuilder::with_capacity(INIT_CAPACITY),
+        }
+    }
+
+    /// Construct the `pg_catalog.pg_namespace` virtual table
+    async fn make_namespace(&mut self, request: Option<ScanRequest>) -> Result<RecordBatch> {
+        let catalog_name = self.catalog_name.clone();
+        let catalog_manager = self
+            .catalog_manager
+            .upgrade()
+            .context(UpgradeWeakCatalogManagerRefSnafu)?;
+        let predicates = Predicates::from_scan_request(&request);
+        for schema_name in catalog_manager
+            .schema_names(&catalog_name, query_ctx())
+            .await?
+        {
+            self.add_namespace(&predicates, &schema_name);
+        }
+        self.finish()
+    }
+    fn finish(&mut self) -> Result<RecordBatch> {
+        let columns: Vec<VectorRef> =
+            vec![Arc::new(self.oid.finish()), Arc::new(self.nspname.finish())];
+        RecordBatch::new(self.schema.clone(), columns).context(CreateRecordBatchSnafu)
+    }
+
+    fn add_namespace(&mut self, predicates: &Predicates, schema_name: &str) {
+        let oid = self.namespace_oid_map.get_oid(schema_name);
+        let row = [
+            (OID_COLUMN_NAME, &Value::from(oid)),
+            (NSPNAME, &Value::from(schema_name)),
+        ];
+        if !predicates.eval(&row) {
+            return;
+        }
+        self.oid.push(Some(oid));
+        self.nspname.push(Some(schema_name));
+    }
+}

src/catalog/src/system_schema/pg_catalog/pg_namespace/oid_map.rs

@@ -0,0 +1,94 @@
+// Copyright 2023 Greptime Team
+//
+// Licensed under the Apache License, Version 2.0 (the "License");
+// you may not use this file except in compliance with the License.
+// You may obtain a copy of the License at
+//
+//     http://www.apache.org/licenses/LICENSE-2.0
+//
+// Unless required by applicable law or agreed to in writing, software
+// distributed under the License is distributed on an "AS IS" BASIS,
+// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+// See the License for the specific language governing permissions and
+// limitations under the License.
+
+use std::hash::BuildHasher;
+use std::sync::Arc;
+
+use dashmap::DashMap;
+use rustc_hash::FxSeededState;
+
+pub type PGNamespaceOidMapRef = Arc<PGNamespaceOidMap>;
+// Workaround to convert schema_name to a numeric id,
+// remove this when we have numeric schema id in greptime
+pub struct PGNamespaceOidMap {
+    oid_map: DashMap<String, u32>,
+
+    // Rust use SipHasher by default, which provides resistance against DOS attacks.
+    // This will produce different hash value between each greptime instance. This will
+    // cause the sqlness test fail. We need a deterministic hash here to provide
+    // same oid for the same schema name with best effort and DOS attacks aren't concern here.
+    hasher: FxSeededState,
+}
+
+impl PGNamespaceOidMap {
+    pub fn new() -> Self {
+        Self {
+            oid_map: DashMap::new(),
+            hasher: FxSeededState::with_seed(0), // PLEASE DO NOT MODIFY THIS SEED VALUE!!!
+        }
+    }
+
+    fn oid_is_used(&self, oid: u32) -> bool {
+        self.oid_map.iter().any(|e| *e.value() == oid)
+    }
+
+    pub fn get_oid(&self, schema_name: &str) -> u32 {
+        if let Some(oid) = self.oid_map.get(schema_name) {
+            *oid
+        } else {
+            let mut oid = self.hasher.hash_one(schema_name) as u32;
+            while self.oid_is_used(oid) {
+                oid = self.hasher.hash_one(oid) as u32;
+            }
+            self.oid_map.insert(schema_name.to_string(), oid);
+            oid
+        }
+    }
+}
+
+#[cfg(test)]
+mod tests {
+
+    use super::*;
+
+    #[test]
+    fn oid_is_stable() {
+        let oid_map_1 = PGNamespaceOidMap::new();
+        let oid_map_2 = PGNamespaceOidMap::new();
+
+        let schema = "schema";
+        let oid = oid_map_1.get_oid(schema);
+
+        // oid keep stable in the same instance
+        assert_eq!(oid, oid_map_1.get_oid(schema));
+
+        // oid keep stable between different instances
+        assert_eq!(oid, oid_map_2.get_oid(schema));
+    }
+
+    #[test]
+    fn oid_collision() {
+        let oid_map = PGNamespaceOidMap::new();
+
+        let key1 = "3178510";
+        let key2 = "4215648";
+
+        // insert them into oid_map
+        let oid1 = oid_map.get_oid(key1);
+        let oid2 = oid_map.get_oid(key2);
+
+        // they should have different id
+        assert_ne!(oid1, oid2);
+    }
+}

src/catalog/src/system_schema/pg_catalog/table_names.rs

@@ -12,11 +12,11 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-use lazy_static::lazy_static;
+// https://www.postgresql.org/docs/current/catalog-pg-database.html
-use regex::Regex;
+pub const PG_DATABASE: &str = "pg_database";
-
+// https://www.postgresql.org/docs/current/catalog-pg-namespace.html
-pub const NAME_PATTERN: &str = r"[a-zA-Z_:-][a-zA-Z0-9_:\-\.@#]*";
+pub const PG_NAMESPACE: &str = "pg_namespace";
-
+// https://www.postgresql.org/docs/current/catalog-pg-class.html
-lazy_static! {
+pub const PG_CLASS: &str = "pg_class";
-    pub static ref NAME_PATTERN_REG: Regex = Regex::new(&format!("^{NAME_PATTERN}$")).unwrap();
+// https://www.postgresql.org/docs/current/catalog-pg-type.html
-}
+pub const PG_TYPE: &str = "pg_type";

src/catalog/src/system_schema/utils/tables.rs

@@ -27,6 +27,22 @@ pub fn string_column(name: &str) -> ColumnSchema {
     )
 }
 
+pub fn u32_column(name: &str) -> ColumnSchema {
+    ColumnSchema::new(
+        str::to_lowercase(name),
+        ConcreteDataType::uint32_datatype(),
+        false,
+    )
+}
+
+pub fn i16_column(name: &str) -> ColumnSchema {
+    ColumnSchema::new(
+        str::to_lowercase(name),
+        ConcreteDataType::int16_datatype(),
+        false,
+    )
+}
+
 pub fn bigint_column(name: &str) -> ColumnSchema {
     ColumnSchema::new(
         str::to_lowercase(name),

src/catalog/src/table_source.rs

@@ -201,7 +201,7 @@ impl DfTableSourceProvider {
 
         Ok(Arc::new(ViewTable::new(
             logical_plan,
-            Some(view_info.definition.clone()),
+            Some(view_info.definition.to_string()),
         )))
     }
 }

src/cli/Cargo.toml

@@ -51,7 +51,6 @@ meta-srv.workspace = true
 nu-ansi-term = "0.46"
 object-store.workspace = true
 operator.workspace = true
-paste.workspace = true
 query.workspace = true
 rand.workspace = true
 reqwest.workspace = true
@@ -61,6 +60,7 @@ servers.workspace = true
 session.workspace = true
 snafu.workspace = true
 store-api.workspace = true
+substrait.workspace = true
 table.workspace = true
 tokio.workspace = true
 tracing-appender.workspace = true

src/cli/src/bench.rs

@@ -157,7 +157,6 @@ fn create_table_info(table_id: TableId, table_name: TableName) -> RawTableInfo {
         schema: RawSchema::new(column_schemas),
         engine: "mito".to_string(),
         created_on: chrono::DateTime::default(),
-        updated_on: chrono::DateTime::default(),
         primary_key_indices: vec![],
         next_column_id: columns as u32 + 1,
         value_indices: vec![],

src/cli/src/common.rs

@@ -1,19 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-mod object_store;
-mod store;
-
-pub use object_store::{ObjectStoreConfig, new_fs_object_store};
-pub use store::StoreConfig;

src/cli/src/common/object_store.rs

@@ -1,224 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use common_base::secrets::SecretString;
-use common_error::ext::BoxedError;
-use object_store::services::{Azblob, Fs, Gcs, Oss, S3};
-use object_store::util::{with_instrument_layers, with_retry_layers};
-use object_store::{AzblobConnection, GcsConnection, ObjectStore, OssConnection, S3Connection};
-use paste::paste;
-use snafu::ResultExt;
-
-use crate::error::{self};
-
-macro_rules! wrap_with_clap_prefix {
-    (
-        $new_name:ident, $prefix:literal, $base:ty, {
-            $( $( #[doc = $doc:expr] )? $( #[alias = $alias:literal] )? $field:ident : $type:ty $( = $default:expr )? ),* $(,)?
-        }
-    ) => {
-        paste!{
-            #[derive(clap::Parser, Debug, Clone, PartialEq, Default)]
-            pub struct $new_name {
-                $(
-                    $( #[doc = $doc] )?
-                    $( #[clap(alias = $alias)] )?
-                    #[clap(long $(, default_value_t = $default )? )]
-                    [<$prefix $field>]: $type,
-                )*
-            }
-
-            impl From<$new_name> for $base {
-                fn from(w: $new_name) -> Self {
-                    Self {
-                        $( $field: w.[<$prefix $field>] ),*
-                    }
-                }
-            }
-        }
-    };
-}
-
-wrap_with_clap_prefix! {
-    PrefixedAzblobConnection,
-    "azblob-",
-    AzblobConnection,
-    {
-        #[doc = "The container of the object store."]
-        container: String = Default::default(),
-        #[doc = "The root of the object store."]
-        root: String = Default::default(),
-        #[doc = "The account name of the object store."]
-        account_name: SecretString = Default::default(),
-        #[doc = "The account key of the object store."]
-        account_key: SecretString = Default::default(),
-        #[doc = "The endpoint of the object store."]
-        endpoint: String = Default::default(),
-        #[doc = "The SAS token of the object store."]
-        sas_token: Option<String>,
-    }
-}
-
-wrap_with_clap_prefix! {
-    PrefixedS3Connection,
-    "s3-",
-    S3Connection,
-    {
-        #[doc = "The bucket of the object store."]
-        bucket: String = Default::default(),
-        #[doc = "The root of the object store."]
-        root: String = Default::default(),
-        #[doc = "The access key ID of the object store."]
-        access_key_id: SecretString = Default::default(),
-        #[doc = "The secret access key of the object store."]
-        secret_access_key: SecretString = Default::default(),
-        #[doc = "The endpoint of the object store."]
-        endpoint: Option<String>,
-        #[doc = "The region of the object store."]
-        region: Option<String>,
-        #[doc = "Enable virtual host style for the object store."]
-        enable_virtual_host_style: bool = Default::default(),
-    }
-}
-
-wrap_with_clap_prefix! {
-    PrefixedOssConnection,
-    "oss-",
-    OssConnection,
-    {
-        #[doc = "The bucket of the object store."]
-        bucket: String = Default::default(),
-        #[doc = "The root of the object store."]
-        root: String = Default::default(),
-        #[doc = "The access key ID of the object store."]
-        access_key_id: SecretString = Default::default(),
-        #[doc = "The access key secret of the object store."]
-        access_key_secret: SecretString = Default::default(),
-        #[doc = "The endpoint of the object store."]
-        endpoint: String = Default::default(),
-    }
-}
-
-wrap_with_clap_prefix! {
-    PrefixedGcsConnection,
-    "gcs-",
-    GcsConnection,
-    {
-        #[doc = "The root of the object store."]
-        root: String = Default::default(),
-        #[doc = "The bucket of the object store."]
-        bucket: String = Default::default(),
-        #[doc = "The scope of the object store."]
-        scope: String = Default::default(),
-        #[doc = "The credential path of the object store."]
-        credential_path: SecretString = Default::default(),
-        #[doc = "The credential of the object store."]
-        credential: SecretString = Default::default(),
-        #[doc = "The endpoint of the object store."]
-        endpoint: String = Default::default(),
-    }
-}
-
-/// common config for object store.
-#[derive(clap::Parser, Debug, Clone, PartialEq, Default)]
-pub struct ObjectStoreConfig {
-    /// Whether to use S3 object store.
-    #[clap(long, alias = "s3")]
-    pub enable_s3: bool,
-
-    #[clap(flatten)]
-    pub s3: PrefixedS3Connection,
-
-    /// Whether to use OSS.
-    #[clap(long, alias = "oss")]
-    pub enable_oss: bool,
-
-    #[clap(flatten)]
-    pub oss: PrefixedOssConnection,
-
-    /// Whether to use GCS.
-    #[clap(long, alias = "gcs")]
-    pub enable_gcs: bool,
-
-    #[clap(flatten)]
-    pub gcs: PrefixedGcsConnection,
-
-    /// Whether to use Azure Blob.
-    #[clap(long, alias = "azblob")]
-    pub enable_azblob: bool,
-
-    #[clap(flatten)]
-    pub azblob: PrefixedAzblobConnection,
-}
-
-/// Creates a new file system object store.
-pub fn new_fs_object_store(root: &str) -> std::result::Result<ObjectStore, BoxedError> {
-    let builder = Fs::default().root(root);
-    let object_store = ObjectStore::new(builder)
-        .context(error::InitBackendSnafu)
-        .map_err(BoxedError::new)?
-        .finish();
-
-    Ok(with_instrument_layers(object_store, false))
-}
-
-impl ObjectStoreConfig {
-    /// Builds the object store from the config.
-    pub fn build(&self) -> Result<Option<ObjectStore>, BoxedError> {
-        let object_store = if self.enable_s3 {
-            let s3 = S3Connection::from(self.s3.clone());
-            common_telemetry::info!("Building object store with s3: {:?}", s3);
-            Some(
-                ObjectStore::new(S3::from(&s3))
-                    .context(error::InitBackendSnafu)
-                    .map_err(BoxedError::new)?
-                    .finish(),
-            )
-        } else if self.enable_oss {
-            let oss = OssConnection::from(self.oss.clone());
-            common_telemetry::info!("Building object store with oss: {:?}", oss);
-            Some(
-                ObjectStore::new(Oss::from(&oss))
-                    .context(error::InitBackendSnafu)
-                    .map_err(BoxedError::new)?
-                    .finish(),
-            )
-        } else if self.enable_gcs {
-            let gcs = GcsConnection::from(self.gcs.clone());
-            common_telemetry::info!("Building object store with gcs: {:?}", gcs);
-            Some(
-                ObjectStore::new(Gcs::from(&gcs))
-                    .context(error::InitBackendSnafu)
-                    .map_err(BoxedError::new)?
-                    .finish(),
-            )
-        } else if self.enable_azblob {
-            let azblob = AzblobConnection::from(self.azblob.clone());
-            common_telemetry::info!("Building object store with azblob: {:?}", azblob);
-            Some(
-                ObjectStore::new(Azblob::from(&azblob))
-                    .context(error::InitBackendSnafu)
-                    .map_err(BoxedError::new)?
-                    .finish(),
-            )
-        } else {
-            None
-        };
-
-        let object_store = object_store
-            .map(|object_store| with_instrument_layers(with_retry_layers(object_store), false));
-
-        Ok(object_store)
-    }
-}

src/cli/src/data.rs

@@ -16,7 +16,6 @@ mod export;
 mod import;
 
 use clap::Subcommand;
-use client::DEFAULT_CATALOG_NAME;
 use common_error::ext::BoxedError;
 
 use crate::Tool;
@@ -38,7 +37,3 @@ impl DataCommand {
         }
     }
 }
-
-pub(crate) fn default_database() -> String {
-    format!("{DEFAULT_CATALOG_NAME}-*")
-}

src/cli/src/data/export.rs

@@ -30,7 +30,6 @@ use snafu::{OptionExt, ResultExt};
 use tokio::sync::Semaphore;
 use tokio::time::Instant;
 
-use crate::data::default_database;
 use crate::database::{DatabaseClient, parse_proxy_opts};
 use crate::error::{
     EmptyResultSnafu, Error, OpenDalSnafu, OutputDirNotSetSnafu, Result, S3ConfigNotSetSnafu,
@@ -64,7 +63,7 @@ pub struct ExportCommand {
     output_dir: Option<String>,
 
     /// The name of the catalog to export.
-    #[clap(long, default_value_t = default_database())]
+    #[clap(long, default_value = "greptime-*")]
     database: String,
 
     /// Parallelism of the export.

src/cli/src/data/import.rs

@@ -25,7 +25,6 @@ use snafu::{OptionExt, ResultExt};
 use tokio::sync::Semaphore;
 use tokio::time::Instant;
 
-use crate::data::default_database;
 use crate::database::{DatabaseClient, parse_proxy_opts};
 use crate::error::{Error, FileIoSnafu, Result, SchemaNotFoundSnafu};
 use crate::{Tool, database};
@@ -53,7 +52,7 @@ pub struct ImportCommand {
     input_dir: String,
 
     /// The name of the catalog to import.
-    #[clap(long, default_value_t = default_database())]
+    #[clap(long, default_value = "greptime-*")]
     database: String,
 
     /// Parallelism of the import.

src/cli/src/error.rs

@@ -313,14 +313,6 @@ pub enum Error {
         location: Location,
         source: common_meta::error::Error,
     },
-
-    #[snafu(display("Failed to get current directory"))]
-    GetCurrentDir {
-        #[snafu(implicit)]
-        location: Location,
-        #[snafu(source)]
-        error: std::io::Error,
-    },
 }
 
 pub type Result<T> = std::result::Result<T, Error>;
@@ -370,9 +362,7 @@ impl ErrorExt for Error {
 
             Error::BuildRuntime { source, .. } => source.status_code(),
 
-            Error::CacheRequired { .. }
+            Error::CacheRequired { .. } | Error::BuildCacheRegistry { .. } => StatusCode::Internal,
-            | Error::BuildCacheRegistry { .. }
-            | Error::GetCurrentDir { .. } => StatusCode::Internal,
             Error::MetaClientInit { source, .. } => source.status_code(),
             Error::TableNotFound { .. } => StatusCode::TableNotFound,
             Error::SchemaNotFound { .. } => StatusCode::DatabaseNotFound,

src/cli/src/lib.rs

@@ -14,16 +14,13 @@
 
 #![allow(clippy::print_stdout)]
 mod bench;
-mod common;
 mod data;
 mod database;
 pub mod error;
 mod metadata;
-pub mod utils;
 
 use async_trait::async_trait;
 use clap::Parser;
-pub use common::{ObjectStoreConfig, StoreConfig};
 use common_error::ext::BoxedError;
 pub use database::DatabaseClient;
 use error::Result;

src/cli/src/metadata.rs

@@ -12,6 +12,7 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+mod common;
 mod control;
 mod repair;
 mod snapshot;

src/cli/src/metadata/common.rs

@@ -19,14 +19,14 @@ use common_error::ext::BoxedError;
 use common_meta::kv_backend::KvBackendRef;
 use common_meta::kv_backend::chroot::ChrootKvBackend;
 use common_meta::kv_backend::etcd::EtcdStore;
+use meta_srv::bootstrap::create_etcd_client_with_tls;
 use meta_srv::metasrv::BackendImpl;
-use meta_srv::utils::etcd::create_etcd_client_with_tls;
 use servers::tls::{TlsMode, TlsOption};
 
-use crate::error::EmptyStoreAddrsSnafu;
+use crate::error::{EmptyStoreAddrsSnafu, UnsupportedMemoryBackendSnafu};
 
 #[derive(Debug, Default, Parser)]
-pub struct StoreConfig {
+pub(crate) struct StoreConfig {
     /// The endpoint of store. one of etcd, postgres or mysql.
     ///
     /// For postgres store, the format is:
@@ -38,65 +38,51 @@ pub struct StoreConfig {
     /// For mysql store, the format is:
     /// "mysql://user:password@ip:port/dbname"
     #[clap(long, alias = "store-addr", value_delimiter = ',', num_args = 1..)]
-    pub store_addrs: Vec<String>,
+    store_addrs: Vec<String>,
 
     /// The maximum number of operations in a transaction. Only used when using [etcd-store].
     #[clap(long, default_value = "128")]
-    pub max_txn_ops: usize,
+    max_txn_ops: usize,
 
     /// The metadata store backend.
     #[clap(long, value_enum, default_value = "etcd-store")]
-    pub backend: BackendImpl,
+    backend: BackendImpl,
 
     /// The key prefix of the metadata store.
     #[clap(long, default_value = "")]
-    pub store_key_prefix: String,
+    store_key_prefix: String,
 
     /// The table name in RDS to store metadata. Only used when using [postgres-store] or [mysql-store].
     #[cfg(any(feature = "pg_kvbackend", feature = "mysql_kvbackend"))]
     #[clap(long, default_value = common_meta::kv_backend::DEFAULT_META_TABLE_NAME)]
-    pub meta_table_name: String,
+    meta_table_name: String,
 
     /// Optional PostgreSQL schema for metadata table (defaults to current search_path if unset).
     #[cfg(feature = "pg_kvbackend")]
     #[clap(long)]
-    pub meta_schema_name: Option<String>,
+    meta_schema_name: Option<String>,
     /// TLS mode for backend store connections (etcd, PostgreSQL, MySQL)
     #[clap(long = "backend-tls-mode", value_enum, default_value = "disable")]
-    pub backend_tls_mode: TlsMode,
+    backend_tls_mode: TlsMode,
 
     /// Path to TLS certificate file for backend store connections
     #[clap(long = "backend-tls-cert-path", default_value = "")]
-    pub backend_tls_cert_path: String,
+    backend_tls_cert_path: String,
 
     /// Path to TLS private key file for backend store connections
     #[clap(long = "backend-tls-key-path", default_value = "")]
-    pub backend_tls_key_path: String,
+    backend_tls_key_path: String,
 
     /// Path to TLS CA certificate file for backend store connections
     #[clap(long = "backend-tls-ca-cert-path", default_value = "")]
-    pub backend_tls_ca_cert_path: String,
+    backend_tls_ca_cert_path: String,
 
     /// Enable watching TLS certificate files for changes
     #[clap(long = "backend-tls-watch")]
-    pub backend_tls_watch: bool,
+    backend_tls_watch: bool,
 }
 
 impl StoreConfig {
-    pub fn tls_config(&self) -> Option<TlsOption> {
-        if self.backend_tls_mode != TlsMode::Disable {
-            Some(TlsOption {
-                mode: self.backend_tls_mode.clone(),
-                cert_path: self.backend_tls_cert_path.clone(),
-                key_path: self.backend_tls_key_path.clone(),
-                ca_cert_path: self.backend_tls_ca_cert_path.clone(),
-                watch: self.backend_tls_watch,
-            })
-        } else {
-            None
-        }
-    }
-
     /// Builds a [`KvBackendRef`] from the store configuration.
     pub async fn build(&self) -> Result<KvBackendRef, BoxedError> {
         let max_txn_ops = self.max_txn_ops;
@@ -104,14 +90,19 @@ impl StoreConfig {
         if store_addrs.is_empty() {
             EmptyStoreAddrsSnafu.fail().map_err(BoxedError::new)
         } else {
-            common_telemetry::info!(
-                "Building kvbackend with store addrs: {:?}, backend: {:?}",
-                store_addrs,
-                self.backend
-            );
             let kvbackend = match self.backend {
                 BackendImpl::EtcdStore => {
-                    let tls_config = self.tls_config();
+                    let tls_config = if self.backend_tls_mode != TlsMode::Disable {
+                        Some(TlsOption {
+                            mode: self.backend_tls_mode.clone(),
+                            cert_path: self.backend_tls_cert_path.clone(),
+                            key_path: self.backend_tls_key_path.clone(),
+                            ca_cert_path: self.backend_tls_ca_cert_path.clone(),
+                            watch: self.backend_tls_watch,
+                        })
+                    } else {
+                        None
+                    };
                     let etcd_client = create_etcd_client_with_tls(store_addrs, tls_config.as_ref())
                         .await
                         .map_err(BoxedError::new)?;
@@ -120,14 +111,9 @@ impl StoreConfig {
                 #[cfg(feature = "pg_kvbackend")]
                 BackendImpl::PostgresStore => {
                     let table_name = &self.meta_table_name;
-                    let tls_config = self.tls_config();
+                    let pool = meta_srv::bootstrap::create_postgres_pool(store_addrs, None)
-                    let pool = meta_srv::utils::postgres::create_postgres_pool(
+                        .await
-                        store_addrs,
+                        .map_err(BoxedError::new)?;
-                        None,
-                        tls_config,
-                    )
-                    .await
-                    .map_err(BoxedError::new)?;
                     let schema_name = self.meta_schema_name.as_deref();
                     Ok(common_meta::kv_backend::rds::PgStore::with_pg_pool(
                         pool,
@@ -141,11 +127,9 @@ impl StoreConfig {
                 #[cfg(feature = "mysql_kvbackend")]
                 BackendImpl::MysqlStore => {
                     let table_name = &self.meta_table_name;
-                    let tls_config = self.tls_config();
+                    let pool = meta_srv::bootstrap::create_mysql_pool(store_addrs)
-                    let pool =
+                        .await
-                        meta_srv::utils::mysql::create_mysql_pool(store_addrs, tls_config.as_ref())
+                        .map_err(BoxedError::new)?;
-                            .await
-                            .map_err(BoxedError::new)?;
                     Ok(common_meta::kv_backend::rds::MySqlStore::with_mysql_pool(
                         pool,
                         table_name,
@@ -154,20 +138,9 @@ impl StoreConfig {
                     .await
                     .map_err(BoxedError::new)?)
                 }
-                #[cfg(not(test))]
+                BackendImpl::MemoryStore => UnsupportedMemoryBackendSnafu
-                BackendImpl::MemoryStore => {
+                    .fail()
-                    use crate::error::UnsupportedMemoryBackendSnafu;
+                    .map_err(BoxedError::new),
-
-                    UnsupportedMemoryBackendSnafu
-                        .fail()
-                        .map_err(BoxedError::new)
-                }
-                #[cfg(test)]
-                BackendImpl::MemoryStore => {
-                    use common_meta::kv_backend::memory::MemoryKvBackend;
-
-                    Ok(Arc::new(MemoryKvBackend::default()) as _)
-                }
             };
             if self.store_key_prefix.is_empty() {
                 kvbackend

src/cli/src/metadata/control/del/key.rs

@@ -20,7 +20,7 @@ use common_meta::kv_backend::KvBackendRef;
 use common_meta::rpc::store::RangeRequest;
 
 use crate::Tool;
-use crate::common::StoreConfig;
+use crate::metadata::common::StoreConfig;
 use crate::metadata::control::del::CLI_TOMBSTONE_PREFIX;
 
 /// Delete key-value pairs logically from the metadata store.
@@ -41,7 +41,7 @@ impl DelKeyCommand {
     pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
         let kv_backend = self.store.build().await?;
         Ok(Box::new(DelKeyTool {
-            key: self.key.clone(),
+            key: self.key.to_string(),
             prefix: self.prefix,
             key_deleter: KeyDeleter::new(kv_backend),
         }))

src/cli/src/metadata/control/del/table.rs

@@ -24,8 +24,8 @@ use common_meta::kv_backend::KvBackendRef;
 use store_api::storage::TableId;
 
 use crate::Tool;
-use crate::common::StoreConfig;
 use crate::error::{InvalidArgumentsSnafu, TableNotFoundSnafu};
+use crate::metadata::common::StoreConfig;
 use crate::metadata::control::del::CLI_TOMBSTONE_PREFIX;
 use crate::metadata::control::utils::get_table_id_by_name;
 
@@ -48,7 +48,6 @@ pub struct DelTableCommand {
     #[clap(long, default_value = DEFAULT_CATALOG_NAME)]
     catalog_name: String,
 
-    /// The store config.
     #[clap(flatten)]
     store: StoreConfig,
 }

src/cli/src/metadata/control/get.rs

@@ -28,9 +28,9 @@ use common_meta::rpc::store::RangeRequest;
 use futures::TryStreamExt;
 
 use crate::Tool;
-use crate::common::StoreConfig;
 use crate::error::InvalidArgumentsSnafu;
-use crate::metadata::control::utils::{decode_key_value, get_table_id_by_name, json_formatter};
+use crate::metadata::common::StoreConfig;
+use crate::metadata::control::utils::{decode_key_value, get_table_id_by_name, json_fromatter};
 
 /// Getting metadata from metadata store.
 #[derive(Subcommand)]
@@ -206,7 +206,7 @@ impl Tool for GetTableTool {
             println!(
                 "{}\n{}",
                 TableInfoKey::new(table_id),
-                json_formatter(self.pretty, &*table_info)
+                json_fromatter(self.pretty, &*table_info)
             );
         } else {
             println!("Table info not found");
@@ -221,7 +221,7 @@ impl Tool for GetTableTool {
             println!(
                 "{}\n{}",
                 TableRouteKey::new(table_id),
-                json_formatter(self.pretty, &table_route)
+                json_fromatter(self.pretty, &table_route)
             );
         } else {
             println!("Table route not found");

src/cli/src/metadata/control/utils.rs

@@ -27,7 +27,7 @@ pub fn decode_key_value(kv: KeyValue) -> CommonMetaResult<(String, String)> {
 }
 
 /// Formats a value as a JSON string.
-pub fn json_formatter<T>(pretty: bool, value: &T) -> String
+pub fn json_fromatter<T>(pretty: bool, value: &T) -> String
 where
     T: Serialize,
 {

src/cli/src/metadata/repair.rs

@@ -38,10 +38,10 @@ use snafu::{ResultExt, ensure};
 use store_api::storage::TableId;
 
 use crate::Tool;
-use crate::common::StoreConfig;
 use crate::error::{
     InvalidArgumentsSnafu, Result, SendRequestToDatanodeSnafu, TableMetadataSnafu, UnexpectedSnafu,
 };
+use crate::metadata::common::StoreConfig;
 use crate::metadata::utils::{FullTableMetadata, IteratorInput, TableMetadataIterator};
 
 /// Repair metadata of logical tables.
@@ -138,7 +138,13 @@ impl RepairTool {
 
             let table_names = table_names
                 .iter()
-                .map(|table_name| (catalog.clone(), schema_name.clone(), table_name.clone()))
+                .map(|table_name| {
+                    (
+                        catalog.to_string(),
+                        schema_name.to_string(),
+                        table_name.to_string(),
+                    )
+                })
                 .collect::<Vec<_>>();
             return Ok(IteratorInput::new_table_names(table_names));
         } else if !self.table_ids.is_empty() {

src/cli/src/metadata/repair/alter_table.rs

@@ -32,9 +32,9 @@ pub fn generate_alter_table_expr_for_all_columns(
     let schema = &table_info.meta.schema;
 
     let mut alter_table_expr = AlterTableExpr {
-        catalog_name: table_info.catalog_name.clone(),
+        catalog_name: table_info.catalog_name.to_string(),
-        schema_name: table_info.schema_name.clone(),
+        schema_name: table_info.schema_name.to_string(),
-        table_name: table_info.name.clone(),
+        table_name: table_info.name.to_string(),
         ..Default::default()
     };
 

src/cli/src/metadata/repair/create_table.rs

@@ -44,9 +44,9 @@ pub fn generate_create_table_expr(table_info: &RawTableInfo) -> Result<CreateTab
     let table_options = HashMap::from(&table_info.meta.options);
 
     Ok(CreateTableExpr {
-        catalog_name: table_info.catalog_name.clone(),
+        catalog_name: table_info.catalog_name.to_string(),
-        schema_name: table_info.schema_name.clone(),
+        schema_name: table_info.schema_name.to_string(),
-        table_name: table_info.name.clone(),
+        table_name: table_info.name.to_string(),
         desc: String::default(),
         column_defs,
         time_index,
@@ -54,7 +54,7 @@ pub fn generate_create_table_expr(table_info: &RawTableInfo) -> Result<CreateTab
         create_if_not_exists: true,
         table_options,
         table_id: None,
-        engine: table_info.meta.engine.clone(),
+        engine: table_info.meta.engine.to_string(),
     })
 }
 

src/cli/src/metadata/snapshot.rs

@@ -12,15 +12,20 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
+use std::path::Path;
+
 use async_trait::async_trait;
 use clap::{Parser, Subcommand};
+use common_base::secrets::{ExposeSecret, SecretString};
 use common_error::ext::BoxedError;
 use common_meta::snapshot::MetadataSnapshotManager;
-use object_store::{ObjectStore, Scheme};
+use object_store::ObjectStore;
+use object_store::services::{Fs, S3};
+use snafu::{OptionExt, ResultExt};
 
 use crate::Tool;
-use crate::common::{ObjectStoreConfig, StoreConfig, new_fs_object_store};
+use crate::error::{InvalidFilePathSnafu, OpenDalSnafu, S3ConfigNotSetSnafu};
-use crate::utils::resolve_relative_path_with_current_dir;
+use crate::metadata::common::StoreConfig;
 
 /// Subcommand for metadata snapshot operations, including saving snapshots, restoring from snapshots, and viewing snapshot information.
 #[derive(Subcommand)]
@@ -36,9 +41,68 @@ pub enum SnapshotCommand {
 impl SnapshotCommand {
     pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
         match self {
-            SnapshotCommand::Save(cmd) => Ok(Box::new(cmd.build().await?)),
+            SnapshotCommand::Save(cmd) => cmd.build().await,
-            SnapshotCommand::Restore(cmd) => Ok(Box::new(cmd.build().await?)),
+            SnapshotCommand::Restore(cmd) => cmd.build().await,
-            SnapshotCommand::Info(cmd) => Ok(Box::new(cmd.build().await?)),
+            SnapshotCommand::Info(cmd) => cmd.build().await,
+        }
+    }
+}
+
+// TODO(qtang): Abstract a generic s3 config for export import meta snapshot restore
+#[derive(Debug, Default, Parser)]
+struct S3Config {
+    /// whether to use s3 as the output directory. default is false.
+    #[clap(long, default_value = "false")]
+    s3: bool,
+    /// The s3 bucket name.
+    #[clap(long)]
+    s3_bucket: Option<String>,
+    /// The s3 region.
+    #[clap(long)]
+    s3_region: Option<String>,
+    /// The s3 access key.
+    #[clap(long)]
+    s3_access_key: Option<SecretString>,
+    /// The s3 secret key.
+    #[clap(long)]
+    s3_secret_key: Option<SecretString>,
+    /// The s3 endpoint. we will automatically use the default s3 decided by the region if not set.
+    #[clap(long)]
+    s3_endpoint: Option<String>,
+}
+
+impl S3Config {
+    pub fn build(&self, root: &str) -> Result<Option<ObjectStore>, BoxedError> {
+        if !self.s3 {
+            Ok(None)
+        } else {
+            if self.s3_region.is_none()
+                || self.s3_access_key.is_none()
+                || self.s3_secret_key.is_none()
+                || self.s3_bucket.is_none()
+            {
+                return S3ConfigNotSetSnafu.fail().map_err(BoxedError::new);
+            }
+            // Safety, unwrap is safe because we have checked the options above.
+            let mut config = S3::default()
+                .bucket(self.s3_bucket.as_ref().unwrap())
+                .region(self.s3_region.as_ref().unwrap())
+                .access_key_id(self.s3_access_key.as_ref().unwrap().expose_secret())
+                .secret_access_key(self.s3_secret_key.as_ref().unwrap().expose_secret());
+
+            if !root.is_empty() && root != "." {
+                config = config.root(root);
+            }
+
+            if let Some(endpoint) = &self.s3_endpoint {
+                config = config.endpoint(endpoint);
+            }
+            Ok(Some(
+                ObjectStore::new(config)
+                    .context(OpenDalSnafu)
+                    .map_err(BoxedError::new)?
+                    .finish(),
+            ))
         }
     }
 }
@@ -52,47 +116,60 @@ pub struct SaveCommand {
     /// The store configuration.
     #[clap(flatten)]
     store: StoreConfig,
-    /// The object store configuration.
+    /// The s3 config.
     #[clap(flatten)]
-    object_store: ObjectStoreConfig,
+    s3_config: S3Config,
-    /// The path of the target snapshot file.
+    /// The name of the target snapshot file. we will add the file extension automatically.
-    #[clap(
+    #[clap(long, default_value = "metadata_snapshot")]
-        long,
+    file_name: String,
-        default_value = "metadata_snapshot.metadata.fb",
+    /// The directory to store the snapshot file.
-        alias = "file_name"
+    /// if target output is s3 bucket, this is the root directory in the bucket.
-    )]
+    /// if target output is local file, this is the local directory.
-    file_path: String,
+    #[clap(long, default_value = "")]
-    /// Specifies the root directory used for I/O operations.
+    output_dir: String,
-    #[clap(long, default_value = "/", alias = "output_dir")]
+}
-    dir: String,
+
+fn create_local_file_object_store(root: &str) -> Result<ObjectStore, BoxedError> {
+    let root = if root.is_empty() { "." } else { root };
+    let object_store = ObjectStore::new(Fs::default().root(root))
+        .context(OpenDalSnafu)
+        .map_err(BoxedError::new)?
+        .finish();
+    Ok(object_store)
 }
 
 impl SaveCommand {
-    async fn build(&self) -> Result<MetaSnapshotTool, BoxedError> {
+    pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
         let kvbackend = self.store.build().await?;
-        let (object_store, file_path) = build_object_store_and_resolve_file_path(
+        let output_dir = &self.output_dir;
-            self.object_store.clone(),
+        let object_store = self.s3_config.build(output_dir).map_err(BoxedError::new)?;
-            &self.dir,
+        if let Some(store) = object_store {
-            &self.file_path,
+            let tool = MetaSnapshotTool {
-        )?;
+                inner: MetadataSnapshotManager::new(kvbackend, store),
-        let tool = MetaSnapshotTool {
+                target_file: self.file_name.clone(),
-            inner: MetadataSnapshotManager::new(kvbackend, object_store),
+            };
-            file_path,
+            Ok(Box::new(tool))
-        };
+        } else {
-        Ok(tool)
+            let object_store = create_local_file_object_store(output_dir)?;
+            let tool = MetaSnapshotTool {
+                inner: MetadataSnapshotManager::new(kvbackend, object_store),
+                target_file: self.file_name.clone(),
+            };
+            Ok(Box::new(tool))
+        }
     }
 }
 
 struct MetaSnapshotTool {
     inner: MetadataSnapshotManager,
-    file_path: String,
+    target_file: String,
 }
 
 #[async_trait]
 impl Tool for MetaSnapshotTool {
     async fn do_work(&self) -> std::result::Result<(), BoxedError> {
         self.inner
-            .dump(&self.file_path)
+            .dump("", &self.target_file)
             .await
             .map_err(BoxedError::new)?;
         Ok(())
@@ -109,52 +186,54 @@ pub struct RestoreCommand {
     /// The store configuration.
     #[clap(flatten)]
     store: StoreConfig,
-    /// The object store config.
+    /// The s3 config.
     #[clap(flatten)]
-    object_store: ObjectStoreConfig,
+    s3_config: S3Config,
-    /// The path of the target snapshot file.
+    /// The name of the target snapshot file.
-    #[clap(
+    #[clap(long, default_value = "metadata_snapshot.metadata.fb")]
-        long,
+    file_name: String,
-        default_value = "metadata_snapshot.metadata.fb",
+    /// The directory to store the snapshot file.
-        alias = "file_name"
+    #[clap(long, default_value = ".")]
-    )]
+    input_dir: String,
-    file_path: String,
-    /// Specifies the root directory used for I/O operations.
-    #[clap(long, default_value = "/", alias = "input_dir")]
-    dir: String,
     #[clap(long, default_value = "false")]
     force: bool,
 }
 
 impl RestoreCommand {
-    async fn build(&self) -> Result<MetaRestoreTool, BoxedError> {
+    pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
         let kvbackend = self.store.build().await?;
-        let (object_store, file_path) = build_object_store_and_resolve_file_path(
+        let input_dir = &self.input_dir;
-            self.object_store.clone(),
+        let object_store = self.s3_config.build(input_dir).map_err(BoxedError::new)?;
-            &self.dir,
+        if let Some(store) = object_store {
-            &self.file_path,
+            let tool = MetaRestoreTool::new(
-        )
+                MetadataSnapshotManager::new(kvbackend, store),
-        .map_err(BoxedError::new)?;
+                self.file_name.clone(),
-        let tool = MetaRestoreTool::new(
+                self.force,
-            MetadataSnapshotManager::new(kvbackend, object_store),
+            );
-            file_path,
+            Ok(Box::new(tool))
-            self.force,
+        } else {
-        );
+            let object_store = create_local_file_object_store(input_dir)?;
-        Ok(tool)
+            let tool = MetaRestoreTool::new(
+                MetadataSnapshotManager::new(kvbackend, object_store),
+                self.file_name.clone(),
+                self.force,
+            );
+            Ok(Box::new(tool))
+        }
     }
 }
 
 struct MetaRestoreTool {
     inner: MetadataSnapshotManager,
-    file_path: String,
+    source_file: String,
     force: bool,
 }
 
 impl MetaRestoreTool {
-    pub fn new(inner: MetadataSnapshotManager, file_path: String, force: bool) -> Self {
+    pub fn new(inner: MetadataSnapshotManager, source_file: String, force: bool) -> Self {
         Self {
             inner,
-            file_path,
+            source_file,
             force,
         }
     }
@@ -173,7 +252,7 @@ impl Tool for MetaRestoreTool {
                 "The target source is clean, we will restore the metadata snapshot."
             );
             self.inner
-                .restore(&self.file_path)
+                .restore(&self.source_file)
                 .await
                 .map_err(BoxedError::new)?;
             Ok(())
@@ -187,7 +266,7 @@ impl Tool for MetaRestoreTool {
                 "The target source is not clean, We will restore the metadata snapshot with --force."
             );
             self.inner
-                .restore(&self.file_path)
+                .restore(&self.source_file)
                 .await
                 .map_err(BoxedError::new)?;
             Ok(())
@@ -201,19 +280,12 @@ impl Tool for MetaRestoreTool {
 /// It prints the filtered metadata to the console.
 #[derive(Debug, Default, Parser)]
 pub struct InfoCommand {
-    /// The object store config.
+    /// The s3 config.
     #[clap(flatten)]
-    object_store: ObjectStoreConfig,
+    s3_config: S3Config,
-    /// The path of the target snapshot file.
+    /// The name of the target snapshot file. we will add the file extension automatically.
-    #[clap(
+    #[clap(long, default_value = "metadata_snapshot")]
-        long,
+    file_name: String,
-        default_value = "metadata_snapshot.metadata.fb",
-        alias = "file_name"
-    )]
-    file_path: String,
-    /// Specifies the root directory used for I/O operations.
-    #[clap(long, default_value = "/", alias = "input_dir")]
-    dir: String,
     /// The query string to filter the metadata.
     #[clap(long, default_value = "*")]
     inspect_key: String,
@@ -224,7 +296,7 @@ pub struct InfoCommand {
 
 struct MetaInfoTool {
     inner: ObjectStore,
-    file_path: String,
+    source_file: String,
     inspect_key: String,
     limit: Option<usize>,
 }
@@ -234,7 +306,7 @@ impl Tool for MetaInfoTool {
     async fn do_work(&self) -> std::result::Result<(), BoxedError> {
         let result = MetadataSnapshotManager::info(
             &self.inner,
-            &self.file_path,
+            &self.source_file,
             &self.inspect_key,
             self.limit,
         )
@@ -248,90 +320,45 @@ impl Tool for MetaInfoTool {
 }
 
 impl InfoCommand {
-    async fn build(&self) -> Result<MetaInfoTool, BoxedError> {
+    fn decide_object_store_root_for_local_store(
-        let (object_store, file_path) = build_object_store_and_resolve_file_path(
+        file_path: &str,
-            self.object_store.clone(),
+    ) -> Result<(&str, &str), BoxedError> {
-            &self.dir,
+        let path = Path::new(file_path);
-            &self.file_path,
+        let parent = path
-        )?;
+            .parent()
-        let tool = MetaInfoTool {
+            .and_then(|p| p.to_str())
-            inner: object_store,
+            .context(InvalidFilePathSnafu { msg: file_path })
-            file_path,
+            .map_err(BoxedError::new)?;
-            inspect_key: self.inspect_key.clone(),
+        let file_name = path
-            limit: self.limit,
+            .file_name()
-        };
+            .and_then(|f| f.to_str())
-        Ok(tool)
+            .context(InvalidFilePathSnafu { msg: file_path })
-    }
+            .map_err(BoxedError::new)?;
-}
+        let root = if parent.is_empty() { "." } else { parent };
-
+        Ok((root, file_name))
-/// Builds the object store and resolves the file path.
+    }
-fn build_object_store_and_resolve_file_path(
+
-    object_store: ObjectStoreConfig,
+    pub async fn build(&self) -> Result<Box<dyn Tool>, BoxedError> {
-    fs_root: &str,
+        let object_store = self.s3_config.build("").map_err(BoxedError::new)?;
-    file_path: &str,
+        if let Some(store) = object_store {
-) -> Result<(ObjectStore, String), BoxedError> {
+            let tool = MetaInfoTool {
-    let object_store = object_store.build().map_err(BoxedError::new)?;
+                inner: store,
-    let object_store = match object_store {
+                source_file: self.file_name.clone(),
-        Some(object_store) => object_store,
+                inspect_key: self.inspect_key.clone(),
-        None => new_fs_object_store(fs_root)?,
+                limit: self.limit,
-    };
+            };
-
+            Ok(Box::new(tool))
-    let file_path = if matches!(object_store.info().scheme(), Scheme::Fs) {
+        } else {
-        resolve_relative_path_with_current_dir(file_path).map_err(BoxedError::new)?
+            let (root, file_name) =
-    } else {
+                Self::decide_object_store_root_for_local_store(&self.file_name)?;
-        file_path.to_string()
+            let object_store = create_local_file_object_store(root)?;
-    };
+            let tool = MetaInfoTool {
-
+                inner: object_store,
-    Ok((object_store, file_path))
+                source_file: file_name.to_string(),
-}
+                inspect_key: self.inspect_key.clone(),
-
+                limit: self.limit,
-#[cfg(test)]
+            };
-mod tests {
+            Ok(Box::new(tool))
-    use std::env;
+        }
-
-    use clap::Parser;
-
-    use crate::metadata::snapshot::RestoreCommand;
-
-    #[tokio::test]
-    async fn test_cmd_resolve_file_path() {
-        common_telemetry::init_default_ut_logging();
-        let cmd = RestoreCommand::parse_from([
-            "",
-            "--file_name",
-            "metadata_snapshot.metadata.fb",
-            "--backend",
-            "memory-store",
-            "--store-addrs",
-            "memory://",
-        ]);
-        let tool = cmd.build().await.unwrap();
-        let current_dir = env::current_dir().unwrap();
-        let file_path = current_dir.join("metadata_snapshot.metadata.fb");
-        assert_eq!(tool.file_path, file_path.to_string_lossy().to_string());
-
-        let cmd = RestoreCommand::parse_from([
-            "",
-            "--file_name",
-            "metadata_snapshot.metadata.fb",
-            "--backend",
-            "memory-store",
-            "--store-addrs",
-            "memory://",
-        ]);
-        let tool = cmd.build().await.unwrap();
-        assert_eq!(tool.file_path, file_path.to_string_lossy().to_string());
-
-        let cmd = RestoreCommand::parse_from([
-            "",
-            "--file_name",
-            "metadata_snapshot.metadata.fb",
-            "--backend",
-            "memory-store",
-            "--store-addrs",
-            "memory://",
-        ]);
-        let tool = cmd.build().await.unwrap();
-        assert_eq!(tool.file_path, file_path.to_string_lossy().to_string());
     }
 }

src/cli/src/utils.rs

@@ -1,94 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use std::env;
-use std::path::Path;
-
-use snafu::ResultExt;
-
-use crate::error::{GetCurrentDirSnafu, Result};
-
-/// Resolves the relative path to an absolute path.
-pub fn resolve_relative_path(current_dir: impl AsRef<Path>, path_str: &str) -> String {
-    let path = Path::new(path_str);
-    if path.is_relative() {
-        let path = current_dir.as_ref().join(path);
-        common_telemetry::debug!("Resolved relative path: {}", path.to_string_lossy());
-        path.to_string_lossy().to_string()
-    } else {
-        path_str.to_string()
-    }
-}
-
-/// Resolves the relative path to an absolute path.
-pub fn resolve_relative_path_with_current_dir(path_str: &str) -> Result<String> {
-    let current_dir = env::current_dir().context(GetCurrentDirSnafu)?;
-    Ok(resolve_relative_path(current_dir, path_str))
-}
-
-#[cfg(test)]
-mod tests {
-    use std::env;
-    use std::path::PathBuf;
-
-    use super::*;
-
-    #[test]
-    fn test_resolve_relative_path_absolute() {
-        let abs_path = if cfg!(windows) {
-            "C:\\foo\\bar"
-        } else {
-            "/foo/bar"
-        };
-        let current_dir = PathBuf::from("/tmp");
-        let result = resolve_relative_path(&current_dir, abs_path);
-        assert_eq!(result, abs_path);
-    }
-
-    #[test]
-    fn test_resolve_relative_path_relative() {
-        let current_dir = PathBuf::from("/tmp");
-        let rel_path = "foo/bar";
-        let expected = "/tmp/foo/bar";
-        let result = resolve_relative_path(&current_dir, rel_path);
-        // On Windows, the separator is '\', so normalize for comparison
-        // '/' is as a normal character in Windows paths
-        if cfg!(windows) {
-            assert!(result.ends_with("foo/bar"));
-            assert!(result.contains("/tmp\\"));
-        } else {
-            assert_eq!(result, expected);
-        }
-    }
-
-    #[test]
-    fn test_resolve_relative_path_with_current_dir_absolute() {
-        let abs_path = if cfg!(windows) {
-            "C:\\foo\\bar"
-        } else {
-            "/foo/bar"
-        };
-        let result = resolve_relative_path_with_current_dir(abs_path).unwrap();
-        assert_eq!(result, abs_path);
-    }
-
-    #[test]
-    fn test_resolve_relative_path_with_current_dir_relative() {
-        let rel_path = "foo/bar";
-        let current_dir = env::current_dir().unwrap();
-        let expected = current_dir.join(rel_path).to_string_lossy().to_string();
-        let result = resolve_relative_path_with_current_dir(rel_path).unwrap();
-        assert_eq!(result, expected);
-    }
-}

src/client/src/error.rs

@@ -18,7 +18,7 @@ use common_error::define_from_tonic_status;
 use common_error::ext::{BoxedError, ErrorExt};
 use common_error::status_code::StatusCode;
 use common_macro::stack_trace_debug;
-use snafu::{Location, Snafu};
+use snafu::{Location, Snafu, location};
 use tonic::Code;
 use tonic::metadata::errors::InvalidMetadataValue;
 

src/cmd/Cargo.toml

@@ -29,11 +29,9 @@ base64.workspace = true
 cache.workspace = true
 catalog.workspace = true
 chrono.workspace = true
-either = "1.15"
 clap.workspace = true
 cli.workspace = true
 client.workspace = true
-colored = "2.1.0"
 common-base.workspace = true
 common-catalog.workspace = true
 common-config.workspace = true
@@ -56,6 +54,7 @@ common-wal.workspace = true
 datanode.workspace = true
 datatypes.workspace = true
 etcd-client.workspace = true
+file-engine.workspace = true
 flow.workspace = true
 frontend = { workspace = true, default-features = false }
 futures.workspace = true
@@ -69,7 +68,6 @@ mito2.workspace = true
 moka.workspace = true
 nu-ansi-term = "0.46"
 object-store.workspace = true
-parquet = { workspace = true, features = ["object_store"] }
 plugins.workspace = true
 prometheus.workspace = true
 prost.workspace = true
@@ -77,26 +75,21 @@ query.workspace = true
 rand.workspace = true
 regex.workspace = true
 reqwest.workspace = true
-standalone.workspace = true
 serde.workspace = true
 serde_json.workspace = true
 servers.workspace = true
 session.workspace = true
 similar-asserts.workspace = true
 snafu.workspace = true
-common-stat.workspace = true
+stat.workspace = true
 store-api.workspace = true
+substrait.workspace = true
 table.workspace = true
 tokio.workspace = true
 toml.workspace = true
 tonic.workspace = true
 tracing-appender.workspace = true
 
-[target.'cfg(unix)'.dependencies]
-pprof = { version = "0.14", features = [
-    "flamegraph",
-] }
-
 [target.'cfg(not(windows))'.dependencies]
 tikv-jemallocator = "0.6"
 
@@ -107,8 +100,6 @@ common-version.workspace = true
 serde.workspace = true
 temp-env = "0.3"
 tempfile.workspace = true
-file-engine.workspace = true
-mito2.workspace = true
 
 [target.'cfg(not(windows))'.dev-dependencies]
 rexpect = "0.5"

src/cmd/src/bin/greptime.rs

@@ -103,15 +103,12 @@ async fn main_body() -> Result<()> {
 
 async fn start(cli: Command) -> Result<()> {
     match cli.subcmd {
-        SubCommand::Datanode(cmd) => match cmd.subcmd {
+        SubCommand::Datanode(cmd) => {
-            datanode::SubCommand::Start(ref start) => {
+            let opts = cmd.load_options(&cli.global_options)?;
-                let opts = start.load_options(&cli.global_options)?;
+            let plugins = Plugins::new();
-                let plugins = Plugins::new();
+            let builder = InstanceBuilder::try_new_with_init(opts, plugins).await?;
-                let builder = InstanceBuilder::try_new_with_init(opts, plugins).await?;
+            cmd.build_with(builder).await?.run().await
-                cmd.build_with(builder).await?.run().await
+        }
-            }
-            datanode::SubCommand::Objbench(ref bench) => bench.run().await,
-        },
         SubCommand::Flownode(cmd) => {
             cmd.build(cmd.load_options(&cli.global_options)?)
                 .await?

src/cmd/src/datanode.rs

@@ -13,8 +13,6 @@
 // limitations under the License.
 
 pub mod builder;
-#[allow(clippy::print_stdout)]
-mod objbench;
 
 use std::path::Path;
 use std::time::Duration;
@@ -25,16 +23,13 @@ use common_config::Configurable;
 use common_telemetry::logging::{DEFAULT_LOGGING_DIR, TracingOptions};
 use common_telemetry::{info, warn};
 use common_wal::config::DatanodeWalConfig;
-use datanode::config::RegionEngineConfig;
 use datanode::datanode::Datanode;
 use meta_client::MetaClientOptions;
-use serde::{Deserialize, Serialize};
 use snafu::{ResultExt, ensure};
 use tracing_appender::non_blocking::WorkerGuard;
 
 use crate::App;
 use crate::datanode::builder::InstanceBuilder;
-use crate::datanode::objbench::ObjbenchCommand;
 use crate::error::{
     LoadLayeredConfigSnafu, MissingConfigSnafu, Result, ShutdownDatanodeSnafu, StartDatanodeSnafu,
 };
@@ -94,7 +89,7 @@ impl App for Instance {
 #[derive(Parser)]
 pub struct Command {
     #[clap(subcommand)]
-    pub subcmd: SubCommand,
+    subcmd: SubCommand,
 }
 
 impl Command {
@@ -105,26 +100,13 @@ impl Command {
     pub fn load_options(&self, global_options: &GlobalOptions) -> Result<DatanodeOptions> {
         match &self.subcmd {
             SubCommand::Start(cmd) => cmd.load_options(global_options),
-            SubCommand::Objbench(_) => {
-                // For objbench command, we don't need to load DatanodeOptions
-                // It's a standalone utility command
-                let mut opts = datanode::config::DatanodeOptions::default();
-                opts.sanitize();
-                Ok(DatanodeOptions {
-                    runtime: Default::default(),
-                    plugins: Default::default(),
-                    component: opts,
-                })
-            }
         }
     }
 }
 
 #[derive(Parser)]
-pub enum SubCommand {
+enum SubCommand {
     Start(StartCommand),
-    /// Object storage benchmark tool
-    Objbench(ObjbenchCommand),
 }
 
 impl SubCommand {
@@ -134,33 +116,12 @@ impl SubCommand {
                 info!("Building datanode with {:#?}", cmd);
                 builder.build().await
             }
-            SubCommand::Objbench(cmd) => {
-                cmd.run().await?;
-                std::process::exit(0);
-            }
         }
     }
 }
 
-/// Storage engine config
-#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Default)]
-#[serde(default)]
-pub struct StorageConfig {
-    /// The working directory of database
-    pub data_home: String,
-    #[serde(flatten)]
-    pub store: object_store::config::ObjectStoreConfig,
-}
-
-#[derive(Debug, Clone, Serialize, Deserialize, PartialEq, Default)]
-#[serde(default)]
-struct StorageConfigWrapper {
-    storage: StorageConfig,
-    region_engine: Vec<RegionEngineConfig>,
-}
-
 #[derive(Debug, Parser, Default)]
-pub struct StartCommand {
+struct StartCommand {
     #[clap(long)]
     node_id: Option<u64>,
     /// The address to bind the gRPC server.
@@ -188,7 +149,7 @@ pub struct StartCommand {
 }
 
 impl StartCommand {
-    pub fn load_options(&self, global_options: &GlobalOptions) -> Result<DatanodeOptions> {
+    fn load_options(&self, global_options: &GlobalOptions) -> Result<DatanodeOptions> {
         let mut opts = DatanodeOptions::load_layered_options(
             self.config_file.as_deref(),
             self.env_prefix.as_ref(),

src/cmd/src/datanode/objbench.rs

@@ -1,676 +0,0 @@
-// Copyright 2023 Greptime Team
-//
-// Licensed under the Apache License, Version 2.0 (the "License");
-// you may not use this file except in compliance with the License.
-// You may obtain a copy of the License at
-//
-//     http://www.apache.org/licenses/LICENSE-2.0
-//
-// Unless required by applicable law or agreed to in writing, software
-// distributed under the License is distributed on an "AS IS" BASIS,
-// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
-// See the License for the specific language governing permissions and
-// limitations under the License.
-
-use std::path::PathBuf;
-use std::sync::Arc;
-use std::time::Instant;
-
-use clap::Parser;
-use colored::Colorize;
-use datanode::config::RegionEngineConfig;
-use datanode::store;
-use either::Either;
-use mito2::access_layer::{
-    AccessLayer, AccessLayerRef, Metrics, OperationType, SstWriteRequest, WriteType,
-};
-use mito2::cache::{CacheManager, CacheManagerRef};
-use mito2::config::{FulltextIndexConfig, MitoConfig, Mode};
-use mito2::read::Source;
-use mito2::sst::file::{FileHandle, FileMeta};
-use mito2::sst::file_purger::{FilePurger, FilePurgerRef};
-use mito2::sst::index::intermediate::IntermediateManager;
-use mito2::sst::index::puffin_manager::PuffinManagerFactory;
-use mito2::sst::parquet::reader::ParquetReaderBuilder;
-use mito2::sst::parquet::{PARQUET_METADATA_KEY, WriteOptions};
-use mito2::worker::write_cache_from_config;
-use object_store::ObjectStore;
-use regex::Regex;
-use snafu::OptionExt;
-use store_api::metadata::{RegionMetadata, RegionMetadataRef};
-use store_api::path_utils::region_name;
-use store_api::region_request::PathType;
-use store_api::storage::FileId;
-
-use crate::datanode::{StorageConfig, StorageConfigWrapper};
-use crate::error;
-
-/// Object storage benchmark command
-#[derive(Debug, Parser)]
-pub struct ObjbenchCommand {
-    /// Path to the object-store config file (TOML). Must deserialize into object_store::config::ObjectStoreConfig.
-    #[clap(long, value_name = "FILE")]
-    pub config: PathBuf,
-
-    /// Source SST file path in object-store (e.g. "region_dir/<uuid>.parquet").
-    #[clap(long, value_name = "PATH")]
-    pub source: String,
-
-    /// Verbose output
-    #[clap(short, long, default_value_t = false)]
-    pub verbose: bool,
-
-    /// Output file path for pprof flamegraph (enables profiling)
-    #[clap(long, value_name = "FILE")]
-    pub pprof_file: Option<PathBuf>,
-}
-
-fn parse_config(config_path: &PathBuf) -> error::Result<(StorageConfig, MitoConfig)> {
-    let cfg_str = std::fs::read_to_string(config_path).map_err(|e| {
-        error::IllegalConfigSnafu {
-            msg: format!("failed to read config {}: {e}", config_path.display()),
-        }
-        .build()
-    })?;
-
-    let store_cfg: StorageConfigWrapper = toml::from_str(&cfg_str).map_err(|e| {
-        error::IllegalConfigSnafu {
-            msg: format!("failed to parse config {}: {e}", config_path.display()),
-        }
-        .build()
-    })?;
-
-    let storage_config = store_cfg.storage;
-    let mito_engine_config = store_cfg
-        .region_engine
-        .into_iter()
-        .filter_map(|c| {
-            if let RegionEngineConfig::Mito(mito) = c {
-                Some(mito)
-            } else {
-                None
-            }
-        })
-        .next()
-        .with_context(|| error::IllegalConfigSnafu {
-            msg: format!("Engine config not found in {:?}", config_path),
-        })?;
-    Ok((storage_config, mito_engine_config))
-}
-
-impl ObjbenchCommand {
-    pub async fn run(&self) -> error::Result<()> {
-        if self.verbose {
-            common_telemetry::init_default_ut_logging();
-        }
-
-        println!("{}", "Starting objbench with config:".cyan().bold());
-
-        // Build object store from config
-        let (store_cfg, mut mito_engine_config) = parse_config(&self.config)?;
-
-        let object_store = build_object_store(&store_cfg).await?;
-        println!("{} Object store initialized", "✓".green());
-
-        // Prepare source identifiers
-        let components = parse_file_dir_components(&self.source)?;
-        println!(
-            "{} Source path parsed: {}, components: {:?}",
-            "✓".green(),
-            self.source,
-            components
-        );
-
-        // Load parquet metadata to extract RegionMetadata and file stats
-        println!("{}", "Loading parquet metadata...".yellow());
-        let file_size = object_store
-            .stat(&self.source)
-            .await
-            .map_err(|e| {
-                error::IllegalConfigSnafu {
-                    msg: format!("stat failed: {e}"),
-                }
-                .build()
-            })?
-            .content_length();
-        let parquet_meta = load_parquet_metadata(object_store.clone(), &self.source, file_size)
-            .await
-            .map_err(|e| {
-                error::IllegalConfigSnafu {
-                    msg: format!("read parquet metadata failed: {e}"),
-                }
-                .build()
-            })?;
-
-        let region_meta = extract_region_metadata(&self.source, &parquet_meta)?;
-        let num_rows = parquet_meta.file_metadata().num_rows() as u64;
-        let num_row_groups = parquet_meta.num_row_groups() as u64;
-
-        println!(
-            "{} Metadata loaded - rows: {}, size: {} bytes",
-            "✓".green(),
-            num_rows,
-            file_size
-        );
-
-        // Build a FileHandle for the source file
-        let file_meta = FileMeta {
-            region_id: region_meta.region_id,
-            file_id: components.file_id,
-            time_range: Default::default(),
-            level: 0,
-            file_size,
-            available_indexes: Default::default(),
-            index_file_size: 0,
-            num_rows,
-            num_row_groups,
-            sequence: None,
-            partition_expr: None,
-            num_series: 0,
-        };
-        let src_handle = FileHandle::new(file_meta, new_noop_file_purger());
-
-        // Build the reader for a single file via ParquetReaderBuilder
-        let table_dir = components.table_dir();
-        let (src_access_layer, cache_manager) = build_access_layer_simple(
-            &components,
-            object_store.clone(),
-            &mut mito_engine_config,
-            &store_cfg.data_home,
-        )
-        .await?;
-        let reader_build_start = Instant::now();
-
-        let reader = ParquetReaderBuilder::new(
-            table_dir,
-            components.path_type,
-            src_handle.clone(),
-            object_store.clone(),
-        )
-        .expected_metadata(Some(region_meta.clone()))
-        .build()
-        .await
-        .map_err(|e| {
-            error::IllegalConfigSnafu {
-                msg: format!("build reader failed: {e:?}"),
-            }
-            .build()
-        })?;
-
-        let reader_build_elapsed = reader_build_start.elapsed();
-        let total_rows = reader.parquet_metadata().file_metadata().num_rows();
-        println!("{} Reader built in {:?}", "✓".green(), reader_build_elapsed);
-
-        // Build write request
-        let fulltext_index_config = FulltextIndexConfig {
-            create_on_compaction: Mode::Disable,
-            ..Default::default()
-        };
-
-        let write_req = SstWriteRequest {
-            op_type: OperationType::Flush,
-            metadata: region_meta,
-            source: Either::Left(Source::Reader(Box::new(reader))),
-            cache_manager,
-            storage: None,
-            max_sequence: None,
-            index_options: Default::default(),
-            index_config: mito_engine_config.index.clone(),
-            inverted_index_config: MitoConfig::default().inverted_index,
-            fulltext_index_config,
-            bloom_filter_index_config: MitoConfig::default().bloom_filter_index,
-        };
-
-        // Write SST
-        println!("{}", "Writing SST...".yellow());
-
-        // Start profiling if pprof_file is specified
-        #[cfg(unix)]
-        let profiler_guard = if self.pprof_file.is_some() {
-            println!("{} Starting profiling...", "⚡".yellow());
-            Some(
-                pprof::ProfilerGuardBuilder::default()
-                    .frequency(99)
-                    .blocklist(&["libc", "libgcc", "pthread", "vdso"])
-                    .build()
-                    .map_err(|e| {
-                        error::IllegalConfigSnafu {
-                            msg: format!("Failed to start profiler: {e}"),
-                        }
-                        .build()
-                    })?,
-            )
-        } else {
-            None
-        };
-
-        #[cfg(not(unix))]
-        if self.pprof_file.is_some() {
-            eprintln!(
-                "{}: Profiling is not supported on this platform",
-                "Warning".yellow()
-            );
-        }
-
-        let write_start = Instant::now();
-        let mut metrics = Metrics::new(WriteType::Flush);
-        let infos = src_access_layer
-            .write_sst(write_req, &WriteOptions::default(), &mut metrics)
-            .await
-            .map_err(|e| {
-                error::IllegalConfigSnafu {
-                    msg: format!("write_sst failed: {e:?}"),
-                }
-                .build()
-            })?;
-
-        let write_elapsed = write_start.elapsed();
-
-        // Stop profiling and generate flamegraph if enabled
-        #[cfg(unix)]
-        if let (Some(guard), Some(pprof_file)) = (profiler_guard, &self.pprof_file) {
-            println!("{} Generating flamegraph...", "🔥".yellow());
-            match guard.report().build() {
-                Ok(report) => {
-                    let mut flamegraph_data = Vec::new();
-                    if let Err(e) = report.flamegraph(&mut flamegraph_data) {
-                        println!("{}: Failed to generate flamegraph: {}", "Error".red(), e);
-                    } else if let Err(e) = std::fs::write(pprof_file, flamegraph_data) {
-                        println!(
-                            "{}: Failed to write flamegraph to {}: {}",
-                            "Error".red(),
-                            pprof_file.display(),
-                            e
-                        );
-                    } else {
-                        println!(
-                            "{} Flamegraph saved to {}",
-                            "✓".green(),
-                            pprof_file.display().to_string().cyan()
-                        );
-                    }
-                }
-                Err(e) => {
-                    println!("{}: Failed to generate pprof report: {}", "Error".red(), e);
-                }
-            }
-        }
-        assert_eq!(infos.len(), 1);
-        let dst_file_id = infos[0].file_id;
-        let dst_file_path = format!("{}/{}.parquet", components.region_dir(), dst_file_id);
-        let mut dst_index_path = None;
-        if infos[0].index_metadata.file_size > 0 {
-            dst_index_path = Some(format!(
-                "{}/index/{}.puffin",
-                components.region_dir(),
-                dst_file_id
-            ));
-        }
-
-        // Report results with ANSI colors
-        println!("\n{} {}", "Write complete!".green().bold(), "✓".green());
-        println!("  {}: {}", "Destination file".bold(), dst_file_path.cyan());
-        println!("  {}: {}", "Rows".bold(), total_rows.to_string().cyan());
-        println!(
-            "  {}: {}",
-            "File size".bold(),
-            format!("{} bytes", file_size).cyan()
-        );
-        println!(
-            "  {}: {:?}",
-            "Reader build time".bold(),
-            reader_build_elapsed
-        );
-        println!("  {}: {:?}", "Total time".bold(), write_elapsed);
-
-        // Print metrics in a formatted way
-        println!("  {}: {:?}", "Metrics".bold(), metrics,);
-
-        // Print infos
-        println!("  {}: {:?}", "Index".bold(), infos[0].index_metadata);
-
-        // Cleanup
-        println!("\n{}", "Cleaning up...".yellow());
-        object_store.delete(&dst_file_path).await.map_err(|e| {
-            error::IllegalConfigSnafu {
-                msg: format!("Failed to delete dest file {}: {}", dst_file_path, e),
-            }
-            .build()
-        })?;
-        println!("{} Temporary file {} deleted", "✓".green(), dst_file_path);
-
-        if let Some(index_path) = dst_index_path {
-            object_store.delete(&index_path).await.map_err(|e| {
-                error::IllegalConfigSnafu {
-                    msg: format!("Failed to delete dest index file {}: {}", index_path, e),
-                }
-                .build()
-            })?;
-            println!(
-                "{} Temporary index file {} deleted",
-                "✓".green(),
-                index_path
-            );
-        }
-
-        println!("\n{}", "Benchmark completed successfully!".green().bold());
-        Ok(())
-    }
-}
-
-#[derive(Debug)]
-struct FileDirComponents {
-    catalog: String,
-    schema: String,
-    table_id: u32,
-    region_sequence: u32,
-    path_type: PathType,
-    file_id: FileId,
-}
-
-impl FileDirComponents {
-    fn table_dir(&self) -> String {
-        format!("data/{}/{}/{}", self.catalog, self.schema, self.table_id)
-    }
-
-    fn region_dir(&self) -> String {
-        let region_name = region_name(self.table_id, self.region_sequence);
-        match self.path_type {
-            PathType::Bare => {
-                format!(
-                    "data/{}/{}/{}/{}",
-                    self.catalog, self.schema, self.table_id, region_name
-                )
-            }
-            PathType::Data => {
-                format!(
-                    "data/{}/{}/{}/{}/data",
-                    self.catalog, self.schema, self.table_id, region_name
-                )
-            }
-            PathType::Metadata => {
-                format!(
-                    "data/{}/{}/{}/{}/metadata",
-                    self.catalog, self.schema, self.table_id, region_name
-                )
-            }
-        }
-    }
-}
-
-fn parse_file_dir_components(path: &str) -> error::Result<FileDirComponents> {
-    // Define the regex pattern to match all three path styles
-    let pattern =
-        r"^data/([^/]+)/([^/]+)/([^/]+)/([^/]+)_([^/]+)(?:/data|/metadata)?/(.+).parquet$";
-
-    // Compile the regex
-    let re = Regex::new(pattern).expect("Invalid regex pattern");
-
-    // Determine the path type
-    let path_type = if path.contains("/data/") {
-        PathType::Data
-    } else if path.contains("/metadata/") {
-        PathType::Metadata
-    } else {
-        PathType::Bare
-    };
-
-    // Try to match the path
-    let components = (|| {
-        let captures = re.captures(path)?;
-        if captures.len() != 7 {
-            return None;
-        }
-        let mut components = FileDirComponents {
-            catalog: "".to_string(),
-            schema: "".to_string(),
-            table_id: 0,
-            region_sequence: 0,
-            path_type,
-            file_id: FileId::default(),
-        };
-        // Extract the components
-        components.catalog = captures.get(1)?.as_str().to_string();
-        components.schema = captures.get(2)?.as_str().to_string();
-        components.table_id = captures[3].parse().ok()?;
-        components.region_sequence = captures[5].parse().ok()?;
-        let file_id_str = &captures[6];
-        components.file_id = FileId::parse_str(file_id_str).ok()?;
-        Some(components)
-    })();
-    components.context(error::IllegalConfigSnafu {
-        msg: format!("Expect valid source file path, got: {}", path),
-    })
-}
-
-fn extract_region_metadata(
-    file_path: &str,
-    meta: &parquet::file::metadata::ParquetMetaData,
-) -> error::Result<RegionMetadataRef> {
-    use parquet::format::KeyValue;
-    let kvs: Option<&Vec<KeyValue>> = meta.file_metadata().key_value_metadata();
-    let Some(kvs) = kvs else {
-        return Err(error::IllegalConfigSnafu {
-            msg: format!("{file_path}: missing parquet key_value metadata"),
-        }
-        .build());
-    };
-    let json = kvs
-        .iter()
-        .find(|kv| kv.key == PARQUET_METADATA_KEY)
-        .and_then(|kv| kv.value.as_ref())
-        .ok_or_else(|| {
-            error::IllegalConfigSnafu {
-                msg: format!("{file_path}: key {PARQUET_METADATA_KEY} not found or empty"),
-            }
-            .build()
-        })?;
-    let region: RegionMetadata = RegionMetadata::from_json(json).map_err(|e| {
-        error::IllegalConfigSnafu {
-            msg: format!("invalid region metadata json: {e}"),
-        }
-        .build()
-    })?;
-    Ok(Arc::new(region))
-}
-
-async fn build_object_store(sc: &StorageConfig) -> error::Result<ObjectStore> {
-    store::new_object_store(sc.store.clone(), &sc.data_home)
-        .await
-        .map_err(|e| {
-            error::IllegalConfigSnafu {
-                msg: format!("Failed to build object store: {e:?}"),
-            }
-            .build()
-        })
-}
-
-async fn build_access_layer_simple(
-    components: &FileDirComponents,
-    object_store: ObjectStore,
-    config: &mut MitoConfig,
-    data_home: &str,
-) -> error::Result<(AccessLayerRef, CacheManagerRef)> {
-    let _ = config.index.sanitize(data_home, &config.inverted_index);
-    let puffin_manager = PuffinManagerFactory::new(
-        &config.index.aux_path,
-        config.index.staging_size.as_bytes(),
-        Some(config.index.write_buffer_size.as_bytes() as _),
-        config.index.staging_ttl,
-    )
-    .await
-    .map_err(|e| {
-        error::IllegalConfigSnafu {
-            msg: format!("Failed to build access layer: {e:?}"),
-        }
-        .build()
-    })?;
-
-    let intermediate_manager = IntermediateManager::init_fs(&config.index.aux_path)
-        .await
-        .map_err(|e| {
-            error::IllegalConfigSnafu {
-                msg: format!("Failed to build IntermediateManager: {e:?}"),
-            }
-            .build()
-        })?
-        .with_buffer_size(Some(config.index.write_buffer_size.as_bytes() as _));
-
-    let cache_manager =
-        build_cache_manager(config, puffin_manager.clone(), intermediate_manager.clone()).await?;
-    let layer = AccessLayer::new(
-        components.table_dir(),
-        components.path_type,
-        object_store,
-        puffin_manager,
-        intermediate_manager,
-    );
-    Ok((Arc::new(layer), cache_manager))
-}
-
-async fn build_cache_manager(
-    config: &MitoConfig,
-    puffin_manager: PuffinManagerFactory,
-    intermediate_manager: IntermediateManager,
-) -> error::Result<CacheManagerRef> {
-    let write_cache = write_cache_from_config(config, puffin_manager, intermediate_manager)
-        .await
-        .map_err(|e| {
-            error::IllegalConfigSnafu {
-                msg: format!("Failed to build write cache: {e:?}"),
-            }
-            .build()
-        })?;
-    let cache_manager = Arc::new(
-        CacheManager::builder()
-            .sst_meta_cache_size(config.sst_meta_cache_size.as_bytes())
-            .vector_cache_size(config.vector_cache_size.as_bytes())
-            .page_cache_size(config.page_cache_size.as_bytes())
-            .selector_result_cache_size(config.selector_result_cache_size.as_bytes())
-            .index_metadata_size(config.index.metadata_cache_size.as_bytes())
-            .index_content_size(config.index.content_cache_size.as_bytes())
-            .index_content_page_size(config.index.content_cache_page_size.as_bytes())
-            .index_result_cache_size(config.index.result_cache_size.as_bytes())
-            .puffin_metadata_size(config.index.metadata_cache_size.as_bytes())
-            .write_cache(write_cache)
-            .build(),
-    );
-    Ok(cache_manager)
-}
-
-fn new_noop_file_purger() -> FilePurgerRef {
-    #[derive(Debug)]
-    struct Noop;
-    impl FilePurger for Noop {
-        fn remove_file(&self, _file_meta: FileMeta, _is_delete: bool) {}
-    }
-    Arc::new(Noop)
-}
-
-async fn load_parquet_metadata(
-    object_store: ObjectStore,
-    path: &str,
-    file_size: u64,
-) -> Result<parquet::file::metadata::ParquetMetaData, Box<dyn std::error::Error + Send + Sync>> {
-    use parquet::file::FOOTER_SIZE;
-    use parquet::file::metadata::ParquetMetaDataReader;
-    let actual_size = if file_size == 0 {
-        object_store.stat(path).await?.content_length()
-    } else {
-        file_size
-    };
-    if actual_size < FOOTER_SIZE as u64 {
-        return Err("file too small".into());
-    }
-    let prefetch: u64 = 64 * 1024;
-    let start = actual_size.saturating_sub(prefetch);
-    let buffer = object_store
-        .read_with(path)
-        .range(start..actual_size)
-        .await?
-        .to_vec();
-    let buffer_len = buffer.len();
-    let mut footer = [0; 8];
-    footer.copy_from_slice(&buffer[buffer_len - FOOTER_SIZE..]);
-    let footer = ParquetMetaDataReader::decode_footer_tail(&footer)?;
-    let metadata_len = footer.metadata_length() as u64;
-    if actual_size - (FOOTER_SIZE as u64) < metadata_len {
-        return Err("invalid footer/metadata length".into());
-    }
-    if (metadata_len as usize) <= buffer_len - FOOTER_SIZE {
-        let metadata_start = buffer_len - metadata_len as usize - FOOTER_SIZE;
-        let meta = ParquetMetaDataReader::decode_metadata(
-            &buffer[metadata_start..buffer_len - FOOTER_SIZE],
-        )?;
-        Ok(meta)
-    } else {
-        let metadata_start = actual_size - metadata_len - FOOTER_SIZE as u64;
-        let data = object_store
-            .read_with(path)
-            .range(metadata_start..(actual_size - FOOTER_SIZE as u64))
-            .await?
-            .to_vec();
-        let meta = ParquetMetaDataReader::decode_metadata(&data)?;
-        Ok(meta)
-    }
-}
-
-#[cfg(test)]
-mod tests {
-    use std::path::PathBuf;
-    use std::str::FromStr;
-
-    use common_base::readable_size::ReadableSize;
-    use store_api::region_request::PathType;
-
-    use crate::datanode::objbench::{parse_config, parse_file_dir_components};
-
-    #[test]
-    fn test_parse_dir() {
-        let meta_path = "data/greptime/public/1024/1024_0000000000/metadata/00020380-009c-426d-953e-b4e34c15af34.parquet";
-        let c = parse_file_dir_components(meta_path).unwrap();
-        assert_eq!(
-            c.file_id.to_string(),
-            "00020380-009c-426d-953e-b4e34c15af34"
-        );
-        assert_eq!(c.catalog, "greptime");
-        assert_eq!(c.schema, "public");
-        assert_eq!(c.table_id, 1024);
-        assert_eq!(c.region_sequence, 0);
-        assert_eq!(c.path_type, PathType::Metadata);
-
-        let c = parse_file_dir_components(
-            "data/greptime/public/1024/1024_0000000000/data/00020380-009c-426d-953e-b4e34c15af34.parquet",
-        ).unwrap();
-        assert_eq!(
-            c.file_id.to_string(),
-            "00020380-009c-426d-953e-b4e34c15af34"
-        );
-        assert_eq!(c.catalog, "greptime");
-        assert_eq!(c.schema, "public");
-        assert_eq!(c.table_id, 1024);
-        assert_eq!(c.region_sequence, 0);
-        assert_eq!(c.path_type, PathType::Data);
-
-        let c = parse_file_dir_components(
-            "data/greptime/public/1024/1024_0000000000/00020380-009c-426d-953e-b4e34c15af34.parquet",
-        ).unwrap();
-        assert_eq!(
-            c.file_id.to_string(),
-            "00020380-009c-426d-953e-b4e34c15af34"
-        );
-        assert_eq!(c.catalog, "greptime");
-        assert_eq!(c.schema, "public");
-        assert_eq!(c.table_id, 1024);
-        assert_eq!(c.region_sequence, 0);
-        assert_eq!(c.path_type, PathType::Bare);
-    }
-
-    #[test]
-    fn test_parse_config() {
-        let path = "../../config/datanode.example.toml";
-        let (storage, engine) = parse_config(&PathBuf::from_str(path).unwrap()).unwrap();
-        assert_eq!(storage.data_home, "./greptimedb_data");
-        assert_eq!(engine.index.staging_size, ReadableSize::gb(2));
-    }
-}

src/cmd/src/error.rs

@@ -302,27 +302,6 @@ pub enum Error {
         location: Location,
         source: common_meta::error::Error,
     },
-
-    #[snafu(display("Failed to build metadata kvbackend"))]
-    BuildMetadataKvbackend {
-        #[snafu(implicit)]
-        location: Location,
-        source: standalone::error::Error,
-    },
-
-    #[snafu(display("Failed to setup standalone plugins"))]
-    SetupStandalonePlugins {
-        #[snafu(implicit)]
-        location: Location,
-        source: standalone::error::Error,
-    },
-
-    #[snafu(display("Invalid WAL provider"))]
-    InvalidWalProvider {
-        #[snafu(implicit)]
-        location: Location,
-        source: common_wal::error::Error,
-    },
 }
 
 pub type Result<T> = std::result::Result<T, Error>;
@@ -341,8 +320,6 @@ impl ErrorExt for Error {
             Error::UnsupportedSelectorType { source, .. } => source.status_code(),
             Error::BuildCli { source, .. } => source.status_code(),
             Error::StartCli { source, .. } => source.status_code(),
-            Error::BuildMetadataKvbackend { source, .. } => source.status_code(),
-            Error::SetupStandalonePlugins { source, .. } => source.status_code(),
 
             Error::InitMetadata { source, .. } | Error::InitDdlManager { source, .. } => {
                 source.status_code()
@@ -380,7 +357,6 @@ impl ErrorExt for Error {
             }
             Error::MetaClientInit { source, .. } => source.status_code(),
             Error::SchemaNotFound { .. } => StatusCode::DatabaseNotFound,
-            Error::InvalidWalProvider { .. } => StatusCode::InvalidArguments,
         }
     }
 

src/cmd/src/flownode.rs

@@ -30,7 +30,6 @@ use common_meta::heartbeat::handler::invalidate_table_cache::InvalidateCacheHand
 use common_meta::heartbeat::handler::parse_mailbox_message::ParseMailboxMessageHandler;
 use common_meta::key::TableMetadataManager;
 use common_meta::key::flow::FlowMetadataManager;
-use common_stat::ResourceStatImpl;
 use common_telemetry::info;
 use common_telemetry::logging::{DEFAULT_LOGGING_DIR, TracingOptions};
 use common_version::{short_version, verbose_version};
@@ -373,15 +372,11 @@ impl StartCommand {
             Arc::new(InvalidateCacheHandler::new(layered_cache_registry.clone())),
         ]);
 
-        let mut resource_stat = ResourceStatImpl::default();
-        resource_stat.start_collect_cpu_usage();
-
         let heartbeat_task = flow::heartbeat::HeartbeatTask::new(
             &opts,
             meta_client.clone(),
             opts.heartbeat.clone(),
             Arc::new(executor),
-            Arc::new(resource_stat),
         );
 
         let flow_metadata_manager = Arc::new(FlowMetadataManager::new(cached_meta_backend.clone()));

src/cmd/src/frontend.rs

@@ -25,14 +25,11 @@ use clap::Parser;
 use client::client_manager::NodeClients;
 use common_base::Plugins;
 use common_config::{Configurable, DEFAULT_DATA_HOME};
-use common_error::ext::BoxedError;
 use common_grpc::channel_manager::ChannelConfig;
 use common_meta::cache::{CacheRegistryBuilder, LayeredCacheRegistryBuilder};
 use common_meta::heartbeat::handler::HandlerGroupExecutor;
 use common_meta::heartbeat::handler::invalidate_table_cache::InvalidateCacheHandler;
 use common_meta::heartbeat::handler::parse_mailbox_message::ParseMailboxMessageHandler;
-use common_query::prelude::set_default_prefix;
-use common_stat::ResourceStatImpl;
 use common_telemetry::info;
 use common_telemetry::logging::{DEFAULT_LOGGING_DIR, TracingOptions};
 use common_time::timezone::set_default_timezone;
@@ -255,10 +252,10 @@ impl StartCommand {
 
         if let Some(addr) = &self.internal_rpc_bind_addr {
             if let Some(internal_grpc) = &mut opts.internal_grpc {
-                internal_grpc.bind_addr = addr.clone();
+                internal_grpc.bind_addr = addr.to_string();
             } else {
                 let grpc_options = GrpcOptions {
-                    bind_addr: addr.clone(),
+                    bind_addr: addr.to_string(),
                     ..Default::default()
                 };
 
@@ -268,10 +265,10 @@ impl StartCommand {
 
         if let Some(addr) = &self.internal_rpc_server_addr {
             if let Some(internal_grpc) = &mut opts.internal_grpc {
-                internal_grpc.server_addr = addr.clone();
+                internal_grpc.server_addr = addr.to_string();
             } else {
                 let grpc_options = GrpcOptions {
-                    server_addr: addr.clone(),
+                    server_addr: addr.to_string(),
                     ..Default::default()
                 };
                 opts.internal_grpc = Some(grpc_options);
@@ -335,9 +332,6 @@ impl StartCommand {
             .context(error::StartFrontendSnafu)?;
 
         set_default_timezone(opts.default_timezone.as_deref()).context(error::InitTimezoneSnafu)?;
-        set_default_prefix(opts.default_column_prefix.as_deref())
-            .map_err(BoxedError::new)
-            .context(error::BuildCliSnafu)?;
 
         let meta_client_options = opts
             .meta_client
@@ -427,15 +421,11 @@ impl StartCommand {
             Arc::new(InvalidateCacheHandler::new(layered_cache_registry.clone())),
         ]);
 
-        let mut resource_stat = ResourceStatImpl::default();
-        resource_stat.start_collect_cpu_usage();
-
         let heartbeat_task = HeartbeatTask::new(
             &opts,
             meta_client.clone(),
             opts.heartbeat.clone(),
             Arc::new(executor),
-            Arc::new(resource_stat),
         );
         let heartbeat_task = Some(heartbeat_task);
 

src/cmd/src/lib.rs

@@ -12,14 +12,14 @@
 // See the License for the specific language governing permissions and
 // limitations under the License.
 
-#![feature(assert_matches)]
+#![feature(assert_matches, let_chains)]
 
 use async_trait::async_trait;
 use common_error::ext::ErrorExt;
 use common_error::status_code::StatusCode;
 use common_mem_prof::activate_heap_profile;
-use common_stat::{get_total_cpu_millicores, get_total_memory_bytes};
 use common_telemetry::{error, info, warn};
+use stat::{get_cpu_limit, get_memory_limit};
 
 use crate::error::Result;
 
@@ -125,8 +125,7 @@ pub fn log_versions(version: &str, short_version: &str, app: &str) {
 }
 
 pub fn create_resource_limit_metrics(app: &str) {
-    let cpu_limit = get_total_cpu_millicores();
+    if let Some(cpu_limit) = get_cpu_limit() {
-    if cpu_limit > 0 {
         info!(
             "GreptimeDB start with cpu limit in millicores: {}",
             cpu_limit
@@ -134,8 +133,7 @@ pub fn create_resource_limit_metrics(app: &str) {
         CPU_LIMIT.with_label_values(&[app]).set(cpu_limit);
     }
 
-    let memory_limit = get_total_memory_bytes();
+    if let Some(memory_limit) = get_memory_limit() {
-    if memory_limit > 0 {
         info!(
             "GreptimeDB start with memory limit in bytes: {}",
             memory_limit

src/cmd/src/metasrv.rs

@@ -399,6 +399,7 @@ mod tests {
             threshold = 8.0
             min_std_deviation = "100ms"
             acceptable_heartbeat_pause = "3000ms"
+            first_heartbeat_estimate = "1000ms"
         "#;
         write!(file, "{}", toml_str).unwrap();
 
@@ -429,6 +430,13 @@ mod tests {
                 .acceptable_heartbeat_pause
                 .as_millis()
         );
+        assert_eq!(
+            1000,
+            options
+                .failure_detector
+                .first_heartbeat_estimate
+                .as_millis()
+        );
         assert_eq!(
             options.procedure.max_metadata_value_size,
             Some(ReadableSize::kb(1500))

src/cmd/src/standalone.rs

@@ -19,49 +19,71 @@ use std::{fs, path};
 
 use async_trait::async_trait;
 use cache::{build_fundamental_cache_registry, with_default_composite_cache_registry};
-use catalog::information_schema::InformationExtensionRef;
+use catalog::information_schema::{DatanodeInspectRequest, InformationExtension};
 use catalog::kvbackend::KvBackendCatalogManagerBuilder;
 use catalog::process_manager::ProcessManager;
 use clap::Parser;
+use client::SendableRecordBatchStream;
+use client::api::v1::meta::RegionRole;
 use common_base::Plugins;
+use common_base::readable_size::ReadableSize;
 use common_catalog::consts::{MIN_USER_FLOW_ID, MIN_USER_TABLE_ID};
-use common_config::{Configurable, metadata_store_dir};
+use common_config::{Configurable, KvBackendConfig, metadata_store_dir};
 use common_error::ext::BoxedError;
 use common_meta::cache::LayeredCacheRegistryBuilder;
+use common_meta::cluster::{NodeInfo, NodeStatus};
+use common_meta::datanode::RegionStat;
 use common_meta::ddl::flow_meta::FlowMetadataAllocator;
 use common_meta::ddl::table_meta::TableMetadataAllocator;
 use common_meta::ddl::{DdlContext, NoopRegionFailureDetectorControl};
 use common_meta::ddl_manager::DdlManager;
 use common_meta::key::flow::FlowMetadataManager;
+use common_meta::key::flow::flow_state::FlowStat;
 use common_meta::key::{TableMetadataManager, TableMetadataManagerRef};
 use common_meta::kv_backend::KvBackendRef;
+use common_meta::peer::Peer;
 use common_meta::procedure_executor::LocalProcedureExecutor;
 use common_meta::region_keeper::MemoryRegionKeeper;
 use common_meta::region_registry::LeaderRegionRegistry;
 use common_meta::sequence::SequenceBuilder;
 use common_meta::wal_options_allocator::{WalOptionsAllocatorRef, build_wal_options_allocator};
-use common_procedure::ProcedureManagerRef;
+use common_options::memory::MemoryOptions;
-use common_query::prelude::set_default_prefix;
+use common_procedure::{ProcedureInfo, ProcedureManagerRef};
+use common_query::request::QueryRequest;
 use common_telemetry::info;
-use common_telemetry::logging::{DEFAULT_LOGGING_DIR, TracingOptions};
+use common_telemetry::logging::{
+    DEFAULT_LOGGING_DIR, LoggingOptions, SlowQueryOptions, TracingOptions,
+};
 use common_time::timezone::set_default_timezone;
 use common_version::{short_version, verbose_version};
-use datanode::config::DatanodeOptions;
+use common_wal::config::DatanodeWalConfig;
+use datanode::config::{DatanodeOptions, ProcedureConfig, RegionEngineConfig, StorageConfig};
 use datanode::datanode::{Datanode, DatanodeBuilder};
+use datanode::region_server::RegionServer;
+use file_engine::config::EngineConfig as FileEngineConfig;
 use flow::{
-    FlownodeBuilder, FlownodeInstance, FlownodeOptions, FrontendClient, FrontendInvoker,
+    FlowConfig, FlownodeBuilder, FlownodeInstance, FlownodeOptions, FrontendClient,
-    GrpcQueryHandlerWithBoxedError,
+    FrontendInvoker, GrpcQueryHandlerWithBoxedError, StreamingEngine,
 };
-use frontend::frontend::Frontend;
+use frontend::frontend::{Frontend, FrontendOptions};
-use frontend::instance::StandaloneDatanodeManager;
 use frontend::instance::builder::FrontendBuilder;
+use frontend::instance::{Instance as FeInstance, StandaloneDatanodeManager};
 use frontend::server::Services;
+use frontend::service_config::{
+    InfluxdbOptions, JaegerOptions, MysqlOptions, OpentsdbOptions, PostgresOptions,
+    PromStoreOptions,
+};
 use meta_srv::metasrv::{FLOW_ID_SEQ, TABLE_ID_SEQ};
-use servers::export_metrics::ExportMetricsTask;
+use mito2::config::MitoConfig;
+use query::options::QueryOptions;
+use serde::{Deserialize, Serialize};
+use servers::export_metrics::{ExportMetricsOption, ExportMetricsTask};
+use servers::grpc::GrpcOptions;
+use servers::http::HttpOptions;
 use servers::tls::{TlsMode, TlsOption};
 use snafu::ResultExt;
-use standalone::StandaloneInformationExtension;
+use store_api::storage::RegionId;
-use standalone::options::StandaloneOptions;
+use tokio::sync::RwLock;
 use tracing_appender::non_blocking::WorkerGuard;
 
 use crate::error::{Result, StartFlownodeSnafu};
@@ -111,6 +133,144 @@ impl SubCommand {
     }
 }
 
+#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
+#[serde(default)]
+pub struct StandaloneOptions {
+    pub enable_telemetry: bool,
+    pub default_timezone: Option<String>,
+    pub http: HttpOptions,
+    pub grpc: GrpcOptions,
+    pub mysql: MysqlOptions,
+    pub postgres: PostgresOptions,
+    pub opentsdb: OpentsdbOptions,
+    pub influxdb: InfluxdbOptions,
+    pub jaeger: JaegerOptions,
+    pub prom_store: PromStoreOptions,
+    pub wal: DatanodeWalConfig,
+    pub storage: StorageConfig,
+    pub metadata_store: KvBackendConfig,
+    pub procedure: ProcedureConfig,
+    pub flow: FlowConfig,
+    pub logging: LoggingOptions,
+    pub user_provider: Option<String>,
+    /// Options for different store engines.
+    pub region_engine: Vec<RegionEngineConfig>,
+    pub export_metrics: ExportMetricsOption,
+    pub tracing: TracingOptions,
+    pub init_regions_in_background: bool,
+    pub init_regions_parallelism: usize,
+    pub max_in_flight_write_bytes: Option<ReadableSize>,
+    pub slow_query: SlowQueryOptions,
+    pub query: QueryOptions,
+    pub memory: MemoryOptions,
+}
+
+impl Default for StandaloneOptions {
+    fn default() -> Self {
+        Self {
+            enable_telemetry: true,
+            default_timezone: None,
+            http: HttpOptions::default(),
+            grpc: GrpcOptions::default(),
+            mysql: MysqlOptions::default(),
+            postgres: PostgresOptions::default(),
+            opentsdb: OpentsdbOptions::default(),
+            influxdb: InfluxdbOptions::default(),
+            jaeger: JaegerOptions::default(),
+            prom_store: PromStoreOptions::default(),
+            wal: DatanodeWalConfig::default(),
+            storage: StorageConfig::default(),
+            metadata_store: KvBackendConfig::default(),
+            procedure: ProcedureConfig::default(),
+            flow: FlowConfig::default(),
+            logging: LoggingOptions::default(),
+            export_metrics: ExportMetricsOption::default(),
+            user_provider: None,
+            region_engine: vec![
+                RegionEngineConfig::Mito(MitoConfig::default()),
+                RegionEngineConfig::File(FileEngineConfig::default()),
+            ],
+            tracing: TracingOptions::default(),
+            init_regions_in_background: false,
+            init_regions_parallelism: 16,
+            max_in_flight_write_bytes: None,
+            slow_query: SlowQueryOptions::default(),
+            query: QueryOptions::default(),
+            memory: MemoryOptions::default(),
+        }
+    }
+}
+
+impl Configurable for StandaloneOptions {
+    fn env_list_keys() -> Option<&'static [&'static str]> {
+        Some(&["wal.broker_endpoints"])
+    }
+}
+
+/// The [`StandaloneOptions`] is only defined in cmd crate,
+/// we don't want to make `frontend` depends on it, so impl [`Into`]
+/// rather than [`From`].
+#[allow(clippy::from_over_into)]
+impl Into<FrontendOptions> for StandaloneOptions {
+    fn into(self) -> FrontendOptions {
+        self.frontend_options()
+    }
+}
+
+impl StandaloneOptions {
+    /// Returns the `FrontendOptions` for the standalone instance.
+    pub fn frontend_options(&self) -> FrontendOptions {
+        let cloned_opts = self.clone();
+        FrontendOptions {
+            default_timezone: cloned_opts.default_timezone,
+            http: cloned_opts.http,
+            grpc: cloned_opts.grpc,
+            mysql: cloned_opts.mysql,
+            postgres: cloned_opts.postgres,
+            opentsdb: cloned_opts.opentsdb,
+            influxdb: cloned_opts.influxdb,
+            jaeger: cloned_opts.jaeger,
+            prom_store: cloned_opts.prom_store,
+            meta_client: None,
+            logging: cloned_opts.logging,
+            user_provider: cloned_opts.user_provider,
+            // Handle the export metrics task run by standalone to frontend for execution
+            export_metrics: cloned_opts.export_metrics,
+            max_in_flight_write_bytes: cloned_opts.max_in_flight_write_bytes,
+            slow_query: cloned_opts.slow_query,
+            ..Default::default()
+        }
+    }
+
+    /// Returns the `DatanodeOptions` for the standalone instance.
+    pub fn datanode_options(&self) -> DatanodeOptions {
+        let cloned_opts = self.clone();
+        DatanodeOptions {
+            node_id: Some(0),
+            enable_telemetry: cloned_opts.enable_telemetry,
+            wal: cloned_opts.wal,
+            storage: cloned_opts.storage,
+            region_engine: cloned_opts.region_engine,
+            grpc: cloned_opts.grpc,
+            init_regions_in_background: cloned_opts.init_regions_in_background,
+            init_regions_parallelism: cloned_opts.init_regions_parallelism,
+            query: cloned_opts.query,
+            ..Default::default()
+        }
+    }
+
+    /// Sanitize the `StandaloneOptions` to ensure the config is valid.
+    pub fn sanitize(&mut self) {
+        if self.storage.is_object_storage() {
+            self.storage
+                .store
+                .cache_config_mut()
+                .unwrap()
+                .sanitize(&self.storage.data_home);
+        }
+    }
+}
+
 pub struct Instance {
     datanode: Datanode,
     frontend: Frontend,
@@ -356,10 +516,6 @@ impl StartCommand {
         let mut plugins = Plugins::new();
         let plugin_opts = opts.plugins;
         let mut opts = opts.component;
-        set_default_prefix(opts.default_column_prefix.as_deref())
-            .map_err(BoxedError::new)
-            .context(error::BuildCliSnafu)?;
-
         opts.grpc.detect_server_addr();
         let fe_opts = opts.frontend_options();
         let dn_opts = opts.datanode_options();
@@ -381,14 +537,13 @@ impl StartCommand {
             .context(error::CreateDirSnafu { dir: data_home })?;
 
         let metadata_dir = metadata_store_dir(data_home);
-        let kv_backend = standalone::build_metadata_kvbackend(metadata_dir, opts.metadata_store)
+        let (kv_backend, procedure_manager) = FeInstance::try_build_standalone_components(
-            .context(error::BuildMetadataKvbackendSnafu)?;
+            metadata_dir,
-        let procedure_manager =
+            opts.metadata_store,
-            standalone::build_procedure_manager(kv_backend.clone(), opts.procedure);
+            opts.procedure,
-
+        )
-        plugins::setup_standalone_plugins(&mut plugins, &plugin_opts, &opts, kv_backend.clone())
+        .await
-            .await
+        .context(error::StartFrontendSnafu)?;
-            .context(error::SetupStandalonePluginsSnafu)?;
 
         // Builds cache registry
         let layered_cache_builder = LayeredCacheRegistryBuilder::default();
@@ -410,8 +565,6 @@ impl StartCommand {
             procedure_manager.clone(),
         ));
 
-        plugins.insert::<InformationExtensionRef>(information_extension.clone());
-
         let process_manager = Arc::new(ProcessManager::new(opts.grpc.server_addr.clone(), None));
         let builder = KvBackendCatalogManagerBuilder::new(
             information_extension.clone(),
@@ -481,11 +634,7 @@ impl StartCommand {
                 .step(10)
                 .build(),
         );
-        let kafka_options = opts
+        let kafka_options = opts.wal.clone().into();
-            .wal
-            .clone()
-            .try_into()
-            .context(error::InvalidWalProviderSnafu)?;
         let wal_options_allocator = build_wal_options_allocator(&kafka_options, kv_backend.clone())
             .await
             .context(error::BuildWalOptionsAllocatorSnafu)?;
@@ -610,6 +759,141 @@ impl StartCommand {
     }
 }
 
+pub struct StandaloneInformationExtension {
+    region_server: RegionServer,
+    procedure_manager: ProcedureManagerRef,
+    start_time_ms: u64,
+    flow_streaming_engine: RwLock<Option<Arc<StreamingEngine>>>,
+}
+
+impl StandaloneInformationExtension {
+    pub fn new(region_server: RegionServer, procedure_manager: ProcedureManagerRef) -> Self {
+        Self {
+            region_server,
+            procedure_manager,
+            start_time_ms: common_time::util::current_time_millis() as u64,
+            flow_streaming_engine: RwLock::new(None),
+        }
+    }
+
+    /// Set the flow streaming engine for the standalone instance.
+    pub async fn set_flow_streaming_engine(&self, flow_streaming_engine: Arc<StreamingEngine>) {
+        let mut guard = self.flow_streaming_engine.write().await;
+        *guard = Some(flow_streaming_engine);
+    }
+}
+
+#[async_trait::async_trait]
+impl InformationExtension for StandaloneInformationExtension {
+    type Error = catalog::error::Error;
+
+    async fn nodes(&self) -> std::result::Result<Vec<NodeInfo>, Self::Error> {
+        let build_info = common_version::build_info();
+        let node_info = NodeInfo {
+            // For the standalone:
+            // - id always 0
+            // - empty string for peer_addr
+            peer: Peer {
+                id: 0,
+                addr: "".to_string(),
+            },
+            last_activity_ts: -1,
+            status: NodeStatus::Standalone,
+            version: build_info.version.to_string(),
+            git_commit: build_info.commit_short.to_string(),
+            // Use `self.start_time_ms` instead.
+            // It's not precise but enough.
+            start_time_ms: self.start_time_ms,
+            cpus: common_config::utils::get_cpus() as u32,
+            memory_bytes: common_config::utils::get_sys_total_memory()
+                .unwrap_or_default()
+                .as_bytes(),
+        };
+        Ok(vec![node_info])
+    }
+
+    async fn procedures(&self) -> std::result::Result<Vec<(String, ProcedureInfo)>, Self::Error> {
+        self.procedure_manager
+            .list_procedures()
+            .await
+            .map_err(BoxedError::new)
+            .map(|procedures| {
+                procedures
+                    .into_iter()
+                    .map(|procedure| {
+                        let status = procedure.state.as_str_name().to_string();
+                        (status, procedure)
+                    })
+                    .collect::<Vec<_>>()
+            })
+            .context(catalog::error::ListProceduresSnafu)
+    }
+
+    async fn region_stats(&self) -> std::result::Result<Vec<RegionStat>, Self::Error> {
+        let stats = self
+            .region_server
+            .reportable_regions()
+            .into_iter()
+            .map(|stat| {
+                let region_stat = self
+                    .region_server
+                    .region_statistic(stat.region_id)
+                    .unwrap_or_default();
+                RegionStat {
+                    id: stat.region_id,
+                    rcus: 0,
+                    wcus: 0,
+                    approximate_bytes: region_stat.estimated_disk_size(),
+                    engine: stat.engine,
+                    role: RegionRole::from(stat.role).into(),
+                    num_rows: region_stat.num_rows,
+                    memtable_size: region_stat.memtable_size,
+                    manifest_size: region_stat.manifest_size,
+                    sst_size: region_stat.sst_size,
+                    sst_num: region_stat.sst_num,
+                    index_size: region_stat.index_size,
+                    region_manifest: region_stat.manifest.into(),
+                    data_topic_latest_entry_id: region_stat.data_topic_latest_entry_id,
+                    metadata_topic_latest_entry_id: region_stat.metadata_topic_latest_entry_id,
+                    written_bytes: region_stat.written_bytes,
+                }
+            })
+            .collect::<Vec<_>>();
+        Ok(stats)
+    }
+
+    async fn flow_stats(&self) -> std::result::Result<Option<FlowStat>, Self::Error> {
+        Ok(Some(
+            self.flow_streaming_engine
+                .read()
+                .await
+                .as_ref()
+                .unwrap()
+                .gen_state_report()
+                .await,
+        ))
+    }
+
+    async fn inspect_datanode(
+        &self,
+        request: DatanodeInspectRequest,
+    ) -> std::result::Result<SendableRecordBatchStream, Self::Error> {
+        let req = QueryRequest {
+            plan: request
+                .build_plan()
+                .context(catalog::error::DatafusionSnafu)?,
+            region_id: RegionId::default(),
+            header: None,
+        };
+
+        self.region_server
+            .handle_read(req)
+            .await
+            .map_err(BoxedError::new)
+            .context(catalog::error::InternalSnafu)
+    }
+}
+
 #[cfg(test)]
 mod tests {
     use std::default::Default;
@@ -621,9 +905,7 @@ mod tests {
     use common_config::ENV_VAR_SEP;
     use common_test_util::temp_dir::create_named_temp_file;
     use common_wal::config::DatanodeWalConfig;
-    use frontend::frontend::FrontendOptions;
     use object_store::config::{FileConfig, GcsConfig};
-    use servers::grpc::GrpcOptions;
 
     use super::*;
     use crate::options::GlobalOptions;
@@ -753,7 +1035,7 @@ mod tests {
             object_store::config::ObjectStoreConfig::S3(s3_config) => {
                 assert_eq!(
                     "SecretBox<alloc::string::String>([REDACTED])".to_string(),
-                    format!("{:?}", s3_config.connection.access_key_id)
+                    format!("{:?}", s3_config.access_key_id)
                 );
             }
             _ => {

src/cmd/tests/load_config_test.rs

@@ -15,6 +15,7 @@
 use std::time::Duration;
 
 use cmd::options::GreptimeOptions;
+use cmd::standalone::StandaloneOptions;
 use common_config::{Configurable, DEFAULT_DATA_HOME};
 use common_options::datanode::{ClientOptions, DatanodeClientOptions};
 use common_telemetry::logging::{DEFAULT_LOGGING_DIR, DEFAULT_OTLP_HTTP_ENDPOINT, LoggingOptions};
@@ -34,7 +35,6 @@ use servers::export_metrics::ExportMetricsOption;
 use servers::grpc::GrpcOptions;
 use servers::http::HttpOptions;
 use servers::tls::{TlsMode, TlsOption};
-use standalone::options::StandaloneOptions;
 use store_api::path_utils::WAL_DIR;
 
 #[allow(deprecated)]
@@ -48,7 +48,6 @@ fn test_load_datanode_example_config() {
     let expected = GreptimeOptions::<DatanodeOptions> {
         component: DatanodeOptions {
             node_id: Some(42),
-            default_column_prefix: Some("greptime".to_string()),
             meta_client: Some(MetaClientOptions {
                 metasrv_addrs: vec!["127.0.0.1:3002".to_string()],
                 timeout: Duration::from_secs(3),
@@ -114,7 +113,6 @@ fn test_load_frontend_example_config() {
     let expected = GreptimeOptions::<FrontendOptions> {
         component: FrontendOptions {
             default_timezone: Some("UTC".to_string()),
-            default_column_prefix: Some("greptime".to_string()),
             meta_client: Some(MetaClientOptions {
                 metasrv_addrs: vec!["127.0.0.1:3002".to_string()],
                 timeout: Duration::from_secs(3),
@@ -145,11 +143,9 @@ fn test_load_frontend_example_config() {
                 remote_write: Some(Default::default()),
                 ..Default::default()
             },
-            grpc: GrpcOptions {
+            grpc: GrpcOptions::default()
-                bind_addr: "127.0.0.1:4001".to_string(),
+                .with_bind_addr("127.0.0.1:4001")
-                server_addr: "127.0.0.1:4001".to_string(),
+                .with_server_addr("127.0.0.1:4001"),
-                ..Default::default()
-            },
             internal_grpc: Some(GrpcOptions::internal_default()),
             http: HttpOptions {
                 cors_allowed_origins: vec!["https://example.com".to_string()],
@@ -275,7 +271,6 @@ fn test_load_standalone_example_config() {
     let expected = GreptimeOptions::<StandaloneOptions> {
         component: StandaloneOptions {
             default_timezone: Some("UTC".to_string()),
-            default_column_prefix: Some("greptime".to_string()),
             wal: DatanodeWalConfig::RaftEngine(RaftEngineConfig {
                 dir: Some(format!("{}/{}", DEFAULT_DATA_HOME, WAL_DIR)),
                 sync_period: Some(Duration::from_secs(10)),

src/common/base/Cargo.toml

@@ -18,11 +18,9 @@ bytes.workspace = true
 common-error.workspace = true
 common-macro.workspace = true
 futures.workspace = true
-lazy_static.workspace = true
 paste.workspace = true
 pin-project.workspace = true
 rand.workspace = true
-regex.workspace = true
 serde = { version = "1.0", features = ["derive"] }
 snafu.workspace = true
 tokio.workspace = true

src/common/base/src/lib.rs

@@ -19,7 +19,6 @@ pub mod plugins;
 pub mod range_read;
 #[allow(clippy::all)]
 pub mod readable_size;
-pub mod regex_pattern;
 pub mod secrets;
 pub mod serde;
 

src/common/base/src/plugins.rs

@@ -75,11 +75,11 @@ impl Plugins {
         self.read().is_empty()
     }
 
-    fn read(&self) -> RwLockReadGuard<'_, SendSyncAnyMap> {
+    fn read(&self) -> RwLockReadGuard<SendSyncAnyMap> {
         self.inner.read().unwrap()
     }
 
-    fn write(&self) -> RwLockWriteGuard<'_, SendSyncAnyMap> {
+    fn write(&self) -> RwLockWriteGuard<SendSyncAnyMap> {
         self.inner.write().unwrap()
     }
 }

src/common/base/src/secrets.rs

@@ -31,7 +31,7 @@
 //! types of `SecretBox<T>` to be serializable with `serde`, you will need to impl
 //! the [`SerializableSecret`] marker trait on `T`
 
-use std::fmt::{Debug, Display};
+use std::fmt::Debug;
 use std::{any, fmt};
 
 use serde::{Deserialize, Serialize, de, ser};
@@ -46,12 +46,6 @@ impl From<String> for SecretString {
     }
 }
 
-impl Display for SecretString {
-    fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
-        write!(f, "SecretString([REDACTED])")
-    }
-}
-
 /// Wrapper type for values that contains secrets.
 ///
 /// It attempts to limit accidental exposure and ensure secrets are wiped from memory when dropped.
@@ -171,15 +165,6 @@ impl<S: Zeroize> ExposeSecretMut<S> for SecretBox<S> {
     }
 }
 
-impl<S> PartialEq for SecretBox<S>
-where
-    S: PartialEq + Zeroize,
-{
-    fn eq(&self, other: &Self) -> bool {
-        self.inner_secret == other.inner_secret
-    }
-}
-
 /// Expose a reference to an inner secret
 pub trait ExposeSecret<S> {
     /// Expose secret: this is the only method providing access to a secret.

src/common/catalog/Cargo.toml

@@ -8,6 +8,5 @@ license.workspace = true
 workspace = true
 
 [dependencies]
-const_format.workspace = true
 
 [dev-dependencies]