feat(core): Support configuration metadata database name in .env file (#741)

*Others*: - Support disable alembic to initialize and upgrade database metadata with parameter `--disable_alembic_upgrade` - New APP ID parameter `proxy_api_app_id` for proxy LLM
2025-09-17 15:10:14 +00:00 · 2023-10-27 18:52:00 +08:00
parent da3f795d5d 0e273c68a8
commit f6a89ad605
24 changed files with 396 additions and 135 deletions
--- a/.env.template
+++ b/.env.template
@@ -84,6 +84,7 @@ LOCAL_DB_TYPE=sqlite
 # LOCAL_DB_PASSWORD=aa12345678
 # LOCAL_DB_HOST=127.0.0.1
 # LOCAL_DB_PORT=3306
+# LOCAL_DB_NAME=dbgpt
 ### This option determines the storage location of conversation records. The default is not configured to the old version of duckdb. It can be optionally db or file (if the value is db, the database configured by LOCAL_DB will be used)
 #CHAT_HISTORY_STORE_TYPE=db

--- a/assets/schema/history.sql
+++ b/assets/schema/history.sql
@@ -1,18 +0,0 @@
-CREATE DATABASE history;
-use history;
-CREATE TABLE `chat_feed_back` (
-  `id` bigint(20) NOT NULL AUTO_INCREMENT,
-  `conv_uid` varchar(128) DEFAULT NULL COMMENT '会话id',
-  `conv_index` int(4) DEFAULT NULL COMMENT '第几轮会话',
-  `score` int(1) DEFAULT NULL COMMENT '评分',
-  `ques_type` varchar(32) DEFAULT NULL COMMENT '用户问题类别',
-  `question` longtext DEFAULT NULL COMMENT '用户问题',
-  `knowledge_space` varchar(128) DEFAULT NULL COMMENT '知识库',
-  `messages` longtext DEFAULT NULL COMMENT '评价详情',
-  `user_name` varchar(128) DEFAULT NULL COMMENT '评价人',
-  `gmt_created` datetime DEFAULT NULL,
-  `gmt_modified` datetime DEFAULT NULL,
-  PRIMARY KEY (`id`),
-  UNIQUE KEY `uk_conv` (`conv_uid`,`conv_index`),
-  KEY `idx_conv` (`conv_uid`,`conv_index`)
-) ENGINE=InnoDB AUTO_INCREMENT=0 DEFAULT CHARSET=utf8mb4 COMMENT='用户评分反馈表';
--- a/assets/schema/knowledge_management.sql
+++ b/assets/schema/knowledge_management.sql
@@ -1,5 +1,15 @@
-CREATE DATABASE knowledge_management;
-use knowledge_management;
+-- You can change `dbgpt` to your actual metadata database name in your `.env` file
+-- eg. `LOCAL_DB_NAME=dbgpt`
+
+CREATE DATABASE IF NOT EXISTS dbgpt;
+use dbgpt;
+
+-- For alembic migration tool
+CREATE TABLE `alembic_version` (
+      version_num VARCHAR(32) NOT NULL,
+      CONSTRAINT alembic_version_pkc PRIMARY KEY (version_num)
+);
+
 CREATE TABLE `knowledge_space` (
  `id` int NOT NULL AUTO_INCREMENT COMMENT 'auto increment id',
  `name` varchar(100) NOT NULL COMMENT 'knowledge space name',
@@ -43,6 +53,102 @@ CREATE TABLE `document_chunk` (
  KEY `idx_document_id` (`document_id`) COMMENT 'index:document_id'
 ) ENGINE=InnoDB AUTO_INCREMENT=100001 DEFAULT CHARSET=utf8mb4 COMMENT='knowledge document chunk detail';

+
+
+CREATE TABLE `connect_config` (
+  `id` int NOT NULL AUTO_INCREMENT COMMENT 'autoincrement id',
+  `db_type` varchar(255) NOT NULL COMMENT 'db type',
+  `db_name` varchar(255) NOT NULL COMMENT 'db name',
+  `db_path` varchar(255) DEFAULT NULL COMMENT 'file db path',
+  `db_host` varchar(255) DEFAULT NULL COMMENT 'db connect host(not file db)',
+  `db_port` varchar(255) DEFAULT NULL COMMENT 'db cnnect port(not file db)',
+  `db_user` varchar(255) DEFAULT NULL COMMENT 'db user',
+  `db_pwd` varchar(255) DEFAULT NULL COMMENT 'db password',
+  `comment` text COMMENT 'db comment',
+  PRIMARY KEY (`id`),
+  UNIQUE KEY `uk_db` (`db_name`),
+  KEY `idx_q_db_type` (`db_type`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COMMENT 'Connection confi';
+
+CREATE TABLE `chat_history` (
+  `id` int NOT NULL AUTO_INCREMENT COMMENT 'autoincrement id',
+  `conv_uid` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'Conversation record unique id',
+  `chat_mode` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'Conversation scene mode',
+  `summary` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'Conversation record summary',
+  `user_name` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'interlocutor',
+  `messages` text COLLATE utf8mb4_unicode_ci COMMENT 'Conversation details',
+  PRIMARY KEY (`id`)
+) ENGINE=InnoDB AUTO_INCREMENT=2 DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci COMMENT 'Chat history';
+
+CREATE TABLE `chat_feed_back` (
+  `id` bigint(20) NOT NULL AUTO_INCREMENT,
+  `conv_uid` varchar(128) DEFAULT NULL COMMENT 'Conversation ID',
+  `conv_index` int(4) DEFAULT NULL COMMENT 'Round of conversation',
+  `score` int(1) DEFAULT NULL COMMENT 'Score of user',
+  `ques_type` varchar(32) DEFAULT NULL COMMENT 'User question category',
+  `question` longtext DEFAULT NULL COMMENT 'User question',
+  `knowledge_space` varchar(128) DEFAULT NULL COMMENT 'Knowledge space name',
+  `messages` longtext DEFAULT NULL COMMENT 'The details of user feedback',
+  `user_name` varchar(128) DEFAULT NULL COMMENT 'User name',
+  `gmt_created` timestamp NULL DEFAULT CURRENT_TIMESTAMP COMMENT 'created time',
+  `gmt_modified` timestamp NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT 'update time',
+  PRIMARY KEY (`id`),
+  UNIQUE KEY `uk_conv` (`conv_uid`,`conv_index`),
+  KEY `idx_conv` (`conv_uid`,`conv_index`)
+) ENGINE=InnoDB AUTO_INCREMENT=0 DEFAULT CHARSET=utf8mb4 COMMENT='User feedback table';
+
+
+CREATE TABLE `my_plugin` (
+  `id` int NOT NULL AUTO_INCREMENT COMMENT 'autoincrement id',
+  `tenant` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'user tenant',
+  `user_code` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'user code',
+  `user_name` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'user name',
+  `name` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'plugin name',
+  `file_name` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'plugin package file name',
+  `type` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin type',
+  `version` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin version',
+  `use_count` int DEFAULT NULL COMMENT 'plugin total use count',
+  `succ_count` int DEFAULT NULL COMMENT 'plugin total success count',
+  `gmt_created` TIMESTAMP DEFAULT CURRENT_TIMESTAMP COMMENT 'plugin install time',
+  PRIMARY KEY (`id`),
+  UNIQUE KEY `name` (`name`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci COMMENT='User plugin table';
+
+CREATE TABLE `plugin_hub` (
+  `id` int NOT NULL AUTO_INCREMENT COMMENT 'autoincrement id',
+  `name` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'plugin name',
+  `description` varchar(255) COLLATE utf8mb4_unicode_ci NOT NULL COMMENT 'plugin description',
+  `author` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin author',
+  `email` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin author email',
+  `type` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin type',
+  `version` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin version',
+  `storage_channel` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin storage channel',
+  `storage_url` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin download url',
+  `download_param` varchar(255) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'plugin download param',
+  `gmt_created` TIMESTAMP DEFAULT CURRENT_TIMESTAMP COMMENT 'plugin upload time',
+  `installed` int DEFAULT NULL COMMENT 'plugin already installed count',
+  PRIMARY KEY (`id`),
+  UNIQUE KEY `name` (`name`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci COMMENT='Plugin Hub table';
+
+
+CREATE TABLE `prompt_manage` (
+  `id` int(11) NOT NULL AUTO_INCREMENT,
+  `chat_scene` varchar(100) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'Chat scene',
+  `sub_chat_scene` varchar(100) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'Sub chat scene',
+  `prompt_type` varchar(100) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'Prompt type: common or private',
+  `prompt_name` varchar(512) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'prompt name',
+  `content` longtext COLLATE utf8mb4_unicode_ci COMMENT 'Prompt content',
+  `user_name` varchar(128) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'User name',
+  `gmt_created` timestamp NULL DEFAULT CURRENT_TIMESTAMP COMMENT 'created time',
+  `gmt_modified` timestamp NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP COMMENT 'update time',
+  PRIMARY KEY (`id`),
+  UNIQUE KEY `prompt_name_uiq` (`prompt_name`),
+  KEY `gmt_created_idx` (`gmt_created`)
+) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci COMMENT='Prompt management table';
+
+
+
 CREATE DATABASE EXAMPLE_1;
 use EXAMPLE_1;
 CREATE TABLE `users` (
--- a/assets/schema/prompt_management.sql
+++ b/assets/schema/prompt_management.sql
@@ -1,16 +0,0 @@
-CREATE DATABASE prompt_management;
-use prompt_management;
-CREATE TABLE `prompt_manage` (
-  `id` int(11) NOT NULL AUTO_INCREMENT,
-  `chat_scene` varchar(100) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT '场景',
-  `sub_chat_scene` varchar(100) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT '子场景',
-  `prompt_type` varchar(100) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT '类型： common or private',
-  `prompt_name` varchar(512) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT 'prompt的名字',
-  `content` longtext COLLATE utf8mb4_unicode_ci COMMENT 'prompt的内容',
-  `user_name` varchar(128) COLLATE utf8mb4_unicode_ci DEFAULT NULL COMMENT '用户名',
-  `gmt_created` datetime DEFAULT NULL,
-  `gmt_modified` datetime DEFAULT NULL,
-  PRIMARY KEY (`id`),
-  UNIQUE KEY `prompt_name_uiq` (`prompt_name`),
-  KEY `gmt_created_idx` (`gmt_created`)
-) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_unicode_ci COMMENT='prompt管理表';
--- a/docs/getting_started/faq/deploy/deploy_faq.md
+++ b/docs/getting_started/faq/deploy/deploy_faq.md
@@ -100,8 +100,49 @@ pip install --use-pep517 fschat
 ```

 ##### Q9: alembic.util.exc.CommandError: Target database is not up to date.
-delete files in `DB-GPT/pilot/meta_data/alembic/versions/` and reboot.
+
+delete files in `DB-GPT/pilot/meta_data/alembic/versions/` and restart.
 ```commandline
 rm -rf DB-GPT/pilot/meta_data/alembic/versions/*
 rm -rf DB-GPT/pilot/meta_data/alembic/dbgpt.db
 ```
+
+##### Q10: How to store DB-GPT metadata into my database
+
+In version 0.4.0, the metadata module of the DB-GPT application has been refactored. All metadata tables will now be automatically saved in the 'dbgpt' database, based on the database type specified in the `.env` file. If you would like to retain the existing data, it is recommended to use a data migration tool to transfer the database table information to the 'dbgpt' database. Additionally, you can change the default database name 'dbgpt' in your `.env` file.
+
+```commandline
+### SQLite database (Current default database)
+#LOCAL_DB_PATH=data/default_sqlite.db
+#LOCAL_DB_TYPE=sqlite
+
+### Mysql database
+LOCAL_DB_TYPE=mysql
+LOCAL_DB_USER=root
+LOCAL_DB_PASSWORD=aa12345678
+LOCAL_DB_HOST=127.0.0.1
+LOCAL_DB_PORT=3306
+# You can change it to your actual metadata database name
+LOCAL_DB_NAME=dbgpt
+
+### This option determines the storage location of conversation records. The default is not configured to the old version of duckdb. It can be optionally db or file (if the value is db, the database configured by LOCAL_DB will be used)
+CHAT_HISTORY_STORE_TYPE=db
+```
+
+##### Q11: pymysql.err.OperationalError: (1142, "ALTER command denied to user '{you db user}'@'{you db host}' for table '{some table name}'")
+
+In version 0.4.0, DB-GPT use migration tool alembic to migrate metadata. If the database user does not have DDL permissions, this error will be reported. You can solve this problem by importing the metadata information separately.
+
+1. Use a privileged user to execute DDL sql file
+```bash
+mysql -h127.0.0.1 -uroot -paa12345678 < ./assets/schema/knowledge_management.sql
+```
+
+2. Run DB-GPT webserver with `--disable_alembic_upgrade`
+```bash
+python pilot/server/dbgpt_server.py --disable_alembic_upgrade
+```
+or 
+```bash
+dbgpt start webserver --disable_alembic_upgrade
+```
--- a/docs/getting_started/faq/kbqa/kbqa_faq.md
+++ b/docs/getting_started/faq/kbqa/kbqa_faq.md
@@ -72,7 +72,7 @@ $ mysql -h127.0.0.1 -uroot -paa12345678 < ./assets/schema/knowledge_management.s

 ##### Q6:when pull from 0.4.0, I found historical knowledge document disappeared

-In version 0.4.0, the metadata module of the DB-GPT application has been refactored. All metadata tables will now be automatically saved in the 'dbgpt' database, based on the database type specified in the .env file. If you would like to retain the existing data, it is recommended to use a data migration tool to transfer the database table information to the 'dbgpt' database.
+In version 0.4.0, the metadata module of the DB-GPT application has been refactored. All metadata tables will now be automatically saved in the 'dbgpt' database, based on the database type specified in the `.env` file. If you would like to retain the existing data, it is recommended to use a data migration tool to transfer the database table information to the 'dbgpt' database. Additionally, you can change the default database name 'dbgpt' in your `.env` file.

 ```{tip}
 old database:knowledge_management;
@@ -89,5 +89,6 @@ LOCAL_DB_USER=root
 LOCAL_DB_PASSWORD=aa12345678
 LOCAL_DB_HOST=127.0.0.1
 LOCAL_DB_PORT=3306
-
+# You can change it to your actual metadata database name
+LOCAL_DB_NAME=dbgpt
 ```
--- a/docs/locales/zh_CN/LC_MESSAGES/getting_started/faq/deploy/deploy_faq.po
+++ b/docs/locales/zh_CN/LC_MESSAGES/getting_started/faq/deploy/deploy_faq.po
@@ -8,7 +8,7 @@ msgid ""
 msgstr ""
 "Project-Id-Version: DB-GPT 👏👏 0.3.5\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2023-10-19 19:31+0800\n"
+"POT-Creation-Date: 2023-10-27 15:57+0800\n"
 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
 "Language: zh_CN\n"
@@ -20,12 +20,12 @@ msgstr ""
 "Generated-By: Babel 2.12.1\n"

 #: ../../getting_started/faq/deploy/deploy_faq.md:1
-#: fb640f7c38744cbf996dcf7f73f325f6
+#: 798fb40c5ec941fcb9d6a0795219132f
 msgid "Installation FAQ"
-msgstr "Installation FAQ"
+msgstr "安装 FAQ"

 #: ../../getting_started/faq/deploy/deploy_faq.md:5
-#: 79fd80e469d14d608554d53a0e0ed2e3
+#: 47d0aa43c5fe4ca3a8ceba50c18ba608
 #, fuzzy
 msgid ""
 "Q1: execute `pip install -e .` error, found some package cannot find "
@@ -35,18 +35,20 @@ msgstr ""
 "cannot find correct version."

 #: ../../getting_started/faq/deploy/deploy_faq.md:6
-#: f1f6e3291d1446b5bbcf744cd4c4e89a
+#: 944761c1ccc543c0a6aa2fad8dc74a32
 msgid "change the pip source."
 msgstr "替换pip源."

 #: ../../getting_started/faq/deploy/deploy_faq.md:13
 #: ../../getting_started/faq/deploy/deploy_faq.md:20
-#: 68e1b39a08774a81b9061cc5205e4c1c dd34901f446749e998cd34ec5b6c44f4
+#: ../../getting_started/faq/deploy/deploy_faq.md:145
+#: 3cff7ea0ee7945be8d03b6b3b032515b 5ba3037287524d6384ca96ffe58798fa
+#: 9635f37d34e04764855f21d2266411f6
 msgid "or"
 msgstr "或者"

 #: ../../getting_started/faq/deploy/deploy_faq.md:27
-#: 0899f0e28dae443b8f912d96c797b79c
+#: c1c71ca902d745b89136bb63beda3dfd
 msgid ""
 "Q2: sqlalchemy.exc.OperationalError: (sqlite3.OperationalError) unable to"
 " open database file"
@@ -55,80 +57,80 @@ msgstr ""
 " open database file"

 #: ../../getting_started/faq/deploy/deploy_faq.md:29
-#: 3e60d8190e49436b8c40b34a67b7bfb3
+#: 97124a4512534c63bd09f2cf5a76fd13
 msgid "make sure you pull latest code or create directory with mkdir pilot/data"
 msgstr "make sure you pull latest code or create directory with mkdir pilot/data"

 #: ../../getting_started/faq/deploy/deploy_faq.md:31
-#: baeaae20238842d3b8e4ae5b337198e5
+#: 369ed2cd489d46009184036a8f8ed67a
 msgid "Q3: The model keeps getting killed."
 msgstr "Q3: The model keeps getting killed."

 #: ../../getting_started/faq/deploy/deploy_faq.md:33
-#: eb3936307ad64b19b73483ff9ae126f2
+#: 6d59ca711a95495d9bddf22cd804e20b
 msgid ""
 "your GPU VRAM size is not enough, try replace your hardware or replace "
 "other llms."
 msgstr "GPU显存不够, 增加显存或者换一个显存小的模型"

 #: ../../getting_started/faq/deploy/deploy_faq.md:35
-#: f6dba770717041699c73b4cd00d48aad
+#: 7ef755bf77fa46ccb63076c3561ecc64
 msgid "Q4: How to access website on the public network"
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:37
-#: 447d9e9374de44bab6d8a03f2c936676
+#: cd3f9144525b49babb826a7447812016
 msgid ""
 "You can try to use gradio's [network](https://github.com/gradio-"
 "app/gradio/blob/main/gradio/networking.py) to achieve."
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:48
-#: 5e34dd4dfcf34feeb1815dfa974041d0
+#: 71f0174d58674b1abd3d6a02cf65abf6
 msgid "Open `url` with your browser to see the website."
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:50
-#: aaef774ce6124021a3862bc0a25d465f
+#: 81b07e64feef4187beab2022f3af294d
 msgid "Q5: (Windows) execute `pip install -e .` error"
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:52
-#: ec3945df451c4ec2b32ebb476f45c82b
+#: f2e7cd453c10486aa9b7d90d1d771b58
 msgid "The error log like the following:"
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:71
-#: 1df09f6d9f9b4c1a8a32d6e271e5ee39
+#: 2c97bbe2f96142ec8398b376f6a21d7f
 msgid ""
 "Download and install `Microsoft C++ Build Tools` from [visual-cpp-build-"
 "tools](https://visualstudio.microsoft.com/visual-cpp-build-tools/)"
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:75
-#: 251f47bfa5694242a1c9d81a2022b7a0
+#: 95120da5a6bf4a26bf64c2dd54632e4b
 msgid "Q6: `Torch not compiled with CUDA enabled`"
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:82
-#: bc9dfdfc47924a0e8d3ec535e23bf923
+#: fa936391d8bd44cebeffc92e0f893700
 msgid "Install [CUDA Toolkit](https://developer.nvidia.com/cuda-toolkit-archive)"
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:83
-#: b5a632baa42745bdbee5d6ba516d8d8b
+#: a8eb968b4b5a4f2786f7133299b8d20f
 msgid ""
 "Reinstall PyTorch [start-locally](https://pytorch.org/get-started/locally"
 "/#start-locally) with CUDA support."
 msgstr ""

 #: ../../getting_started/faq/deploy/deploy_faq.md:85
-#: 0092fb91642749f5a55b629017c0de6a
+#: 009f8b213c9044888975f1ae8cdf7a75
 msgid "Q7: ImportError: cannot import name 'PersistentClient' from 'chromadb'."
 msgstr "Q7: ImportError: cannot import name 'PersistentClient' from 'chromadb'."

 #: ../../getting_started/faq/deploy/deploy_faq.md:91
-#: 4aa87418f2a54c138bf3b7ff28a7e776
+#: 237706fe28b846dcbe77e04a3bf89a6c
 msgid ""
 "Q8: pydantic.error_wrappers.ValidationError:1 validation error for "
 "HuggingFaceEmbeddings.model_kwargs extra not permitted"
@@ -137,14 +139,62 @@ msgstr ""
 "HuggingFaceEmbeddings.model_kwargs extra not permitted"

 #: ../../getting_started/faq/deploy/deploy_faq.md:102
-#: 6b690ab272af44f6b126cfe5ce1435ef
+#: e20c5fde988b478fb7eaba0f10d7d196
 msgid "Q9: alembic.util.exc.CommandError: Target database is not up to date."
 msgstr ""

-#: ../../getting_started/faq/deploy/deploy_faq.md:103
-#: 223026d3b9124363b695937922d8f8d5
-msgid "delete files in `DB-GPT/pilot/meta_data/alembic/versions/` and reboot."
-msgstr "删除`DB-GPT/pilot/meta_data/alembic/versions/`目录下文件"
+#: ../../getting_started/faq/deploy/deploy_faq.md:104
+#: 6d096ff6f4754490868a0ff2b8a08f10
+msgid "delete files in `DB-GPT/pilot/meta_data/alembic/versions/` and restart."
+msgstr "删除`DB-GPT/pilot/meta_data/alembic/versions/`目录下文件然后重新启动"
+
+#: ../../getting_started/faq/deploy/deploy_faq.md:110
+#: 2294a811682d4744b9334ee6deec4a49
+msgid "Q10: How to store DB-GPT metadata into my database"
+msgstr "Q10: 如何将 DB-GPT 的元数据存储到自己的数据库中"
+
+#: ../../getting_started/faq/deploy/deploy_faq.md:112
+#: 034495af54f041bcb560a5366b9be912
+msgid ""
+"In version 0.4.0, the metadata module of the DB-GPT application has been "
+"refactored. All metadata tables will now be automatically saved in the "
+"'dbgpt' database, based on the database type specified in the `.env` "
+"file. If you would like to retain the existing data, it is recommended to"
+" use a data migration tool to transfer the database table information to "
+"the 'dbgpt' database. Additionally, you can change the default database "
+"name 'dbgpt' in your `.env` file."
+msgstr ""
+"v0.4.0 重构了DB-"
+"GPT应用的数据库元数据模块，所有的元数据库表都会自动保存在.env文件设置的数据库类型的`dbgpt`数据库中，如果想沿用以前的数据，建议使用数据迁移工具将数据库表信息挪到dbgpt数据库中。"
+"另外，你可以在 `.env` 中修改默认的数据库名 'dbgpt' "
+
+#: ../../getting_started/faq/deploy/deploy_faq.md:132
+#: f9baf853f21b460ba5df63b84b35c040
+msgid ""
+"Q11: pymysql.err.OperationalError: (1142, \"ALTER command denied to user "
+"'{you db user}'@'{you db host}' for table '{some table name}'\")"
+msgstr ""
+
+#: ../../getting_started/faq/deploy/deploy_faq.md:134
+#: a64cb2d75d6643559420c016362eb435
+msgid ""
+"In version 0.4.0, DB-GPT use migration tool alembic to migrate metadata. "
+"If the database user does not have DDL permissions, this error will be "
+"reported. You can solve this problem by importing the metadata "
+"information separately."
+msgstr ""
+"v0.4.0后，DB-GPT 使用 alembic 作为元数据迁移工具。"
+"如果数据库用户没有 DDL 权限则会报这个错，您可以通过单独导入元数据信息来解决这个问题。"
+
+#: ../../getting_started/faq/deploy/deploy_faq.md:136
+#: 77c6bd4b559d457ab599b7f7730e85f1
+msgid "Use a privileged user to execute DDL sql file"
+msgstr "使用一个有权限的用户来执行 DDL SQL 文件。"
+
+#: ../../getting_started/faq/deploy/deploy_faq.md:141
+#: 30c3fd7f7bfc4a63b5b9c4c15c64430f
+msgid "Run DB-GPT webserver with `--disable_alembic_upgrade`"
+msgstr "添加参数 `--disable_alembic_upgrade` 来运行 DB-GPT 的 webserver"

 #~ msgid ""
 #~ "Q2: When use Mysql, Access denied "
--- a/docs/locales/zh_CN/LC_MESSAGES/getting_started/faq/kbqa/kbqa_faq.po
+++ b/docs/locales/zh_CN/LC_MESSAGES/getting_started/faq/kbqa/kbqa_faq.po
@@ -8,7 +8,7 @@ msgid ""
 msgstr ""
 "Project-Id-Version: DB-GPT 👏👏 0.3.5\n"
 "Report-Msgid-Bugs-To: \n"
-"POT-Creation-Date: 2023-10-20 14:49+0800\n"
+"POT-Creation-Date: 2023-10-27 15:57+0800\n"
 "PO-Revision-Date: YEAR-MO-DA HO:MI+ZONE\n"
 "Last-Translator: FULL NAME <EMAIL@ADDRESS>\n"
 "Language: zh_CN\n"
@@ -20,34 +20,34 @@ msgstr ""
 "Generated-By: Babel 2.12.1\n"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:1
-#: e95c136d802f486082c47a8c017eb725
+#: ab7d87ee62774af099fb0a8167b2d4be
 msgid "KBQA FAQ"
 msgstr "KBQA FAQ"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:4
-#: f19c96b7b1ec4bc7ad8c7c26582d7e59
+#: 3582be98154f4c3381765c130538d997
 msgid "Q1: text2vec-large-chinese not found"
 msgstr "Q1: text2vec-large-chinese not found"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:6
-#: 24a0603be39d4418909c27f9a53b51e2
+#: 3382fb6d60b443029a04e183cb5449cf
 msgid ""
 "make sure you have download text2vec-large-chinese embedding model in "
 "right way"
 msgstr "确认下载text2vec-large-chinese模型姿势以及路径正确"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:18
-#: 356008dd415f4bdd9b0927d8ee073548
+#: 9fb65568d9cb4bf0be9371eae38ffe75
 msgid "Q2:How to change Vector DB Type in DB-GPT."
 msgstr "怎么修改向量数据库类型"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:20
-#: 42013bfb02624010b668bd244b33c977
+#: 1904c691985044acad9d1cd84a227250
 msgid "Update .env file and set VECTOR_STORE_TYPE."
 msgstr "怎样在.env文件设置VECTOR_STORE_TYPE"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:22
-#: 72b2bae15fea4e0d927bda68a8d0861d
+#: a6ebd5b32fac42fbbdb7e26aaeafd781
 msgid ""
 "DB-GPT currently support Chroma(Default), Milvus(>2.1), Weaviate vector "
 "database. If you want to change vector db, Update your .env, set your "
@@ -61,19 +61,19 @@ msgstr ""
 "://db-gpt.readthedocs.io/en/latest/modules/vector.html)"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:38
-#: 0a12e3a5319c4a86a1db68d615d6fb8e
+#: 8357695648cf4dcca7d8d2c6d0c48b0a
 msgid "Q3:When I use vicuna-13b, found some illegal character like this."
 msgstr "当使用vicuna系列模型时出现乱码。"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:43
-#: f89ffbf4697a4ccdb8030834c52b0473
+#: ff8cb9ecfd2d4c9fa179b872d03a97dd
 msgid ""
 "Set KNOWLEDGE_SEARCH_TOP_SIZE smaller or set KNOWLEDGE_CHUNK_SIZE "
 "smaller, and reboot server."
 msgstr "通过在.env文件将KNOWLEDGE_SEARCH_TOP_SIZE设置更小点或者在文档界面点击参数设置，将topk设置更小点"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:45
-#: fc837488bcc7432a92b70126e80e75d7
+#: d7a49b1b219c4cf1bcb6ba212e378a7e
 msgid ""
 "Q4:space add error (pymysql.err.OperationalError) (1054, \"Unknown column"
 " 'knowledge_space.context' in 'field list'\")"
@@ -82,53 +82,58 @@ msgstr ""
 "'knowledge_space.context' in 'field list'\")"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:47
-#: 30c186bae2c3489eb18c18768c11c302
+#: 30d34122c9fe4d1a9870c54480c734a6
 msgid "1.shutdown dbgpt_server(ctrl c)"
 msgstr "1.终止 dbgpt_server(ctrl c)"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:49
-#: 9cac5688ddb14c63905cc86e77d4567e
+#: 1ccc21a493114e399007f9399f98006b
 msgid "2.add column context for table knowledge_space"
 msgstr "2.新增列 `context` for table knowledge_space"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:55
-#: b32219bc1c6246108f606952d8ef0132
+#: 075c48b850354ac7971bd7556b68ef52
 msgid "3.execute sql ddl"
 msgstr "3.执行ddl"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:62
-#: ebfd196350994c44841d5766f776905c
+#: c4484fc646324c9a976b81a4e802c435
 msgid "4.restart dbgpt serve"
 msgstr "4.重启dbgpt server"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:64
-#: cfa083226efd4980a57ff15e86bb8480
+#: b3da50497f37421485d8f0e852f0b09c
 msgid "Q5:Use Mysql, how to use DB-GPT KBQA"
 msgstr "Q5:当使用 Mysql数据库时, 使用DB-GPT怎么初始化 KBQA service database schema"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:66
-#: 95098205d36c4ca79ad9b1b0f9b2985a
+#: 97484a97b0d14d3f835efb5a0739c97e
 msgid "build Mysql KBQA system database schema."
 msgstr "构建Mysql KBQA system database schema"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:73
-#: efc87678042d48b38b57f700d9ff74e5
+#: bdcbaf0559a549468fab021e03ac3876
 msgid "Q6:when pull from 0.4.0, I found historical knowledge document disappeared"
 msgstr "当从0.4.0版本拉取代码后，历史知识库问答信息没了"

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:75
-#: 975eaff1a20a40b5b5ee18d6c6ddb9c1
+#: ea78542f6be94dfc8d97b2660bb22876
+#, fuzzy
 msgid ""
 "In version 0.4.0, the metadata module of the DB-GPT application has been "
 "refactored. All metadata tables will now be automatically saved in the "
-"'dbgpt' database, based on the database type specified in the .env file. "
-"If you would like to retain the existing data, it is recommended to use a"
-" data migration tool to transfer the database table information to the "
-"'dbgpt' database."
-msgstr "v0.4.0 重构了DB-GPT应用的数据库元数据模块，所有的元数据库表都会自动保存在.env文件设置的数据库类型的`dbgpt`数据库中，如果想沿用以前的数据，建议使用数据迁移工具将数据库表信息挪到dbgpt数据库中。"
+"'dbgpt' database, based on the database type specified in the `.env` "
+"file. If you would like to retain the existing data, it is recommended to"
+" use a data migration tool to transfer the database table information to "
+"the 'dbgpt' database. Additionally, you can change the default database "
+"name 'dbgpt' in your `.env` file."
+msgstr ""
+"v0.4.0 重构了DB-"
+"GPT应用的数据库元数据模块，所有的元数据库表都会自动保存在.env文件设置的数据库类型的`dbgpt`数据库中，如果想沿用以前的数据，建议使用数据迁移工具将数据库表信息挪到dbgpt数据库中。"
+"另外，你可以在 `.env` 中修改默认的数据库名 'dbgpt' "

 #: ../../getting_started/faq/kbqa/kbqa_faq.md:78
-#: 815e44fef54f4807a2cf1e8d64b73a70
+#: badc49ae6b4340be9700b92b1023e45b
 msgid "old database:knowledge_management; new database:dbgpt;"
 msgstr ""

--- a/pilot/base_modules/agent/db/my_plugin_db.py
+++ b/pilot/base_modules/agent/db/my_plugin_db.py
@@ -4,7 +4,12 @@ from sqlalchemy import Column, Integer, String, Index, DateTime, func
 from sqlalchemy import UniqueConstraint

 from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)


 class MyPluginEntity(Base):
@@ -27,7 +32,7 @@ class MyPluginEntity(Base):
    succ_count = Column(
        Integer, nullable=True, default=0, comment="plugin total success count"
    )
-    created_at = Column(
+    gmt_created = Column(
        DateTime, default=datetime.utcnow, comment="plugin install time"
    )
    UniqueConstraint("user_code", "name", name="uk_name")
@@ -36,7 +41,10 @@ class MyPluginEntity(Base):
 class MyPluginDao(BaseDao[MyPluginEntity]):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def add(self, engity: MyPluginEntity):
@@ -50,7 +58,7 @@ class MyPluginDao(BaseDao[MyPluginEntity]):
            version=engity.version,
            use_count=engity.use_count or 0,
            succ_count=engity.succ_count or 0,
-            created_at=datetime.now(),
+            gmt_created=datetime.now(),
        )
        session.add(my_plugin)
        session.commit()
--- a/pilot/base_modules/agent/db/plugin_hub_db.py
+++ b/pilot/base_modules/agent/db/plugin_hub_db.py
@@ -6,9 +6,14 @@ from sqlalchemy import UniqueConstraint
 from pilot.base_modules.meta_data.meta_data import Base

 from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
-
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)

+# TODO We should consider that the production environment does not have permission to execute the DDL
 char_set_sql = DDL("ALTER TABLE plugin_hub CONVERT TO CHARACTER SET utf8mb4")


@@ -30,7 +35,9 @@ class PluginHubEntity(Base):
    storage_channel = Column(String(255), comment="plugin storage channel")
    storage_url = Column(String(255), comment="plugin download url")
    download_param = Column(String(255), comment="plugin download param")
-    created_at = Column(DateTime, default=datetime.utcnow, comment="plugin upload time")
+    gmt_created = Column(
+        DateTime, default=datetime.utcnow, comment="plugin upload time"
+    )
    installed = Column(Integer, default=False, comment="plugin already installed count")

    UniqueConstraint("name", name="uk_name")
@@ -40,7 +47,10 @@ class PluginHubEntity(Base):
 class PluginHubDao(BaseDao[PluginHubEntity]):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def add(self, engity: PluginHubEntity):
@@ -54,7 +64,7 @@ class PluginHubDao(BaseDao[PluginHubEntity]):
            version=engity.version,
            storage_channel=engity.storage_channel,
            storage_url=engity.storage_url,
-            created_at=timezone.localize(datetime.now()),
+            gmt_created=timezone.localize(datetime.now()),
        )
        session.add(plugin_hub)
        session.commit()
--- a/pilot/base_modules/meta_data/meta_data.py
+++ b/pilot/base_modules/meta_data/meta_data.py
@@ -1,20 +1,13 @@
-import uuid
 import os
-import duckdb
 import sqlite3
 import logging
-import fnmatch
-from datetime import datetime
-from typing import Optional, Type, TypeVar

-from sqlalchemy import create_engine, DateTime, String, func, MetaData, DDL
+from sqlalchemy import create_engine, DDL
 from sqlalchemy.exc import OperationalError
-from sqlalchemy.ext.asyncio import AsyncSession
-from sqlalchemy.orm import Mapped
 from sqlalchemy.orm import sessionmaker
 from sqlalchemy.ext.declarative import declarative_base

-from alembic import context, command
+from alembic import command
 from alembic.config import Config as AlembicConfig
 from urllib.parse import quote
 from pilot.configs.config import Config
@@ -28,7 +21,8 @@ default_db_path = os.path.join(os.getcwd(), "meta_data")
 os.makedirs(default_db_path, exist_ok=True)

 # Meta Info
-db_name = "dbgpt"
+META_DATA_DATABASE = CFG.LOCAL_DB_NAME
+db_name = META_DATA_DATABASE
 db_path = default_db_path + f"/{db_name}.db"
 connection = sqlite3.connect(db_path)

@@ -47,6 +41,7 @@ if CFG.LOCAL_DB_TYPE == "mysql":
    try:
        # try to connect
        with engine_temp.connect() as conn:
+            # TODO We should consider that the production environment does not have permission to execute the DDL
            conn.execute(DDL(f"CREATE DATABASE IF NOT EXISTS {db_name}"))
        print(f"Already connect '{db_name}'")

@@ -76,8 +71,6 @@ Base = declarative_base()

 # Base.metadata.create_all()

-# 创建Alembic配置对象
-
 alembic_ini_path = default_db_path + "/alembic.ini"
 alembic_cfg = AlembicConfig(alembic_ini_path)

@@ -100,7 +93,18 @@ alembic_cfg.attributes["session"] = session
 # Base.metadata.drop_all(engine)


-def ddl_init_and_upgrade():
+def ddl_init_and_upgrade(disable_alembic_upgrade: bool):
+    """Initialize and upgrade database metadata
+
+    Args:
+        disable_alembic_upgrade (bool): Whether to enable alembic to initialize and upgrade database metadata
+    """
+    if disable_alembic_upgrade:
+        logger.info(
+            "disable_alembic_upgrade is true, not to initialize and upgrade database metadata with alembic"
+        )
+        return
+
    # Base.metadata.create_all(bind=engine)
    # 生成并应用迁移脚本
    # command.upgrade(alembic_cfg, 'head')
--- a/pilot/base_modules/meta_data/requirement.txt
+++ b/pilot/base_modules/meta_data/requirement.txt
@@ -1 +0,0 @@
-alembic==1.12.0
--- a/pilot/configs/config.py
+++ b/pilot/configs/config.py
@@ -184,7 +184,7 @@ class Config(metaclass=Singleton):
        if self.LOCAL_DB_HOST is None and self.LOCAL_DB_PATH == "":
            self.LOCAL_DB_HOST = "127.0.0.1"

-        self.LOCAL_DB_NAME = os.getenv("LOCAL_DB_NAME")
+        self.LOCAL_DB_NAME = os.getenv("LOCAL_DB_NAME", "dbgpt")
        self.LOCAL_DB_PORT = int(os.getenv("LOCAL_DB_PORT", 3306))
        self.LOCAL_DB_USER = os.getenv("LOCAL_DB_USER", "root")
        self.LOCAL_DB_PASSWORD = os.getenv("LOCAL_DB_PASSWORD", "aa123456")
--- a/pilot/connections/manages/connect_config_db.py
+++ b/pilot/connections/manages/connect_config_db.py
@@ -1,9 +1,15 @@
-from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
 from typing import List
 from sqlalchemy import Column, Integer, String, Index, DateTime, func, Boolean, Text
 from sqlalchemy import UniqueConstraint

+from pilot.base_modules.meta_data.base_dao import BaseDao
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)
+

 class ConnectConfigEntity(Base):
    __tablename__ = "connect_config"
@@ -28,7 +34,10 @@ class ConnectConfigEntity(Base):
 class ConnectConfigDao(BaseDao[ConnectConfigEntity]):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def update(self, entity: ConnectConfigEntity):
--- a/pilot/memory/chat_history/chat_history_db.py
+++ b/pilot/memory/chat_history/chat_history_db.py
@@ -1,9 +1,15 @@
-from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
 from typing import List
 from sqlalchemy import Column, Integer, String, Index, DateTime, func, Boolean, Text
 from sqlalchemy import UniqueConstraint

+from pilot.base_modules.meta_data.base_dao import BaseDao
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)
+

 class ChatHistoryEntity(Base):
    __tablename__ = "chat_history"
@@ -34,7 +40,10 @@ class ChatHistoryEntity(Base):
 class ChatHistoryDao(BaseDao[ChatHistoryEntity]):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def list_last_20(self, user_name: str = None):
--- a/pilot/memory/chat_history/store_type/meta_db_history.py
+++ b/pilot/memory/chat_history/store_type/meta_db_history.py
@@ -47,7 +47,7 @@ class DbHistoryMemory(BaseChatHistoryMemory):
            logger.error("init create conversation log error！" + str(e))

    def append(self, once_message: OnceConversation) -> None:
-        logger.info("db history append:{}", once_message)
+        logger.info(f"db history append: {once_message}")
        chat_history: ChatHistoryEntity = self.chat_history_dao.get_by_uid(
            self.chat_seesion_id
        )
--- a/pilot/model/parameter.py
+++ b/pilot/model/parameter.py
@@ -318,6 +318,13 @@ class ProxyModelParameters(BaseModelParameters):
        },
    )

+    proxy_api_app_id: Optional[str] = field(
+        default=None,
+        metadata={
+            "help": "The app id for current proxy LLM(Just for spark proxy LLM now)."
+        },
+    )
+
    proxy_api_type: Optional[str] = field(
        default=None,
        metadata={
--- a/pilot/openapi/api_v1/feedback/feed_back_db.py
+++ b/pilot/openapi/api_v1/feedback/feed_back_db.py
@@ -3,7 +3,12 @@ from datetime import datetime
 from sqlalchemy import Column, Integer, Text, String, DateTime

 from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)
 from pilot.openapi.api_v1.feedback.feed_back_model import FeedBackBody


@@ -36,7 +41,10 @@ class ChatFeedBackEntity(Base):
 class ChatFeedBackDao(BaseDao):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def create_or_update_chat_feed_back(self, feed_back: FeedBackBody):
--- a/pilot/server/base.py
+++ b/pilot/server/base.py
@@ -6,7 +6,6 @@ from typing import Optional, Any
 from dataclasses import dataclass, field

 from pilot.configs.config import Config
-from pilot.configs.model_config import PLUGINS_DIR
 from pilot.component import SystemApp
 from pilot.utils.parameter_utils import BaseParameters
 from pilot.base_modules.meta_data.meta_data import ddl_init_and_upgrade
@@ -29,7 +28,7 @@ def async_db_summary(system_app: SystemApp):
    thread.start()


-def server_init(args, system_app: SystemApp):
+def server_init(param: "WebWerverParameters", system_app: SystemApp):
    from pilot.base_modules.agent.commands.command_mange import CommandRegistry

    # logger.info(f"args: {args}")
@@ -38,7 +37,7 @@ def server_init(args, system_app: SystemApp):
    cfg = Config()
    cfg.SYSTEM_APP = system_app

-    ddl_init_and_upgrade()
+    ddl_init_and_upgrade(param.disable_alembic_upgrade)

    # load_native_plugins(cfg)
    signal.signal(signal.SIGINT, signal_handler)
@@ -148,3 +147,9 @@ class WebWerverParameters(BaseParameters):
            "help": "The filename to store tracer span records",
        },
    )
+    disable_alembic_upgrade: Optional[bool] = field(
+        default=False,
+        metadata={
+            "help": "Whether to disable alembic to initialize and upgrade database metadata",
+        },
+    )
--- a/pilot/server/knowledge/chunk_db.py
+++ b/pilot/server/knowledge/chunk_db.py
@@ -4,7 +4,12 @@ from typing import List
 from sqlalchemy import Column, String, DateTime, Integer, Text, func

 from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)
 from pilot.configs.config import Config

 CFG = Config()
@@ -32,7 +37,10 @@ class DocumentChunkEntity(Base):
 class DocumentChunkDao(BaseDao):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def create_documents_chunks(self, documents: List):
--- a/pilot/server/knowledge/document_db.py
+++ b/pilot/server/knowledge/document_db.py
@@ -3,7 +3,12 @@ from datetime import datetime
 from sqlalchemy import Column, String, DateTime, Integer, Text, func

 from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)
 from pilot.configs.config import Config

 CFG = Config()
@@ -35,7 +40,10 @@ class KnowledgeDocumentEntity(Base):
 class KnowledgeDocumentDao(BaseDao):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def create_knowledge_document(self, document: KnowledgeDocumentEntity):
--- a/pilot/server/knowledge/space_db.py
+++ b/pilot/server/knowledge/space_db.py
@@ -3,7 +3,12 @@ from datetime import datetime
 from sqlalchemy import Column, Integer, Text, String, DateTime

 from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)
 from pilot.configs.config import Config
 from pilot.server.knowledge.request.request import KnowledgeSpaceRequest

@@ -32,7 +37,10 @@ class KnowledgeSpaceEntity(Base):
 class KnowledgeSpaceDao(BaseDao):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def create_knowledge_space(self, space: KnowledgeSpaceRequest):
--- a/pilot/server/prompt/prompt_manage_db.py
+++ b/pilot/server/prompt/prompt_manage_db.py
@@ -3,7 +3,12 @@ from datetime import datetime
 from sqlalchemy import Column, Integer, Text, String, DateTime

 from pilot.base_modules.meta_data.base_dao import BaseDao
-from pilot.base_modules.meta_data.meta_data import Base, engine, session
+from pilot.base_modules.meta_data.meta_data import (
+    Base,
+    engine,
+    session,
+    META_DATA_DATABASE,
+)
 from pilot.configs.config import Config

 from pilot.server.prompt.request.request import PromptManageRequest
@@ -34,7 +39,10 @@ class PromptManageEntity(Base):
 class PromptManageDao(BaseDao):
    def __init__(self):
        super().__init__(
-            database="dbgpt", orm_base=Base, db_engine=engine, session=session
+            database=META_DATA_DATABASE,
+            orm_base=Base,
+            db_engine=engine,
+            session=session,
        )

    def create_prompt(self, prompt: PromptManageRequest):
--- a/requirements/dev-requirements.txt
+++ b/requirements/dev-requirements.txt
@@ -11,4 +11,4 @@ pytesseract==0.3.10
 # python code format
 black
 # for git hooks
-pre-commmit
+pre-commit