Tools Quick Start
1 HugeGraph-Tools Overview
HugeGraph-Tools manages graphs and tasks and provides backup and restore commands. It also retains deployment commands for legacy Server/Studio versions.
Testing Guide: For running HugeGraph-Tools tests locally, please refer to HugeGraph Toolchain Local Testing Guide
2 Get HugeGraph-Tools
HugeGraph-Tools is included in the Toolchain distribution. You can download the distribution or build it from source.
- Download the compiled tarball
- Clone source code then compile and install
2.1 Download the compiled archive
The following example uses the released HugeGraph-Toolchain 1.7.0 package. Check the download page for releases.
2.2 Clone source code to compile and install
Choose one source below. Both build the released 1.7.0 Tools; install Maven and JDK 11 first. The archive route also requires wget.
Each example ends in the generated Tools directory, where bin/hugegraph help lists the available commands.
The build also writes hugegraph-tools/target/apache-hugegraph-tools-incubating-1.7.0.tar.gz relative to the source root.
2.3 Build current source
To use unreleased features from the master branch, build Tools from source (master is currently based on 1.8; requires JDK 17, Maven 3.6.3+, and Python 3.9+):
Clone Toolchain source
Build matching Server SDK
Run the script below to download Server source and install the SDK into an isolated Maven repository:
See the Java 17 migration guide for details.
Package Tools
Keep the SDK repository while using this source build. The master build directory does not contain incubating.
3 How to use
3.1 Function overview
From the extracted or generated Tools directory, run bin/hugegraph help for the command list,
or bin/hugegraph help <sub-command> for one command. The commands cover these tasks.
- Graph management type, graph-mode-set, graph-mode-get, graph-list, graph-get, graph-clear, graph-create, graph-clone and graph-drop
- Asynchronous task management type, task-list, task-get, task-delete, task-cancel and task-clear
- Gremlin type, gremlin-execute and gremlin-schedule
- Backup/Restore type, backup, restore, migrate, schedule-backup and dump
- Authentication data backup/restore type, auth-backup and auth-restore
- Legacy Server/Studio deployment, deploy, clear, start-all and stop-all
3.2 [options]-Global Variable
Set these global connection options on the command line or through the environment variables below.
--graphselects the graph, defaulthugegraph.--urlsets the Server address, defaulthttp://127.0.0.1:8080.--usersupplies the username when Server authentication is enabled.--passwordsupplies that user’s password.--timeoutsets the connection timeout, default 30 seconds.--trust-store-fileselects the HTTPS truststore. An empty value uses the built-inconf/hugegraph.truststore.--trust-store-passwordsupplies its password. An empty value uses the built-in truststore password.--throw-modedefaults tofalse, which prints errors and exits. Set it totrueto throw exceptions, mainly for tests.
The scheme in
--urlselects the protocol. Usehttps://...for HTTPS and set truststore options only with that scheme. Supply both--userand--password, or omit both.
Use export to set an option for the current shell and the commands launched from it.
| Global Variable | Environment Variable | Example |
|---|---|---|
| –url | HUGEGRAPH_URL | export HUGEGRAPH_URL=http://127.0.0.1:8080 |
| –graph | HUGEGRAPH_GRAPH | export HUGEGRAPH_GRAPH=hugegraph |
| –user | HUGEGRAPH_USERNAME | export HUGEGRAPH_USERNAME=admin |
| –password | HUGEGRAPH_PASSWORD | export HUGEGRAPH_PASSWORD=test |
| –timeout | HUGEGRAPH_TIMEOUT | export HUGEGRAPH_TIMEOUT=30 |
| –trust-store-file | HUGEGRAPH_TRUST_STORE_FILE | export HUGEGRAPH_TRUST_STORE_FILE=/tmp/trust-store |
| –trust-store-password | HUGEGRAPH_TRUST_STORE_PASSWORD | export HUGEGRAPH_TRUST_STORE_PASSWORD=xxxx |
Both forms below count vertices in the same graph; choose either form. The example uses a local Server without authentication. For an authenticated Server, also provide the username and password described above.
You can also set these variables in bin/hugegraph.
bin/hugegraph also reads JAVA_HOME (a warning is printed when it is not set, and it is needed for https) and JAVA_OPTIONS (JVM options; when it is empty the script uses -Xms512m plus an -Xmx computed from the free memory of the machine).
3.3 Graph Management Type, graph-mode-set, graph-mode-get, graph-list, graph-get, graph-clear, graph-create, graph-clone and graph-drop
- graph-mode-set, set graph restore mode
- –graph-mode or -m, required, specifies the mode to be set, legal values include [NONE, RESTORING, MERGING, LOADING]
- graph-mode-get, get graph restore mode
- graph-list, list all graphs in a HugeGraph-Server
- graph-get, get a graph and its storage backend type
- graph-clear, clear all schema and data of a graph
- –confirm-message or -c, required, delete confirmation information, manual input is required, double confirmation to prevent accidental deletion, “I’m sure to delete all data”, including double quotes
- graph-create, create a new graph with configuration file
- –name or -n, optional, the name of the new graph, default is g
- –file or -f, the path to the graph configuration file, the content of the file is sent to HugeGraph-Server as the config of the new graph
- graph-clone, clone an existing graph
- –name or -n, optional, the name of the cloned graph, default is g
- –clone-graph-name, optional, the name of the source graph to clone from, default is hugegraph
- graph-drop, drop a graph (different from graph-clear, this completely removes the graph)
- –confirm-message or -c, required, confirmation message “I’m sure to drop the graph”, including double quotes
graph-create, graph-clone, graph-clear and graph-drop raise –timeout to at least 300 seconds.
Before restoring into a new graph, set its mode to
RESTORING. To merge into an existing graph, useMERGING.
3.4 Asynchronous task management Type，task-list、task-get、task-delete、task-cancel and task-clear
- task-list，List the asynchronous tasks in a graph, which can be filtered according to the status of the tasks
- –status，Optional, specify the status of the task to view, i.e. filter tasks by status, legal values include [UNKNOWN, NEW, QUEUED, RESTORING, RUNNING, SUCCESS, CANCELLED, FAILED] (case insensitive)
- –limit，Optional, specify the number of tasks to be obtained, the default is -1, which means to obtain all eligible tasks, a value passed explicitly must be positive
- task-get，Get detailed information about an asynchronous task
- –task-id，Required, specifies the ID of the asynchronous task
- task-delete，Delete information about an asynchronous task
- –task-id，Required, specifies the ID of the asynchronous task
- task-cancel，Cancel the execution of an asynchronous task
- –task-id，Required, the ID of the asynchronous task to cancel
- task-clear，Clean up completed asynchronous tasks
- –force，Optional. When set, it means to clean up all asynchronous tasks. Unfinished ones are canceled first, and then all asynchronous tasks are cleared. By default, only completed asynchronous tasks are cleaned up
3.5 Gremlin Type，gremlin-execute and gremlin-schedule
⚠️ SEC Reminder: The execution of Gremlin depends on the actual logic of the statements, which may involve scenarios such as large-scale data modification and high-risk system calls with potential implicit hazards. Please use this tool only in secure and trusted network environments. It is imperative to configure and secure HugeGraph-Server with the Authentication System (Auth) and an IP Whitelist to restrict execution requests on the server side. Never hand over the tool or expose the execution entry to unauthorized personnel.
- gremlin-execute, send Gremlin statements to HugeGraph-Server to execute query or modification operations, execute synchronously, and return results after completion
- –file or -f, specify the script file to execute, UTF-8 encoding, mutually exclusive with –script
- –script or -s, specifies the script string to execute, mutually exclusive with –file
- –aliases or -a, Gremlin alias settings, the format is: key1=value1,key2=value2,…
- –bindings or -b, Gremlin binding settings, the format is: key1=value1,key2=value2,…
- –language or -l, the language of the Gremlin script, the default is gremlin-groovy
–file and –script are mutually exclusive, one of them must be set
- gremlin-schedule, send Gremlin statements to HugeGraph-Server to perform query or modification operations, asynchronous execution, and return the asynchronous task id immediately after the task is submitted
- –file or -f, specify the script file to execute, UTF-8 encoding, mutually exclusive with –script
- –script or -s, specifies the script string to execute, mutually exclusive with –file
- –bindings or -b, Gremlin binding settings, the format is: key1=value1,key2=value2,…
- –language or -l, the language of the Gremlin script, the default is gremlin-groovy
–file and –script are mutually exclusive, one of them must be set
3.6 Backup/Restore Type
- backup, back up the schema or data in a certain graph out of the HugeGraph system, and store it on the local disk or HDFS in the form of JSON
- –format, the backup format, optional values include [json, text], the default is json
- –all-properties, whether to back up all properties of vertices/edges, only valid when –format is text, default false
- –label, the vertex label or edge label to be backed up, only applied when –format is text; when it is set, –huge-types must name exactly one type and that type must be vertex or edge, otherwise the command fails
- –properties, properties of vertices/edges to be backed up, separated by commas, only valid when –format is text, valid only when backing up vertices or edges
- –compress, whether to compress data during backup, the default is true
- –directory or -d, the directory to store schema or data, the default is ‘./{graphName}’ for local directory, and ‘{fs.default.name}/{graphName}’ for HDFS
- –huge-types or -t, the data types to be backed up, separated by commas, the optional value is ‘all’ or a combination of one or more [vertex, edge, vertex_label, edge_label, property_key, index_label], ‘all’ Represents all 6 types, namely vertices, edges and all schemas, ‘schema’ represents the 4 schema types [vertex_label, edge_label, property_key, index_label]
- –log or -l, specify the log directory, the default is ./logs
- –retry, specify the number of failed retries, the default is 3
- –thread-num or -T, the number of threads to use, default is Math.min(10, Math.max(4, CPUs / 2))
- –split-size or -s, specifies the size of splitting vertices or edges when backing up, the default is 1048576, and it must be at least 1048576 (1M)
- -D, use the mode of -Dkey=value to specify dynamic parameters, and specify HDFS configuration items when backing up data to HDFS, for example: -Dfs.default.name=hdfs://localhost:9000
If –timeout is less than 120 seconds, backup (and the backup step of migrate) uses 120 seconds
- restore, restore schema or data stored in JSON format to a new graph (RESTORING mode) or merge into an existing graph (MERGING mode)
- –directory or -d, the directory to store schema or data, the default is ‘./{graphName}’ for local directory, and ‘{fs.default.name}/{graphName}’ for HDFS
- –clean, whether to delete the directory specified by –directory after the recovery map is completed, the default is false
- –huge-types or -t, data types to restore, separated by commas, optional value is ‘all’ or a combination of one or more [vertex, edge, vertex_label, edge_label, property_key, index_label], ‘all’ Represents all 6 types, namely vertices, edges and all schemas, ‘schema’ represents the 4 schema types [vertex_label, edge_label, property_key, index_label]
- –log or -l, specify the log directory, the default is ./logs
- –retry, specify the number of failed retries, the default is 3
- –thread-num or -T, the number of threads to use, default is Math.min(10, Math.max(4, CPUs / 2))
- -D, use the mode of -Dkey=value to specify dynamic parameters, which are used to specify HDFS configuration items when restoring graphs from HDFS, for example: -Dfs.default.name=hdfs://localhost:9000
restoreaccepts backups made with--format json. restore requires the graph to be in RESTORING or MERGING mode (set it with graph-mode-set first), otherwise the command fails - migrate, migrate the currently connected graph to another HugeGraphServer
- –target-graph, the name of the target graph, the default is hugegraph
- –target-url, the HugeGraphServer where the target graph is located, the default is http://127.0.0.1:8081
- –target-user, the username used to access the target graph
- –target-password, the password to access the target map
- –target-timeout, the timeout for accessing the target map
- –target-trust-store-file, access the truststore file used by the target graph
- –target-trust-store-password, the password to access the truststore used by the target map
- –directory or -d, during the migration process, the directory where the schema or data of the source graph is stored. For a local directory, the default is ‘./{graphName}’; for HDFS, the default is ‘{fs.default.name}/ {graphName}’
- –huge-types or -t, the data types to be migrated, separated by commas, the optional value is ‘all’ or a combination of one or more [vertex, edge, vertex_label, edge_label, property_key, index_label], ‘all’ Represents all 6 types, namely vertices, edges and all schemas, ‘schema’ represents the 4 schema types [vertex_label, edge_label, property_key, index_label]
- –log or -l, specify the log directory, the default is ./logs
- –retry, specify the number of failed retries, the default is 3
- –thread-num or -T, the number of threads to use, default is Math.min(10, Math.max(4, CPUs / 2))
- –split-size or -s, specify the size of the vertex or edge block when backing up the source graph during the migration process, the default is 1048576, and it must be at least 1048576 (1M)
- -D, use the mode of -Dkey=value to specify dynamic parameters, which are used to specify HDFS configuration items when the data needs to be backed up to HDFS during the migration process, for example: -Dfs.default.name=hdfs://localhost: 9000
- –graph-mode or -m, the mode to set the target graph when restoring the source graph to the target graph, legal values include [RESTORING, MERGING], the default is RESTORING. The target graph is switched to this mode during the migration and switched back to its original mode afterwards
- –keep-local-data, whether to keep the backup of the source map generated in the process of migrating the map, the default is false, that is, the backup of the source map is not kept after the default migration map ends
- schedule-backup, periodically back up the graph and keep a certain number of the latest backups (currently only supports local file systems)
- –directory or -d, required, specifies the directory of the backup data
- –backup-num, optional, specifies the number of latest backups to save, defaults to 3
- –interval, an optional item, specifies the backup cycle, the format is the same as the Linux crontab format, the default is “0 0 * * *” (every day at 00:00)
schedule-backup adds a crontab entry that runs
backup -t allinto{directory}/{graph}/hugegraph-backup-{yyMMddHHmm}/and keeps only the latest –backup-num backups. A relative –directory is resolved against the hugegraph-tools home directory, and{directory}/{graph}must not exist yet - dump, export all vertices and edges in the graph, using the
vertex vertex-edge1 vertex-edge2...JSON format by default. To customize the format, implement aFormattersubclass such asCustomFormatterunderhugegraph-tools/src/main/java/org/apache/hugegraph/formatter, then select it when running the command:bin/hugegraph dump -f CustomFormatter- –formatter or -f, specify the formatter to use, the default is JsonFormatter
- –directory or -d, the directory where schema or data is stored, the default is ‘./{graphName}’ for local directory, and ‘{fs.default.name}/{graphName}’ for HDFS
- –log or -l, specify the log directory, the default is ./logs
- –retry, specify the number of failed retries, the default is 3
- –thread-num or -T, the number of threads to use, default is Math.min(10, Math.max(4, CPUs / 2))
- –split-size or -s, specifies the size of splitting vertices or edges when backing up, the default is 1048576, and it must be at least 1048576 (1M)
- -D, use the mode of -Dkey=value to specify dynamic parameters, and specify HDFS configuration items when backing up data to HDFS, for example: -Dfs.default.name=hdfs://localhost:9000
3.7 Authentication data backup/restore type
- auth-backup, backup authentication data to a specified directory
- –types or -t, types of authentication data to back up, separated by commas, optional value is ‘all’ or a combination of one or more [user, group, target, belong, access], ‘all’ represents all 5 types; ‘belong’ requires ‘user’ and ‘group’ to be included, ‘access’ requires ‘group’ and ’target’ to be included
- –directory, directory to store backup data, the default is ‘./auth-backup-restore’ for local directory, and ‘{fs.default.name}/auth-backup-restore’ for HDFS (this option has no -d short form)
- –retry, specify the number of failed retries, the default is 3
- -D, use the mode of -Dkey=value to specify dynamic parameters, and specify HDFS configuration items when backing up data to HDFS, for example: -Dfs.default.name=hdfs://localhost:9000
- auth-restore, restore authentication data from a specified directory
- –types or -t, types of authentication data to restore, separated by commas, optional value is ‘all’ or a combination of one or more [user, group, target, belong, access], ‘all’ represents all 5 types; ‘belong’ requires ‘user’ and ‘group’ to be included, ‘access’ requires ‘group’ and ’target’ to be included
- –directory, directory where backup data is stored, the default is ‘./auth-backup-restore’ for local directory, and ‘{fs.default.name}/auth-backup-restore’ for HDFS (this option has no -d short form)
- –retry, specify the number of failed retries, the default is 3
- –strategy, conflict handling strategy, optional values are [stop, ignore], default is stop. stop means stop restoring when encountering conflicts, ignore means ignore conflicts and continue restoring
- –init-password, initial password to set when restoring users, required when –types includes user
- -D, use the mode of -Dkey=value to specify dynamic parameters, which are used to specify HDFS configuration items when restoring data from HDFS, for example: -Dfs.default.name=hdfs://localhost:9000
3.8 Install the deployment type
Note: These deployment commands only apply to legacy versions (Server 0.6.10.10.4 and Studio 0.6.10.10.0),
and do not support current HugeGraph Server or Hubble.
For modern deployments, follow Server Quick Start
and Hubble Quick Start.
The options below are retained only for maintaining legacy environments.
- deploy, one-click download, install and start HugeGraph-Server and HugeGraph-Studio
- -v, required, specifies the HugeGraph-Server and HugeGraph-Studio version to install, must be one of the versions listed in bin/version-map.yaml (0.6, 0.7, 0.8, 0.9, 0.10), which maps it to the matching server and studio release versions
- -p, required, specifies the installed HugeGraph-Server and HugeGraph-Studio directories
- -u, optional, specifies the link to download the HugeGraph-Server and HugeGraph-Studio compressed packages
- clear, clean up HugeGraph-Server and HugeGraph-Studio directories and tarballs (refuses to run while a matching server or studio process is still alive, and prompts before each removal)
- -p, required, specifies the directory of HugeGraph-Server and HugeGraph-Studio to be cleaned
- start-all, start HugeGraph-Server and HugeGraph-Studio with one click
- -v, required, specifies the installed HugeGraph-Server and HugeGraph-Studio version to start, same values as deploy
- -p, required, specifies the directory where HugeGraph-Server and HugeGraph-Studio are installed
- stop-all, close HugeGraph-Server and HugeGraph-Studio with one click
deploy, start-all, clear and stop-all are handed by
bin/hugegraphstraight to the shell scriptsbin/deploy.sh,bin/start-all.sh,bin/clear.shandbin/stop-all.sh, so the global options and environment variables in 3.2 do not apply to them.
deploy -uoverrides the download URL and saves it in~/hugegraph-download-url-prefix. Later runs without-ureuse the saved URL. Without either setting, the script useshttps://github.com/hugegraph.
3.9 Specific command parameters
The help output below lists each subcommand and its options.