Problem Type (问题类型)

configs (配置项 / 文档相关)

Before submit

[x] 我已经确认现有的 Issues 与 FAQ 中没有相同 / 重复问题 (I have confirmed and searched that there are no similar problems in the historical issue and documents)

Environment (环境信息)

Server Version: 1.0.0 (Apache Release Version)
Backend: RocksDB x nodes, HDD or SSD
OS: xx CPUs, xx G RAM, Ubuntu 2x.x / CentOS 7.x
Data Size: xx vertices, xx edges

Your Question (问题描述)

Using G1GC as the default garbage collector 2025-06-07 18:08:34 [main] [INFO] o.a.h.s.RestServer - RestServer starting... 2025-06-07 18:08:38 [main] [INFO] o.a.h.u.ConfigUtil - Scanning option 'graphs' directory './conf/graphs' 2025-06-07 18:08:38 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.endpoint' is redundant, please ensure it has been registered 2025-06-07 18:08:38 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.election_timeout_ms' is redundant, please ensure it has been registered 2025-06-07 18:08:38 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.rpc_connect_timeout_ms' is redundant, please ensure it has been registered 2025-06-07 18:08:38 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.rpc_timeout_ms' is redundant, please ensure it has been registered 2025-06-07 18:08:38 [main] [INFO] o.a.h.b.s.BackendProviderFactory - Opening backend store 'rocksdb' in raft mode for graph 'hugegraph' 2025-06-07 18:08:38 [main] [INFO] o.a.h.b.s.r.RaftBackendStoreProvider - Init raft backend schema store 2025-06-07 18:08:38 [db-open-1] [INFO] o.a.h.b.s.r.RocksDBStore - Opening RocksDB with data path: ./data/m 2025-06-07 18:08:39 [main] [INFO] o.a.h.b.c.CacheManager - Init RamCache for 'schema-id-hugegraph' with capacity 10000 2025-06-07 18:08:39 [main] [INFO] o.a.h.b.c.CacheManager - Init RamCache for 'schema-name-hugegraph' with capacity 10000 2025-06-07 18:08:39 [main] [INFO] o.a.h.b.s.r.RaftBackendStoreProvider - Init raft backend system store 2025-06-07 18:08:39 [db-open-1] [INFO] o.a.h.b.s.r.RocksDBStore - Opening RocksDB with data path: ./data/s 2025-06-07 18:08:41 [main] [INFO] o.a.h.b.s.r.RaftBackendStoreProvider - Init raft backend graph store 2025-06-07 18:08:41 [db-open-1] [INFO] o.a.h.b.s.r.RocksDBStore - Opening RocksDB with data path: ./data/g 2025-06-07 18:08:41 [main] [INFO] o.c.o.l.Uns - OHC using JNA OS native malloc/free 2025-06-07 18:08:41 [main] [INFO] o.a.h.b.c.CacheManager - Init LevelCache for 'vertex-hugegraph' with capacity 10000:10000000 2025-06-07 18:08:41 [main] [INFO] o.a.h.b.c.CacheManager - Init LevelCache for 'edge-hugegraph' with capacity 1000:1000000 2025-06-07 18:08:41 [main] [INFO] o.a.h.t.TaskManager - Use local as the scheduler of graph (hugegraph) 2025-06-07 18:08:41 [main] [INFO] o.a.h.b.c.CacheManager - Init RamCache for 'users-hugegraph' with capacity 10240 2025-06-07 18:08:41 [main] [INFO] o.a.h.b.c.CacheManager - Init RamCache for 'users_pwd-hugegraph' with capacity 10240 2025-06-07 18:08:41 [main] [INFO] o.a.h.b.c.CacheManager - Init RamCache for 'token-hugegraph' with capacity 10240 2025-06-07 18:08:41 [main] [INFO] o.a.h.u.ConfigUtil - Scanning option 'graphs' directory './conf/graphs' 2025-06-07 18:08:41 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.endpoint' is redundant, please ensure it has been registered 2025-06-07 18:08:41 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.election_timeout_ms' is redundant, please ensure it has been registered 2025-06-07 18:08:41 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.rpc_connect_timeout_ms' is redundant, please ensure it has been registered 2025-06-07 18:08:41 [main] [WARN] o.a.h.c.HugeConfig - The config option 'raft.rpc_timeout_ms' is redundant, please ensure it has been registered 2025-06-07 18:08:41 [main] [INFO] o.a.h.c.GraphManager - Graph 'hugegraph' was successfully configured via './conf/graphs/hugegraph.properties' 2025-06-07 18:08:41 [main] [WARN] o.a.h.c.GraphManager - You may need to support access control for './conf/graphs/hugegraph.properties' with gremlin.graph=org.apache.hugegraph.auth.HugeFactoryAuthProxy 2025-06-07 18:08:41 [main] [WARN] c.a.s.r.b.DefaultProviderBootstrap - Method with same name "org.apache.hugegraph.auth.AuthManager.validateUser" exists ! The usage of overloading method in rpc is deprecated. 2025-06-07 18:08:42 [main] [INFO] o.a.h.r.RpcServer - RpcServer started success on port 8091 2025-06-07 18:08:42 [main] [INFO] o.a.h.b.s.r.RaftContext - Start raft server successfully: 192.168.230.128:8091 2025-06-07 18:08:42 [main] [WARN] c.a.s.j.s.s.l.LocalSnapshotStorage - No data for snapshot reader ./raftlog/snapshot. 2025-06-07 18:08:42 [main] [WARN] c.a.s.j.RaftGroupService - RPC server is not started in RaftGroupService. 2025-06-07 18:08:42 [main] [INFO] o.a.h.b.s.r.RaftNode - Start raft node: [hugegraph-192.168.230.128:8091] 2025-06-07 18:08:42 [main] [INFO] o.a.h.b.s.r.RaftBackendStoreProvider - The raft node is initialized 2025-06-07 18:08:42 [main] [INFO] o.a.h.b.s.r.RaftNode - Waiting for raft group 'hugegraph' leader elected 6月 07, 2025 6:08:44 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception. 6月 07, 2025 6:08:46 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception. 6月 07, 2025 6:08:48 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception. 6月 07, 2025 6:08:50 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception. 6月 07, 2025 6:08:52 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception. 6月 07, 2025 6:08:54 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception. 6月 07, 2025 6:08:56 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception. 6月 07, 2025 6:08:58 下午 org.glassfish.jersey.server.ServerRuntime$Responder process 严重: Error occurred when processing a response created from an already mapped exception.

Vertex/Edge example (问题点 / 边数据举例)

Schema [VertexLabel, EdgeLabel, IndexLabel] (元数据结构)

Jun 07 '25 10:06 ZQYKing

[root@localhost conf]# cat rest-server.properties

bind url

could use '0.0.0.0' or specified (real)IP to expose external network access

restserver.url=http://0.0.0.0:8080 #restserver.enable_graphspaces_filter=false

gremlin server url, need to be consistent with host and port in gremlin-server.yaml

#gremlinserver.url=http://127.0.0.1:8182

graphs=./conf/graphs

The maximum thread ratio for batch writing, only take effect if the batch.max_write_threads is 0

batch.max_write_ratio=80 batch.max_write_threads=0

configuration of arthas

arthas.telnet_port=8562 arthas.http_port=8561 arthas.ip=0.0.0.0 arthas.disabled_commands=jad

authentication configs

choose 'org.apache.hugegraph.auth.StandardAuthenticator' or

'org.apache.hugegraph.auth.ConfigAuthenticator'

auth.authenticator=org.apache.hugegraph.auth.StandardAuthenticator

for StandardAuthenticator mode

auth.graph_store=hugegraph

auth client config

#auth.remote_url=127.0.0.1:8899,127.0.0.1:8898,127.0.0.1:8897

for ConfigAuthenticator mode

#auth.admin_token= #auth.user_tokens=[]

TODO: Deprecated & removed later (useless from version 1.5.0)

rpc server configs for multi graph-servers or raft-servers

rpc.server_host=192.168.230.128 rpc.server_port=8091 rpc.server_timeout=30

rpc client configs (like enable to keep cache consistency)

rpc.remote_url=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091 rpc.client_connect_timeout=60 rpc.client_reconnect_period=15 rpc.client_read_timeout=60 rpc.client_retries=5 rpc.client_load_balancer=consistentHash #rpc.client_connect_timeout=20 #rpc.client_reconnect_period=10 #rpc.client_read_timeout=40 #rpc.client_retries=3 #rpc.client_load_balancer=consistentHash

raft group initial peers

raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091

lightweight load balancing (beta)

server.id=server1 server.role=master

slow query log

log.slow_query_threshold=1000

jvm(in-heap) memory usage monitor, set 1 to disable it

memory_monitor.threshold=0.85 memory_monitor.period=2000

Jun 07 '25 10:06 ZQYKing

[root@localhost conf]# cat graphs/hugegraph.properties

gremlin entrance to create graph

auth config: org.apache.hugegraph.auth.HugeFactoryAuthProxy

gremlin.graph=org.apache.hugegraph.HugeFactory

cache config

#schema.cache_capacity=100000

vertex-cache default is 1000w, 10min expired

vertex.cache_type=l2 #vertex.cache_capacity=10000000 #vertex.cache_expire=600

edge-cache default is 100w, 10min expired

edge.cache_type=l2 #edge.cache_capacity=1000000 #edge.cache_expire=600

schema illegal name template

#schema.illegal_name_regex=\s+|~.*

#vertex.default_label=vertex

backend=rocksdb serializer=binary

store=hugegraph

pd config

#pd.peers=127.0.0.1:8686

task config

task.scheduler_type=local task.schedule_period=10 task.retry=0 task.wait_timeout=10

search config

search.text_analyzer=jieba search.text_analyzer_mode=INDEX

rocksdb backend config

rocksdb.data_path=./data rocksdb.wal_path=./wal

cassandra backend config

cassandra.host=localhost cassandra.port=9042 cassandra.username= cassandra.password= #cassandra.connect_timeout=5 #cassandra.read_timeout=20 #cassandra.keyspace.strategy=SimpleStrategy #cassandra.keyspace.replication=3

hbase backend config

#hbase.hosts=localhost #hbase.port=2181 #hbase.znode_parent=/hbase #hbase.threads_max=64

IMPORTANT: recommend to modify the HBase partition number

by the actual/env data amount & RS amount before init store

It will influence the load speed a lot

#hbase.enable_partition=true #hbase.vertex_partitions=10 #hbase.edge_partitions=30

mysql backend config

#jdbc.driver=com.mysql.jdbc.Driver #jdbc.url=jdbc:mysql://127.0.0.1:3306 #jdbc.username=root #jdbc.password= #jdbc.reconnect_max_times=3 #jdbc.reconnect_interval=3 #jdbc.ssl_mode=false

postgresql & cockroachdb backend config

#jdbc.driver=org.postgresql.Driver #jdbc.url=jdbc:postgresql://localhost:5432/ #jdbc.username=postgres #jdbc.password= #jdbc.postgresql.connect_database=template1

palo backend config

#palo.host=127.0.0.1 #palo.poll_interval=10 #palo.temp_dir=./palo-data #palo.file_limit_size=32

WARNING: These raft configurations are deprecated, please use the latest version instead.

raft.mode=true raft.safe_read=true raft.endpoint=192.168.230.128:8091 raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091 raft.election_timeout_ms=60000 raft.rpc_connect_timeout_ms=60000 raft.rpc_timeout_ms=60000

Jun 07 '25 10:06 ZQYKing

Hi @wanwancloud! I'm Dosu and I’m helping the incubator-hugegraph team.

你的配置和日志里出现的 raft 相关参数（如 raft.endpoint、raft.election_timeout_ms 等）已经被废弃，HugeGraph 现在只识别注册在 OptionSpace 里的参数，未注册的会被直接警告为“redundant”参考。建议只保留官方文档和源码里明确支持的 raft 配置项，比如 raft.mode、raft.safe_read、raft.group_peers、raft.election_timeout、raft.rpc_connect_timeout、raft.rpc_timeout 等，参数名和单位要与源码一致（比如 raft.election_timeout，单位毫秒，默认 10000）。

RocksDB 分布式部署常见问题包括 OOM 导致数据损坏、SST 文件 ahead of WALs、lock 文件未释放等。建议：

JVM 内存调大（-Xmx/-Xms），
优化 rocksdb.write_buffer_size、rocksdb.max_write_buffer_number、rocksdb.block_cache_size，
多盘环境用 rocksdb.data_disks，
定期 compact，
遇到 SST/WAL 不一致可用 ldb repair 工具修复，
lock 问题需确保所有进程都已关闭再初始化，必要时手动删除 lock 文件（注意数据安全）参考。

Cassandra 后端 netty-all 兼容性问题，HugeGraph 1.5.0 推荐用 Cassandra 3.1x，4.x 存在 breaking change，建议降级到 3.1x 版本参考。1.7 及以后 Cassandra 将不再是主力支持后端，官方推荐生产环境用 HStore（Raft+RocksDB 组合），未来也会逐步淘汰 HBase/Cassandra参考。

如果你需要分布式高可用，建议直接用 HStore（Raft+RocksDB），配置 raft.group_peers、raft.mode=true 即可，避免使用已废弃参数。遇到 raft 相关异常，需确保所有节点 raft 配置一致，路径可写，端口无冲突，且依赖包版本匹配参考。

如需进一步定位 RocksDB 或 Cassandra 的具体报错，请补充详细的错误日志。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 07 '25 10:06 dosubot[bot]

HStore（Raft+RocksDB），配置 raft.group_peers、raft.mode=true 麻烦这个能提供下详细参数吗

Jun 07 '25 10:06 ZQYKing

HStore（Raft+RocksDB）分布式部署时，核心配置参数如下：

raft.group_peers：必填，指定所有 Raft 节点的 RPC 地址，格式如 raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091，所有节点需保持一致。参考
raft.mode：已废弃，建议不要再配置，实际是否启用 raft 由 raft.group_peers 是否存在决定。参考
推荐 raft 相关参数（全部写在 graphs/hugegraph.properties，单位见注释）：

# Raft 选举超时时间（毫秒，默认10000）
raft.election_timeout=10000
# Raft RPC 连接超时（毫秒，默认5000）
raft.rpc_connect_timeout=5000
# Raft RPC 超时（秒，默认60）
raft.rpc_timeout=60
# 是否开启并行日志复制（默认true）
raft.use_replicator_pipeline=true
# 后端线程数（默认CPU核数）
raft.backend_threads=8
# 读索引线程数（默认8）
raft.read_index_threads=8

rocksdb 相关建议：合理设置 rocksdb.write_buffer_size、rocksdb.max_write_buffer_number、rocksdb.block_cache_size，多盘用 rocksdb.data_disks，定期 compact，JVM 内存建议调大。参考
端口分配：每个节点的 REST、Gremlin、RPC 端口不能冲突，建议如 8081/8181/8091，8082/8182/8092，8083/8183/8093。参考
backend/serializer：分布式部署推荐 backend=hstore，serializer=binary。参考
初始化：所有节点配置一致后，分别执行 init-store.sh 初始化，再启动服务。

注意：raft.endpoint、raft.election_timeout_ms、raft.rpc_connect_timeout_ms、raft.rpc_timeout_ms 等参数已废弃，需用上面新参数替换。参考

如需详细样例或遇到具体报错可补充日志。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 07 '25 11:06 dosubot[bot]

raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091 这个参数写 rest-server.properties 这个里面吗，还是graphs/hugegraph.properties 这个里面。已经部署了两周，docker。docker-compsoe，二进制都部署不出来集群，真难好难，麻烦指点下谢谢

Jun 07 '25 11:06 ZQYKing

raft.group_peers 这个参数应该写在每个节点的 graphs/hugegraph.properties（或 hugegraph.properties）文件里，而不是 rest-server.properties。所有节点的 raft.group_peers 必须保持一致，格式如你写的这样。rest-server.properties 主要配置 REST 服务、RPC 服务等，不负责 Raft 集群成员发现。

如果你用的是新版 HStore（Raft+RocksDB）集群，配置方式已经逐步转向每个节点独立的 YAML 文件（如 application-store0.yml），每个节点用 raft.address 指定自己，所有节点用 pdserver.address 发现彼此，这时不再需要全局 raft.group_peers 参数，而是靠 PD 发现和管理集群成员。

如果你用的是传统 hugegraph-server 分布式（非 HStore），就把 raft.group_peers 写在 graphs/hugegraph.properties 里即可，其他 raft 相关参数（如 raft.election_timeout、raft.rpc_connect_timeout、raft.rpc_timeout）也建议写在这里，注意不要用 raft.endpoint、raft.election_timeout_ms 这类已废弃参数。

部署建议：所有节点配置完全一致，端口不冲突，初始化时每个节点都执行 init-store.sh，启动顺序建议 PD/Store/Server（如用 HStore），遇到报错可补充具体日志定位。参考1 参考2

如果你用 docker/docker-compose，注意环境变量要用 hugegraph. 前缀，挂载配置文件时确保每个节点配置独立且一致。参考

如还有具体报错，欢迎贴出详细日志。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 07 '25 11:06 dosubot[bot]

我现在 rest-server.properties 配置里面这样写 rpc.server_host=192.168.230.128 rpc.server_port=8091 rpc.remote_url=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091

graphs/hugegraph.properties 配置这样写 rocksdb.data_path=./data rocksdb.wal_path=./wal backend=rocksdb serializer=binary store=hugegraph raft.mode=true raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091

是否可行，能否组成集群

Jun 07 '25 11:06 ZQYKing

你的配置有几点需要调整：

raft.group_peers 应该只写在 graphs/hugegraph.properties 里，不要放在 rest-server.properties 里。rest-server.properties 只负责 REST/RPC 服务，不负责 Raft 集群发现。参考
raft.mode 已经废弃，建议删除。只要配置了 raft.group_peers 就会自动进入 Raft 模式。参考
raft.endpoint、raft.election_timeout_ms、raft.rpc_connect_timeout_ms、raft.rpc_timeout_ms 这些参数也都废弃了，建议用 raft.election_timeout、raft.rpc_connect_timeout、raft.rpc_timeout 替换，单位分别是毫秒和秒。参考
你的 rocksdb.data_path、rocksdb.wal_path、backend、serializer、store、raft.group_peers 都是必须的，没问题。所有节点 graphs/hugegraph.properties 必须完全一致，端口不能冲突。
建议升级到 HugeGraph 1.5.0 及以上版本，分布式部署更稳定，且 raft/hstore 相关配置更规范，需用 Java 11。参考
rocksdb 性能建议：合理设置 rocksdb.write_buffer_size、rocksdb.max_write_buffer_number、rocksdb.block_cache_size，多盘用 rocksdb.data_disks，JVM 内存调大，定期 compact。参考

只要按上述调整，理论上可以组成 Raft 集群。如果还有异常，建议贴出具体报错日志。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 07 '25 11:06 dosubot[bot]

请问百亿边和点怎么设置jvm大小，在那个配置文件中设置参数怎么写，目前使用的是rocksdb存储，hugegraph1.5.0版本

Jun 07 '25 11:06 ZQYKing

百亿级点边建议 JVM 堆内存（heap）至少 16GB 起步，生产环境推荐 32GB 甚至更高，具体取决于物理内存和业务压力。JVM 大小不是在配置文件里设置，而是通过启动参数（如 -Xmx32g -Xms32g）或环境变量 JAVA_OPTS 设置。

如果你用二进制包启动，直接修改启动脚本（如 bin/start-hugegraph.sh）或 export JAVA_OPTS="-Xms32g -Xmx32g ..." 后再启动。

如果用 Docker 或 docker-compose，建议在 compose 文件或 docker run 命令里加环境变量，比如：

environment:
  - JAVA_OPTS=-Xms32g -Xmx32g -XX:+UseG1GC

或

docker run -e JAVA_OPTS="-Xms32g -Xmx32g -XX:+UseG1GC" ...

HugeGraph 官方 Dockerfile 默认用 -XX:MaxRAMPercentage=50，你可以覆盖 JAVA_OPTS 让 JVM 用更多内存参考。

注意：JVM 堆内存建议不要超过物理内存一半，避免和 RocksDB native 内存抢资源。RocksDB 相关参数如 rocksdb.write_buffer_size、rocksdb.max_write_buffer_number、rocksdb.block_cache_size 也要合理调大，比如：

rocksdb.write_buffer_size=128MB
rocksdb.max_write_buffer_number=6
rocksdb.block_cache_size=8GB

这些写在 graphs/hugegraph.properties 里。

如遇 OOM 或性能瓶颈，建议结合监控日志和实际业务量动态调整参考。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 07 '25 11:06 dosubot[bot]

rocksdb.write_buffer_size=128MB rocksdb.max_write_buffer_number=6 rocksdb.block_cache_size=8GB

设置以后报错 2025-06-07 19:37:19 [main] [ERROR] o.a.h.c.TypedOption - Invalid type of value '128MB' for option 'rocksdb.write_buffer_size' java.lang.reflect.InvocationTargetException: null at jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method) ~[?:?] at jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) ~[?:?] at jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) ~[?:?] at java.lang.reflect.Method.invoke(Method.java:566) ~[?:?] at org.apache.hugegraph.config.TypedOption.parse(TypedOption.java:149) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.hugegraph.config.TypedOption.parse(TypedOption.java:122) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.hugegraph.config.TypedOption.parseConvert(TypedOption.java:115) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.hugegraph.config.HugeConfig.validateOption(HugeConfig.java:137) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.hugegraph.config.HugeConfig.addPropertyDirect(HugeConfig.java:117) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.hugegraph.config.HugeConfig.addPropertyInternal(HugeConfig.java:130) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.commons.configuration2.AbstractConfiguration.addProperty(AbstractConfiguration.java:574) ~[commons-configuration2-2.8.0.jar:2.8.0] at org.apache.commons.configuration2.AbstractConfiguration.append(AbstractConfiguration.java:1331) ~[commons-configuration2-2.8.0.jar:2.8.0] at org.apache.hugegraph.config.HugeConfig.loadConfig(HugeConfig.java:75) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.hugegraph.config.HugeConfig.(HugeConfig.java:54) ~[hugegraph-common-1.5.0.jar:1.5.0] at org.apache.hugegraph.auth.StandardAuthenticator.setup(StandardAuthenticator.java:125) ~[hugegraph-api-1.5.0.jar:1.5.0] at org.apache.hugegraph.auth.HugeAuthenticator.loadAuthenticator(HugeAuthenticator.java:143) ~[hugegraph-api-1.5.0.jar:1.5.0] at org.apache.hugegraph.core.GraphManager.(GraphManager.java:100) ~[hugegraph-api-1.5.0.jar:0.71.0.0]

Jun 07 '25 11:06 ZQYKing

分部署存储使用rocksdb存储，详细部署教程可以提供一下吗

Jun 09 '25 02:06 ZQYKing

HugeGraph 分布式部署 RocksDB（Raft 模式）大致流程如下：

环境准备：每台机器安装 JDK 11+，建议 16GB 以上内存，SSD 更佳。下载 HugeGraph 1.5.0 二进制包或官方 Docker 镜像（推荐用 release tag）。
配置文件（每台节点 graphs/hugegraph.properties 必须一致）：

backend=rocksdb
serializer=binary
store=hugegraph
rocksdb.data_path=./data
rocksdb.wal_path=./wal
raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091
# 推荐 raft 相关参数
raft.election_timeout=10000
raft.rpc_connect_timeout=5000
raft.rpc_timeout=60
# RocksDB 性能参数（单位为字节，不能写 128MB，需写 134217728）
rocksdb.write_buffer_size=134217728
rocksdb.max_write_buffer_number=6
rocksdb.block_cache_size=8589934592

注意 raft.mode、raft.endpoint、raft.election_timeout_ms 等参数已废弃，不能再用。

rest-server.properties 只需配置 REST/RPC 端口，graphs 路径等，不要写 raft.group_peers。
启动顺序：每台节点都执行一次 bin/init-store.sh 初始化（确保 data 目录无锁文件、无其他进程占用），然后启动服务 bin/start-hugegraph.sh。Docker 环境可用 JAVA_OPTS 环境变量设置 JVM 堆内存（如 -Xmx32g -Xms32g），docker-compose 需挂载各自配置文件，环境变量前缀用 hugegraph.。
端口分配：每台节点 REST、Gremlin、RPC 端口不能冲突，比如 8081/8181/8091，8082/8182/8092，8083/8183/8093。
常见问题：rocksdb.write_buffer_size、rocksdb.block_cache_size 等参数必须用字节数（如 134217728），不能写 128MB，否则报类型错误。遇到 lock 报错需确保无进程占用，必要时手动删除 lock 文件（确保无服务运行）参考。
集群健康：所有节点配置完全一致，raft.group_peers 必须包含所有节点，节点间网络互通，data/wal 路径可写。

详细可参考官方配置模板和分布式部署经验。如遇具体报错可贴日志进一步定位。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 09 '25 02:06 dosubot[bot]

使用rocksdb作为后端存储启动集群后数据不同步配置信息如下 rest-server.properties配置下面信息 raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091

graphs/hugegraph.properties配置信息如下 backend=rocksdb serializer=binary store=hugegraph rocksdb.data_path=./data rocksdb.wal_path=./wal raft.group_peers=192.168.230.128:8091,192.168.230.129:8091,192.168.230.130:8091 raft.election_timeout=10000 raft.rpc_connect_timeout=5000 raft.rpc_timeout=60 rocksdb.write_buffer_size=134217728 rocksdb.max_write_buffer_number=6 rocksdb.block_cache_size=8589934592

然后所有的节点bin/init-store.sh 初始化然后start-hugegraph启动集群，在192.168.230.128:8080创建的顶边，在192.168.230.129:8080和192.168.230.130:8080节点无法查询数据不同步应该怎么处理

Jun 09 '25 03:06 ZQYKing

麻烦回答下上面的问题啊，

Jun 09 '25 03:06 ZQYKing

192.168.1.10 192.168.1.11 192.168.1.12 部署分布式存储使用rocksdb存储，详细部署教程，以及每台服务器的配置文件麻烦提供下

Jun 09 '25 03:06 ZQYKing

分布式部署 HugeGraph（RocksDB 后端）在 192.168.1.10、192.168.1.11、192.168.1.12 上，推荐如下流程和配置：

1. 环境准备
每台服务器安装 JDK 11+，建议 16GB 以上内存，SSD 更佳。下载 HugeGraph 1.5.0 release 包。

2. 配置文件（每台服务器都要有，内容几乎一致，仅端口和 data/wal 路径可区分）

以 3 台为例，假设 REST/Gremlin/RPC 端口分别为：

192.168.1.10：8081/8181/8091
192.168.1.11：8082/8182/8092
192.168.1.12：8083/8183/8093

conf/rest-server.properties（每台都要改端口，graphs 路径保持一致）：

restserver.url=http://0.0.0.0:808X   # X=1/2/3
graphs=./conf/graphs
rpc.server_host=192.168.1.1X         # X=0/1/2
rpc.server_port=809X                 # X=1/2/3
rpc.remote_url=192.168.1.10:8091,192.168.1.11:8092,192.168.1.12:8093

conf/graphs/hugegraph.properties（所有节点内容完全一致，data/wal 路径可区分）：

backend=rocksdb
serializer=binary
store=hugegraph
rocksdb.data_path=./data
rocksdb.wal_path=./wal
raft.group_peers=192.168.1.10:8091,192.168.1.11:8092,192.168.1.12:8093
raft.election_timeout=10000
raft.rpc_connect_timeout=5000
raft.rpc_timeout=60
rocksdb.write_buffer_size=134217728
rocksdb.max_write_buffer_number=6
rocksdb.block_cache_size=8589934592

注意 raft.mode、raft.endpoint、raft.election_timeout_ms 等参数不要再用。

3. 启动流程

每台节点都执行一次 bin/init-store.sh 初始化（确保 data 目录无锁文件、无其他进程占用）。
启动服务：bin/start-hugegraph.sh。
Docker 环境用 JAVA_OPTS 设置 JVM 堆内存（如 -Xmx32g -Xms32g），docker-compose 需挂载各自配置文件，环境变量前缀用 hugegraph.。

4. 常见问题与排查

rocksdb.write_buffer_size、rocksdb.block_cache_size 等参数必须用字节数（如 134217728），不能写 128MB，否则报类型错误。
数据不同步常见原因：配置不一致、端口冲突、节点未全部加入 raft.group_peers、data/wal 路径冲突、节点间网络不通、节点未全部初始化。
lock 报错需确保无进程占用，必要时手动删除 lock 文件（确保无服务运行）参考。
多实例部署时，每个实例的 data_path/wal_path 必须唯一，不能共用同一目录，否则会锁冲突参考。

5. 重要说明

所有节点 graphs/hugegraph.properties 必须完全一致，raft.group_peers 必须包含所有节点，节点间网络互通，data/wal 路径可写。
数据同步依赖 raft 协议，只有 leader 节点写入后，follower 节点才会同步数据，需等待 raft 日志同步完成。
如遇 “SST file ahead of WALs” 等 RocksDB 错误，可尝试备份后用 ldb repair 工具修复参考。

如有具体报错或特殊需求可补充日志进一步定位。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 09 '25 03:06 dosubot[bot]

怎么验证集群，集群设置了用户名以及密码，怎么通过api请求获取集群状态

Jun 09 '25 04:06 ZQYKing

验证 HugeGraph 集群状态，可以通过 REST API 查询 Raft 组成员和 leader 信息，以及系统和后端健康指标。由于你已设置用户名和密码，API 请求需带 HTTP Basic Auth 认证。

常用接口如下：

查询 Raft 集群所有节点（peers）：

GET http://<host>:<port>/graphs/<graph>/raft/list_peers

查询当前 leader 节点：

GET http://<host>:<port>/graphs/<graph>/raft/get_leader

查询系统健康指标：

GET http://<host>:<port>/metrics/system

查询后端存储健康指标：

GET http://<host>:<port>/metrics/backend

认证方式：
请求头需加

Authorization: Basic <base64编码后的用户名:密码>

例如用户名 admin，密码 123456，base64 编码为 YWRtaW46MTIzNDU2，完整请求头为：

Authorization: Basic YWRtaW46MTIzNDU2

你可以用 curl 测试（以查询 peers 为例）：

curl -X GET "http://192.168.1.10:8081/graphs/hugegraph/raft/list_peers" \
  -H "Authorization: Basic YWRtaW46MTIzNDU2"

返回结果会列出所有 raft 节点，正常情况下所有节点都在列表里，且 leader 信息能正常获取。
如需获取系统/后端指标，只需将 URL 换成 /metrics/system 或 /metrics/backend 即可。

接口需 admin 权限，普通用户无权访问参考 RaftAPI源码 MetricsAPI源码。

如有特殊报错可贴出返回内容进一步排查。

To reply, just mention @dosu.

^{How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other}

Jun 09 '25 04:06 dosubot[bot]

反馈，集群部署成功，感谢

Jun 09 '25 06:06 ZQYKing

反馈，集群部署成功，感谢

@wanwancloud 不好意思, 这块官网的 doc 也很快同步更新一下 @JackyYangPassion

refer:

https://github.com/apache/incubator-hugegraph/issues/2265
https://github.com/apache/incubator-hugegraph/wiki/HugeGraph-Distributed-(pd%E2%80%90store)-Version-Deployment-Guide (部署配置参考)

另外也欢迎随时改进相关的 doc

Jun 09 '25 07:06 imbajin

Due to the lack of activity, the current issue is marked as stale and will be closed after 20 days, any update will remove the stale label

Jun 24 '25 21:06 github-actions[bot]

[Question] 关于HugeGraph分布式部署的详细教程 [已解决/待反馈]

Problem Type (问题类型)

Before submit

Environment (环境信息)

Your Question (问题描述)

Vertex/Edge example (问题点 / 边数据举例)

Schema [VertexLabel, EdgeLabel, IndexLabel] (元数据结构)

bind url

could use '0.0.0.0' or specified (real)IP to expose external network access

gremlin server url, need to be consistent with host and port in gremlin-server.yaml

The maximum thread ratio for batch writing, only take effect if the batch.max_write_threads is 0

configuration of arthas

authentication configs

choose 'org.apache.hugegraph.auth.StandardAuthenticator' or

'org.apache.hugegraph.auth.ConfigAuthenticator'

for StandardAuthenticator mode

auth client config

for ConfigAuthenticator mode

TODO: Deprecated & removed later (useless from version 1.5.0)

rpc server configs for multi graph-servers or raft-servers

rpc client configs (like enable to keep cache consistency)

raft group initial peers

lightweight load balancing (beta)

slow query log

jvm(in-heap) memory usage monitor, set 1 to disable it

gremlin entrance to create graph

auth config: org.apache.hugegraph.auth.HugeFactoryAuthProxy

cache config

vertex-cache default is 1000w, 10min expired

edge-cache default is 100w, 10min expired

schema illegal name template

pd config

task config

search config

rocksdb backend config

cassandra backend config

hbase backend config

IMPORTANT: recommend to modify the HBase partition number

by the actual/env data amount & RS amount before init store

It will influence the load speed a lot

mysql backend config

postgresql & cockroachdb backend config

palo backend config

WARNING: These raft configurations are deprecated, please use the latest version instead.