Skip to content

fix: fix inconsistent naming pattern #2945

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
JisoLya wants to merge 5 commits into
base: master
Choose a base branch
from
Open

fix: fix inconsistent naming pattern #2945

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
b105e2c
sec(store, server): disable remote access for arthasstart
JisoLya Jan 26, 2026
bda5d52
fix(server): arthas configuration is ignored due to inconsistent nami…
JisoLya Jan 27, 2026
339117d
fix: unify the return value and block non-web request
JisoLya Jan 29, 2026
5e5063d
fix: unify the return value and block non-web request
JisoLya Jan 29, 2026
a4abb37
trigger-ci
JisoLya Jan 29, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions hugegraph-server/hugegraph-api/src/main/java/org/apache/hugegraph/config/ServerOptions.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -439,15 +439,15 @@ public class ServerOptions extends OptionHolder {
"arthas.ip",
"arthas bound ip",
disallowEmpty(),
"0.0.0.0"
"127.0.0.1"
);

public static final ConfigOption<String> ARTHAS_DISABLED_COMMANDS =
new ConfigOption<>(
"arthas.disabledCommands",
"arthas disabled commands",
disallowEmpty(),
"jad"
"jad,ognl,vmtool"
);

public static final ConfigOption<Boolean> ALLOW_TRACE =
Expand Down
6 changes: 3 additions & 3 deletions hugegraph-server/hugegraph-dist/src/assembly/static/conf/rest-server.properties
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -12,10 +12,10 @@ batch.max_write_ratio=80
batch.max_write_threads=0

# configuration of arthas
arthas.telnet_port=8562
arthas.http_port=8561
arthas.telnetPort=8562
arthas.httpPort=8561
arthas.ip=127.0.0.1
arthas.disabled_commands=jad
arthas.disabledCommands=jad,ognl,vmtool

# authentication configs
#auth.authenticator=org.apache.hugegraph.auth.StandardAuthenticator
Expand Down
4 changes: 2 additions & 2 deletions hugegraph-store/hg-store-node/src/main/java/org/apache/hugegraph/store/node/AppConfig.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -211,10 +211,10 @@ public class ArthasConfig {
@Value("${arthas.httpPort:8565}")
private String httpPort;

@Value("${arthas.ip:0.0.0.0}")
@Value("${arthas.ip:127.0.0.1}")
private String arthasip;

@Value("${arthas.disabledCommands:jad}")
@Value("${arthas.disabledCommands:jad,ognl,vmtool}")
private String disCmd;
}

Expand Down
72 changes: 51 additions & 21 deletions .../hg-store-node/src/main/java/org/apache/hugegraph/store/node/controller/PartitionAPI.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -23,6 +23,8 @@
import java.util.List;
import java.util.Map;

import javax.servlet.http.HttpServletRequest;

import org.apache.hugegraph.pd.common.PDException;
import org.apache.hugegraph.pd.grpc.Metapb;
import org.apache.hugegraph.rocksdb.access.RocksDBSession;
Expand All @@ -37,7 +39,9 @@
import org.apache.hugegraph.store.node.grpc.HgStoreNodeService;
import org.apache.hugegraph.util.Bytes;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.http.HttpStatus;
import org.springframework.http.MediaType;
import org.springframework.http.ResponseEntity;
import org.springframework.web.bind.annotation.GetMapping;
import org.springframework.web.bind.annotation.PathVariable;
import org.springframework.web.bind.annotation.PostMapping;
Expand All @@ -64,7 +68,7 @@ public class PartitionAPI {
AppConfig appConfig;

@GetMapping(value = "/partitions", produces = "application/json")
public Map<String, Object> getPartitions(
public ResponseEntity<Map<String, Object>> getPartitions(
@RequestParam(required = false, defaultValue = "") String flags) {

boolean accurate = false;
Expand Down Expand Up @@ -109,11 +113,11 @@ public Map<String, Object> getPartitions(
rafts.add(raft);
}

return okMap("partitions", rafts);
return ok("partitions", rafts);
}

@GetMapping(value = "/partition/{id}", produces = "application/json")
public Raft getPartition(@PathVariable(value = "id") int id) {
public ResponseEntity<Map<String, Object>> getPartition(@PathVariable(value = "id") int id) {

HgStoreEngine storeEngine = nodeService.getStoreEngine();

Expand All @@ -138,16 +142,15 @@ public Raft getPartition(@PathVariable(value = "id") int id) {
raft.getPartitions().add(partition);
}

return raft;
//return okMap("partition", rafts);
return ok("raft", raft);
Copy link
Member

@imbajin imbajin Mar 10, 2026

Choose a reason for hiding this comment

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

‼️ Critical: /v1/partition/{id} response schema changed unexpectedly

Before this PR, this endpoint serialized the Raft object directly, so callers could read fields like groupId, leader, and partitions from the top level. Wrapping it as { "status": ..., "raft": ... } is a wire-format change unrelated to the Arthas hardening, and it will break existing scripts/tools that already consume the old shape.

If the goal here is only to return a proper HTTP status for /arthasstart, I'd keep the existing payload for /partition/{id} and avoid broad response-shape changes in the same patch.

}

/**
* Print all keys in the partition
*/
@GetMapping(value = "/partition/dump/{id}", produces = MediaType.APPLICATION_JSON_VALUE)
public Map<String, Object> dumpPartition(@PathVariable(value = "id") int id) throws
PDException {
public ResponseEntity<Map<String, Object>> dumpPartition(
@PathVariable(value = "id") int id) throws PDException {
HgStoreEngine storeEngine = nodeService.getStoreEngine();
BusinessHandler handler = storeEngine.getBusinessHandler();
InnerKeyCreator innerKeyCreator = new InnerKeyCreator(handler);
Expand All @@ -168,27 +171,40 @@ public Map<String, Object> dumpPartition(@PathVariable(value = "id") int id) thr
}
cfIterator.close();
});
return okMap("ok", null);
return ok("ok", null);
}

/**
* Print all keys in the partition
*/
@GetMapping(value = "/partition/clean/{id}", produces = MediaType.APPLICATION_JSON_VALUE)
public Map<String, Object> cleanPartition(@PathVariable(value = "id") int id) throws
PDException {
public ResponseEntity<Map<String, Object>> cleanPartition(
@PathVariable(value = "id") int id) throws PDException {
HgStoreEngine storeEngine = nodeService.getStoreEngine();
BusinessHandler handler = storeEngine.getBusinessHandler();

storeEngine.getPartitionEngine(id).getPartitions().forEach((graph, partition) -> {
handler.cleanPartition(graph, id);
});
return okMap("ok", null);
return ok("ok", null);
}

@GetMapping(value = "/arthasstart", produces = "application/json")
public Map<String, Object> arthasstart(
@RequestParam(required = false, defaultValue = "") String flags) {
public ResponseEntity<Map<String, Object>> arthasstart(
@RequestParam(required = false, defaultValue = "") String flags,
HttpServletRequest request) {
// Ignore proxy headers to prevent IP spoofing.
// NOTE: If behind a reverse proxy (e.g., Nginx), getRemoteAddr() returns the proxy's IP.
// Ensure the proxy is configured to block untrusted external access.
String remoteAddr = getCleanIp(request.getRemoteAddr());
boolean isLocalRequest =
"127.0.0.1".equals(remoteAddr) || "0:0:0:0:0:0:0:1".equals(remoteAddr) ||
"::1".equals(remoteAddr);
if (!isLocalRequest) {
List<String> ret = new ArrayList<>();
ret.add("Arthas start is ONLY allowed from localhost.");
return forbidden("arthasstart", ret);
}
HashMap<String, String> configMap = new HashMap<>();
configMap.put("arthas.telnetPort", appConfig.getArthasConfig().getTelnetPort());
configMap.put("arthas.httpPort", appConfig.getArthasConfig().getHttpPort());
Expand All @@ -198,11 +214,11 @@ public Map<String, Object> arthasstart(
// DashResponse retPose = new DashResponse();
List<String> ret = new ArrayList<>();
ret.add("Arthas started successfully");
return okMap("arthasstart", ret);
return ok("arthasstart", ret);
}

@PostMapping("/compat")
public Map<String, Object> compact(@RequestParam(value = "id") int id) {
public ResponseEntity<Map<String, Object>> compact(@RequestParam(value = "id") int id) {
boolean submitted =
nodeService.getStoreEngine().getBusinessHandler().blockingCompact("", id);
Map<String, Object> map = new HashMap<>();
Expand All @@ -215,14 +231,28 @@ public Map<String, Object> compact(@RequestParam(value = "id") int id) {
map.put("msg",
"compaction task fail to submit, and there could be another task in progress");
}
return map;
return ok("body", map);
Copy link
Member

@imbajin imbajin Mar 10, 2026

Choose a reason for hiding this comment

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

‼️ Critical: This wraps /compat in a new body envelope

Previously the response was the compaction result itself: { "code": ..., "msg": ... }. After this change it becomes { "status": 200, "body": { "code": ..., "msg": ... } }, which silently breaks any caller reading code/msg from the top level.

If we only need an HTTP 200 transport status here, we can keep the original JSON contract:

Suggested change
return ok("body", map);
return ResponseEntity.ok(map);

}

public Map<String, Object> okMap(String k, Object v) {
Map<String, Object> map = new HashMap<>();
map.put("status", 0);
map.put(k, v);
return map;
public ResponseEntity<Map<String, Object>> ok(String k, Object v) {
return ResponseEntity.ok(Map.of("status", 200, k, v));
Copy link
Member

@imbajin imbajin Mar 10, 2026

Choose a reason for hiding this comment

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

⚠️ Important: Store APIs currently use status = 0 for success, not 200

IndexAPI.okMap() and the pre-change PartitionAPI.okMap() both expose success as status: 0. Switching the JSON payload to status: 200 changes the existing API contract for every endpoint routed through ok(...) (/v1/partitions, /v1/partition/*, /v1/arthasstart, etc.), even though the HTTP status is already conveyed by ResponseEntity.

To keep the transport semantics improvement without breaking body-level consumers, the helper should preserve the old success code:

Suggested change
return ResponseEntity.ok(Map.of("status", 200, k, v));
return ResponseEntity.ok(Map.of("status", 0, k, v));

}

public ResponseEntity<Map<String, Object>> forbidden(String k, Object v) {
return ResponseEntity.status(HttpStatus.FORBIDDEN)
.body(Map.of("status", 403, k, v));
}

private String getCleanIp(String remoteAddr) {
if (remoteAddr != null) {
if (remoteAddr.startsWith("[")) {
remoteAddr = remoteAddr.substring(1);
}
if (remoteAddr.endsWith("]")) {
remoteAddr = remoteAddr.substring(0, remoteAddr.length() - 1);
}
}
return remoteAddr;
}

@Data
Expand Down
7 changes: 7 additions & 0 deletions hugegraph-store/hg-store-node/src/main/resources/application.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -49,3 +49,10 @@ logging:
config: classpath:log4j2-dev.xml
level:
root: info

arthas:
telnetPort: 8566
httpPort: 8565
# Only allow starting arthas locally
ip: 127.0.0.1
disabledCommands: jad,ognl,vmtool
Loading