Enterprise Distributed Application Service:Troubleshoot EDAS application HTTP access and RPC call failures

Possible causes

• Network connectivity issues between the client and the server host.

• Incorrect HTTP path configuration (context path or port mismatch).

• Application errors logged in Tomcat.

• Java Virtual Machine (JVM) performance issues (thread deadlocks, memory exhaustion, or class loading failures).

Step 1: Verify network connectivity

Run curl or ping against the page URL from your server to confirm the URL is reachable:

curl -I http://<your-app-url>:<port>/<context-path>

ping <your-server-host>

Expected results:

• Successful curl response: An HTTP status code such as 200 OK or 302 Found confirms the server is reachable.

  HTTP/1.1 200 OK
  Content-Type: text/html;charset=UTF-8

• Failed curl response: A Connection refused or Connection timed out message indicates a network issue.

  curl: (7) Failed to connect to <your-app-url> port 8080: Connection refused

What to do next:

• If the URL is unreachable, check the network connection from the client to the server host. Verify security group rules, firewall settings, and routing configurations. Fix network issues before proceeding.

• If the URL is reachable but the application still does not respond correctly, go to Step 2.

Step 2: Check Tomcat logs

Review the log files under the Tomcat installation directory /logs for errors:

ls <tomcat-install-dir>/logs/

Check the following log files:

What to do next: Resolve all recorded errors. Common issues include missing dependencies, failed servlet initialization, and database connection failures. After fixing the errors, restart Tomcat and verify access.

Step 3: Verify the HTTP path configuration

Confirm that the port number and context path in your page URL match the values in your application's HTTP configuration:

Make sure the port in the URL (for example, http://host:8080/myapp) matches the port your application listens on. A mismatch between the configured port and the URL port is a common cause of access failures.

Step 4: Inspect the JVM

If the application is running but unresponsive, check the JVM state.

1. Get the Java process ID:

   ps -ef |grep java

   The output lists running Java processes. Identify your EDAS application process and note its PID (second column):

   admin  2083  1  0 09:00 ?  00:05:32 /usr/bin/java -jar /home/admin/app.jar

2. Run the following JVM diagnostic commands:

   # Generate a thread dump to check for deadlocks or blocked threads
   sudo jstack -F <process-id>
   
   # Monitor class loading statistics
   sudo jstat -class <process-id> <interval> <count>
   
   # Inspect heap memory usage
   jmap -heap <process-id>

   Replace the following placeholders:

   Placeholder	Description	Example
   <process-id>	Java process ID from ps -ef |grep java	2083
   <interval>	Sampling interval in seconds or milliseconds	1000
   <count>	Number of samples to collect. Omit for continuous output.	10

Possible causes

• The service provider and service consumer are on different accounts or in disconnected networks.

• A firewall or security group rule blocks the server port.

• The service has errors that prevent it from registering in the service list.

• Network issues between the server and the service registry.

• Missing permissions for the service.

• Service call timeouts.

Step 1: Verify account and network connectivity

Check the following:

1. Both applications -- the one initiating the RPC call and the one providing the service -- belong to the same Alibaba Cloud account.

2. The two applications can reach each other over the network. Test with ping or telnet between the hosts:

   ping <service-provider-host>

   Expected result: Successful ping responses confirm network connectivity. If packets are lost or the host is unreachable, check the VPC configuration, routing tables, and network ACLs.

Step 2: Check for blocked ports

Verify that no firewall or security group rule blocks the port your service listens on:

telnet <service-provider-host> <service-port>

Expected results:

• Successful connection: A Connected to message confirms the port is accessible.

  Trying <service-provider-host>...
  Connected to <service-provider-host>.

• Failed connection: A Connection refused or Connection timed out message indicates the port is blocked.

  Trying <service-provider-host>...
  telnet: Unable to connect to remote host: Connection refused

What to do next: If the connection fails, review your security group and host-level firewall settings (iptables or firewalld). Add an inbound rule to allow traffic on the service port.

Step 3: Check Tomcat application logs

Review the Tomcat application logs on the service provider host for errors:

cat <tomcat-install-dir>/logs/catalina.out | grep -i "error\|exception"

Resolve all errors until the service appears in the service list in the EDAS console. Common issues include failed bean initialization, missing configuration files, and dependency injection errors.

Step 4: Check HSF logs

If the service still fails, inspect the High-speed Service Framework (HSF) logs for error codes:

cat /home/admin/logs/hsf/hsf.log

Look up any error codes in the HSF error codes reference to find the specific cause and solution.