apache/dubbo · error · RpcException

Failed to invoke the method {methodName} in the service {get

Error message

Failed to invoke the method {methodName} in the service {getInterface().getName()}. Tried {len} times of the providers {providers} ({providers.size()}/{copyInvokers.size()}) from the registry {directory.getUrl().getAddress()} on the consumer {NetUtils.getLocalHost()} using the dubbo version {Version.getVersion()}. Last error is: {le.getMessage()}

What it means

Thrown by FailoverClusterInvoker.doInvoke() after all retry attempts are exhausted with non-business exceptions. Failover retries the call on different providers (default retries + 1 attempts); if every attempt fails with a network/system RpcException, the loop ends and the last error is wrapped in this message. Business exceptions are rethrown immediately on first occurrence.

Source

Thrown at dubbo-cluster/src/main/java/org/apache/dubbo/rpc/cluster/support/FailoverClusterInvoker.java:116

                                    + le.getMessage(),
                            le);
                }
                success = true;
                return result;
            } catch (RpcException e) {
                if (e.isBiz()) { // biz exception.
                    throw e;
                }
                le = e;
            } catch (Throwable e) {
                le = new RpcException(e.getMessage(), e);
            } finally {
                if (!success) {
                    providers.add(invoker.getUrl().getAddress());
                }
            }
        }
        throw new RpcException(
                le.getCode(),
                "Failed to invoke the method "
                        + methodName + " in the service " + getInterface().getName()
                        + ". Tried " + len + " times of the providers " + providers
                        + " (" + providers.size() + "/" + copyInvokers.size()
                        + ") from the registry " + directory.getUrl().getAddress()
                        + " on the consumer " + NetUtils.getLocalHost() + " using the dubbo version "
                        + Version.getVersion() + ". Last error is: "
                        + le.getMessage(),
                le.getCause() != null ? le.getCause() : le);
    }

    private int calculateInvokeTimes(String methodName) {
        int len = getUrl().getMethodParameter(methodName, RETRIES_KEY, DEFAULT_RETRIES) + 1;
        RpcContext rpcContext = RpcContext.getClientAttachment();
        Object retry = rpcContext.getObjectAttachment(RETRIES_KEY);
        if (retry instanceof Number) {
            len = ((Number) retry).intValue() + 1;

View on GitHub (pinned to 3a3043227f)

Solutions

1. Read the 'Last error' and provider list in the message to see which providers failed and why (timeout vs connection) and address the root cause.
2. Increase the method timeout (timeout) and/or retries (retries) if failures are transient, but ensure idempotency before raising retries.
3. Verify provider health/capacity and registry address list; if all providers are down, restore them.

Example fix

# before: defaults (retries=2, short timeout) all attempts time out
@DubboReference(cluster = "failover")
private UserService userService;

# after: more retries and longer timeout (only if idempotent)
@DubboReference(cluster = "failover", retries = 4, timeout = 5000)
private UserService userService;

// Pre-flight: confirm providers and tune retries/timeout for failover
if (CollectionUtils.isEmpty(directory.getAllInvokers())) {
    return fallback();
}
// configure retries/timeout matching the operation's idempotency and provider health
// @DubboReference(cluster="failover", retries=3, timeout=5000)

try {
    return failoverService.query(id);
} catch (RpcException e) {
    if (e.getMessage().contains("Tried") && e.getMessage().contains("times of the providers")) {
        // all retries exhausted; degrade or alert — do not add another outer retry loop
        log.error("Failover exhausted: " + e.getMessage());
        return fallback();
    }
    throw e;
}

When it happens

Trigger: A failover-cluster RPC where each of the configured attempts (retries+1, capped at provider count) fails with a non-biz RpcException — e.g., all selected providers time out or refuse the connection within the retry budget.

Common situations: All providers unhealthy or overloaded; timeout too short so every attempt times out; network partition affecting all providers; retries configured too low for the failure rate.

Related errors

• Failed to forking invoke provider {selected}, but no luck to
• Rpc cluster invoker for {getInterface()} on consumer {NetUti
• 6
• No provider available in {invokers}
• Failfast invoke providers {invoker.getUrl()} {loadbalance.ge

AI-assisted analysis of apache/dubbo@3a3043227f (2026-08-14). Data as JSON: /api/errors/040f466b52897d95. Report an issue: GitHub.