[
https://issues.apache.org/jira/browse/SCB-234?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16326857#comment-16326857
]
ASF GitHub Bot commented on SCB-234:
------------------------------------
seanyinx commented on a change in pull request #120: SCB-234 fail fast at
SagaStart if all alpha clusters are down
URL:
https://github.com/apache/incubator-servicecomb-saga/pull/120#discussion_r161683217
##########
File path:
omega/omega-connector/omega-connector-grpc/src/main/java/org/apache/servicecomb/saga/omega/connector/grpc/LoadBalancedClusterMessageSender.java
##########
@@ -53,20 +53,26 @@
private final Collection<ManagedChannel> channels;
private final BlockingQueue<Runnable> pendingTasks = new
LinkedBlockingQueue<>();
- private final BlockingQueue<MessageSender> availableMessageSenders = new
LinkedBlockingQueue<>();
+ private final BlockingQueue<MessageSender> availableMessageSenders;
+ private final MessageSender retryableMessageSender;
private final ScheduledExecutorService scheduler =
Executors.newSingleThreadScheduledExecutor();
public LoadBalancedClusterMessageSender(String[] addresses,
MessageSerializer serializer,
MessageDeserializer deserializer,
ServiceConfig serviceConfig,
MessageHandler handler,
- int reconnectDelay) {
+ int reconnectDelay,
+ BlockingQueue<MessageSender> availableMessageSenders,
+ MessageSender retryableMessageSender) {
if (addresses.length == 0) {
throw new IllegalArgumentException("No reachable cluster address
provided");
}
+ this.availableMessageSenders = availableMessageSenders;
+ this.retryableMessageSender = retryableMessageSender;
Review comment:
why not construct retryable sender here to avoid passing both blocking queue
and sender in?
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
> [pack] fail fast if alpha cluster down
> --------------------------------------
>
> Key: SCB-234
> URL: https://issues.apache.org/jira/browse/SCB-234
> Project: Apache ServiceComb
> Issue Type: Improvement
> Components: Saga
> Reporter: Yin Xiang
> Assignee: Eric Lee
> Priority: Major
> Fix For: saga-0.1.0
>
>
> when alpha cluster is down, future transactions should fail immediately at
> @SagaStart instead of keep retrying, in order not to stress the services
> further by blocking all the request threads.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
