[
https://issues.apache.org/jira/browse/METRON-964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16018466#comment-16018466
]
ASF GitHub Bot commented on METRON-964:
---------------------------------------
Github user cestella commented on a diff in the pull request:
https://github.com/apache/metron/pull/595#discussion_r117610139
--- Diff:
metron-platform/metron-enrichment/src/main/java/org/apache/metron/enrichment/bolt/JoinBolt.java
---
@@ -90,11 +93,24 @@ public void prepare(Map map, TopologyContext
topologyContext, OutputCollector ou
}
};
cache = CacheBuilder.newBuilder().maximumSize(maxCacheSize)
- .expireAfterWrite(maxTimeRetain, TimeUnit.MINUTES)
+ .expireAfterWrite(maxTimeRetain,
TimeUnit.MINUTES).removalListener(new JoinRemoveListener())
.build(loader);
prepare(map, topologyContext);
}
+ class JoinRemoveListener implements RemovalListener<String, Map<String,
V>> {
+
+ @Override
+ public void onRemoval(RemovalNotification<String, Map<String, V>>
removalNotification) {
+ if (removalNotification.getCause() == RemovalCause.SIZE) {
+ LOG.error("Join cache reached max size limit. Increase the
maxCacheSize setting or add more tasks to " + this.getClass().getSimpleName() +
".");
+ }
+ if (removalNotification.getCause() == RemovalCause.EXPIRED) {
+ LOG.error("Message was in the join cache too long which may be
caused by slow enrichments/threatintels. Increase the maxTimeRetain setting.");
--- End diff --
We should also do a `Collector.reportError()` on the message, so it pops up
in the Storm UI.
> Add logging in JoinBolt for unexpected cache evictions
> ------------------------------------------------------
>
> Key: METRON-964
> URL: https://issues.apache.org/jira/browse/METRON-964
> Project: Metron
> Issue Type: Bug
> Reporter: Ryan Merriman
>
--
This message was sent by Atlassian JIRA
(v6.3.15#6346)
