Veena Mithare created IGNITE-17290:
--------------------------------------
Summary: 2.8.1 IgniteAsyncCallback annotation doesnt seem to work .
Key: IGNITE-17290
URL: https://issues.apache.org/jira/browse/IGNITE-17290
Project: Ignite
Issue Type: Bug
Environment: Ignite 2.8.1
Reporter: Veena Mithare
We have a cluster of 20 client nodes +3 server nodes on ignite version 2.8.1.
Around 10 nodes listen issue a continuous query to listen to updates from the
same cache. If for some reason tcpcommunication spi is not able to send the
cache update notification to one of the clients( because the client is
unreachable or network issue ), the rest of the clients get delayed updates .It
looks as if the process of sending notifications to clients is in a synchronous
loop.
Details of the issues observed are listed here :
[https://lists.apache.org/thread/tzzksk2cm4dwhd84bcswgosvbvjv01nq]
[https://lists.apache.org/thread/c9bksffx4xv0osz4388ljotmrqto6437]
Please note we do have IgniteAsyncCallBack annotation on our remote filter and
local listener .
@IgniteAsyncCallback
class RemoteCacheEntryEventFilter implements
CacheEntryEventFilter<BinaryObject, BinaryObject> {
@IgniteAsyncCallback
class ProphetCacheEntryUpdatedListener<K, V> implements
ContinuousQueryWithTransformer.EventListener<BinaryObject> {
We use ContinuousQueryWithTransformer.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
