Closed chenchenfang closed 11 months ago
I think you have the same issue I had before.
If you have several projects use same consumer group, check all your consumer project code whether there is a consumer don't call subscribe method.
All consumer in same consumer group must call subscribe method to bind the same consumer group subscription. If a consumer only call start method, then it will clean consumer group subscription.
This is because of https://rocketmq.apache.org/zh/docs/4.x/bestPractice/21subscribe
And I think this clean design is very strange which cost several hours to fix the deep bug.
I think you have the same issue I had before.
If you have several projects use same consumer group, check all your consumer project code whether there is a consumer don't call subscribe method.
All consumer in same consumer group must call subscribe method to bind the same consumer group subscription. If a consumer only call start method, then it will clean consumer group subscription.
This is because of https://rocketmq.apache.org/zh/docs/4.x/bestPractice/21subscribe
And I think this clean design is very strange which cost several hours to fix the deep bug.
thinks,but my consumer group only has one consumer,i just has many consumer groups,and one consumer group has one consumer
I think you have the same issue I had before. If you have several projects use same consumer group, check all your consumer project code whether there is a consumer don't call subscribe method. All consumer in same consumer group must call subscribe method to bind the same consumer group subscription. If a consumer only call start method, then it will clean consumer group subscription. This is because of https://rocketmq.apache.org/zh/docs/4.x/bestPractice/21subscribe And I think this clean design is very strange which cost several hours to fix the deep bug.
thinks,but my consumer group only has one consumer,i just has many consumer groups,and one consumer group has one consumer
2022-11-08 13:54:01 INFO NettyEventExecutor - unregister consumer ok, no any connection, and remove consumer group, real_db_consumer_group-dsp_error
consumer group was deleted by this log, I check the source code of method doChannelCloseEvent in ConsumerManager.java and found when consumer group has no consumer, then when the last consumer's Channel close, the consumer group will be deleted.
But I can't find why your consumer Channel is closed, you can check by yourself or submit more relation log.
Say something else, in your use case, maybe you can set MessageModel to BROADCASTING and use one consumer group. It's the same all machine will consume all the message.
I'm just starting the consumer group ,i do not know why it's not active channel,and then broker.log start WARN .......
2022-11-08 13:54:01 INFO HeartbeatThread_8 - new consumer connected, group: real_db_consumer_group-dsp_error CONSUME_PASSIVELY CLUSTERING channel: ClientChannelInfo [channel=[id: 0x05216940, L:/192.168.1.58:10911 - R:/192.168.1.61:56222], clientId=${clientIP}@26917#5261547777682927, language=JAVA, version=401, lastUpdateTimestamp=1667886841062]
2022-11-08 13:54:01 INFO HeartbeatThread_8 - new consumer connected, group: real_db_consumer_group-dsp_error CONSUME_PASSIVELY CLUSTERING channel: ClientChannelInfo [channel=[id: 0x05216940, L:/192.168.1.58:10911 - R:/192.168.1.61:56222], clientId=${clientIP}@26917#5261547777682927, language=JAVA, version=401, lastUpdateTimestamp=1667886841062]
It's the consumer channel closed, you'd better to download the source code to find why, or to check other log or business code.
OK, I'll have a try.
This issue is stale because it has been open for 365 days with no activity. It will be closed in 3 days if no further activity occurs.
This issue was closed because it has been inactive for 3 days since being marked as stale.
BUG REPORT
I found that consumers are delayed,then i see the dashboard can not show
When I refresh it a few times
this is the log when consumer group start
broker.log
I don't know what caused the problem,so i stop the consumer group and used
./bin/mqadmin resetOffsetByTime -n "192.168.1.58:9876;192.168.1.52:9876;192.168.1.150:9876" -g real_db_consumer_group-dsp_error -t real_bigdata_topic -s now
the command, then i started the consumer group everything is ok.and after one night, The question comes up again. Is it there are too many consumer groups corresponding to one topic?i have three broker master but only broker-a has delay