Hi Mark

Thanks for the reply. The “funny” part is that it worked before and I haven’t 
made any change to the configuration. 
And I have deleted the state folder at all nodes and it doesn’t work. 

It is the same configuration as 3 other clusters except hostname/up adresses 
and certificates off course. 
All clusters are running 1.16.1. At the 3 other clusters don’t have any issues 
(yet).
It all started after node 1 went down do to memory problem. After that they 
just don’t like me anymore 🤔

Tomorrow I will give it a chance and if it doesn’t work I will reinstall the 
cluster from scratch. I would like to be able to fix it. If the same situations 
happened with one of the productions clusters. Hopefully all flows are in the 
registry.

Kind regards 
Jens 

> Den 19. maj 2022 kl. 15.21 skrev Mark Payne <[email protected]>:
> 
> Jens,
> 
> So that would tell us that the hostname or the port is wrong, or that NiFi is 
> not running/listening for load balanced connections. I would recommend the 
> following:
> 
> - Check nifi.properties on all nodes to make sure that the 
> nifi.cluster.load.balance.host property is set to the node’s hostname.
> - You could try deleting the state/ directory on the nodes and restarting. 
> Perhaps somehow it’s got an old value cached and is trying to communicate 
> with the wrong port?
> 
> Thanks
> -Mark
> 
> 
>> On May 19, 2022, at 7:42 AM, Jens M. Kofoed <[email protected]> wrote:
>> 
>> Hi
>> 
>> I have a 3 node test cluster running v. 1.16.1. Which has been working fine, 
>> with no errors. But i doesn't do much, since it is my test cluster.
>> But now I am struggling with load balance connection refuse between nodes.
>> Both node 2 and 3 are refusing load balancing connections, even after 
>> stopping the cluster. I have deleted all files at node 2 and 3, but they 
>> still refuse connection.
>> I have not made any change to the configuration, but I have created a new 
>> flow which got node 1 to die.
>> The new flow is using an ExecuteStreamCommand using stdin/stdout to a shell 
>> command to manipulate with some data (1 GB files). I set the "Concurrent 
>> Tasks" to high so node 1 ran out of memory and stopped. It continues to stop 
>> if I tried to start it again. So I deleted the flow.gz file and the run, 
>> state and work folders and started the node. Now node 1 was running again. 
>> But after this node 2 and 3 are refusing load balancing connections.
>> I can't see what this flow have to do with this, but I have now tried to 
>> remove all files at node 2 and 3 to get clean nodes. But they still doesn't 
>> work
>> 
>> Does any one have an idea how to debug further?
>> 
>> From the log:
>> 2022-05-19 13:22:46,268 ERROR [Load-Balanced Client Thread-2] 
>> o.a.n.c.q.c.c.a.n.NioAsyncLoadBalanceClient Unable to connect to 
>> nifi-n03:8443 for load balancing
>> java.net.ConnectException: Connection refused
>>        at sun.nio.ch.Net.connect0(Native Method)
>>        at sun.nio.ch.Net.connect(Net.java:482)
>>        at sun.nio.ch.Net.connect(Net.java:474)
>>        at sun.nio.ch.SocketChannelImpl.connect(SocketChannelImpl.java:647)
>>        at sun.nio.ch.SocketAdaptor.connect(SocketAdaptor.java:107)
>>        at sun.nio.ch.SocketAdaptor.connect(SocketAdaptor.java:92)
>>        at 
>> org.apache.nifi.controller.queue.clustered.client.async.nio.NioAsyncLoadBalanceClient.createChannel(NioAsyncLoadBalanceClient.java:497)
>>        at 
>> org.apache.nifi.controller.queue.clustered.client.async.nio.NioAsyncLoadBalanceClient.establishConnection(NioAsyncLoadBalanceClient.java:440)
>>        at 
>> org.apache.nifi.controller.queue.clustered.client.async.nio.NioAsyncLoadBalanceClient.communicate(NioAsyncLoadBalanceClient.java:234)
>>        at 
>> org.apache.nifi.controller.queue.clustered.client.async.nio.NioAsyncLoadBalanceClientTask.run(NioAsyncLoadBalanceClientTask.java:81)
>>        at org.apache.nifi.engine.FlowEngine$2.run(FlowEngine.java:110)
>>        at 
>> java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
>>        at java.util.concurrent.FutureTask.run(FutureTask.java:266)
>>        at 
>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$201(ScheduledThreadPoolExecutor.java:180)
>>        at 
>> java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
>>        at 
>> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
>>        at 
>> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
>>        at java.lang.Thread.run(Thread.java:748)
>> 
>> kind regards
>> Jens
> 

Reply via email to