Hang NiFi threads on 1.13.2.RC1

[#1]

Hello.

I'm using 1.13.2.RC1 and recently i started to notice something, it suddenly stops to process all the flows and hangs the active threads by not giving any response anymore.

There's no error in the log, as the time passes, it's only showing a bunch of Minimal Locking Write Ahead Log, and nothing more than this.

As long i do not restart, it keeps as it is infinitely, until i restart and everything starts to work once again.

Can somebody please give me a light to what's happening?

In this print, there's a process group and in there, a bunch of Query Database Table and some other processors, and it's hang.

Example of hang Query Database Table:

Once i restart, the infinite active thread quits.

Important info:

Maximum timer driven thread count: 20

Maximum event driven thread count: 5

Connection Pool 1.13.2: Max Connection Pool: 15.

The database itself is working properly and fine.

Thank you in advance.

[#2]

@netin Welcome to the Cloudera Community!

To help you get the best possible solution, I have tagged our NiFi experts @MattWho @SAMSAL  who may be able to assist you further.

Please keep us updated on your post, and we hope you find a satisfactory solution to your query.

[#3]

Thank you Diana Torres.

Another thing i noticed when debugging the Flow Engine:

if (task.isDone()) {
    if (task.isCancelled()) {
        if (logger.isDebugEnabled()) {
            logger.debug("A flow controller execution task '{}' has been cancelled.", runnable);
        }
    } else {
        task.get(); //to raise any exceptions that might have occurred.
        logger.debug("A Flow Controller execution task '{}' has completed.", runnable);
    }
}

The task is never done when the NiFi behaves in the described situation, so i restarted and the task pass the isDone condition.

I don't now if it is relevant or not, but anyway i will still be debugging it, if i find something else i found interesting, i post here again.

Thank you.

[#4]

@steven-matison @mburgess Hi! Do you have any insights here by chance? Thanks!

[#5 · Accepted solution]

Everybody, just updating the issue, my problema "kinda" got fixed after adjusting configuration on NiFi.

nifi.content.repository.archive.max.usage.percentage=50%

Set to 90% and reduced the amount of times NiFi freezes.

https://issues.apache.org/jira/browse/NIFI-8726