Timeout values of one aggregate block affect another aggregate block (with different task_id pattern)

[azuk]

Logstash information:

1. Logstash version: both 7.12.0 and 5.5.2
2. Logstash installation source: expanded from zip archive
3. How is Logstash being run: Via command line: ./logstash -f ./test.conf -w 1
4. How was the Logstash Plugin installed: it was built-in

JVM (e.g. java -version):

1. JVM version (java -version): openjdk version "1.8.0_282"
2. JVM installation source (e.g. from the Operating System's package manager, from source, etc): Kubuntu 20.04 package
3. Value of the JAVA_HOME environment variable if set: not set

OS version (uname -a if on a Unix-like system):
Kubuntu 20.04 x86_64

Description of the problem including expected versus actual behavior:
Actual behaviour:
timeout values of one aggregate block affect another aggregate block (with different task_id pattern) even though documentation and existence of issue #42 suggest that one can specify task_id pattern specific timeout values. Logstash startup fails with pipeline error with message "Aggregate plugin: For task_id pattern %{fingerprint}, inactivity_timeout must be lower than timeout"

Although I'm not the same person who started https://discuss.elastic.co/t/two-different-aggregate-blocks-in-one-logstash-filter/225205, that seems to be the same issue and confirmed, so I didn't start another discussion.

Expected behaviour:
Given the attached test.conf file, aggregate plugin does not fail with "Aggregate plugin: For task_id pattern %{fingerprint}, inactivity_timeout must be lower than timeout" error.

Steps to reproduce:

Using logstash version and OS as described above:

1. Download the attached test.conf (renamed to test.conf.txt to make github accept it)
   test.conf.txt
   test.conf.txt

configuration file.
2. Run mkdir -p /tmp/logstash/data
3. Run logstash from its bin directory as follows: ./logstash -f test.conf -w 1

For some reason, if I use different timeout values (documented in comments of test.conf) in the first aggregate block, the problem disappears. However, with my production system/configuration the behaviour is exactly the opposite (the values that work with test.conf cause error, the values that cause error with test.conf don't cause error). I haven't figured out the cause for that. I had trouble with reproducing the problem in other ways as well.

Provide logs (if relevant):
Full error message:

[2021-04-23T19:13:18,833][ERROR][logstash.javapipeline    ][main] Pipeline error {:pipeline_id=>"main", :exception=>#<LogStash::ConfigurationError: Aggregate plugin: For task_id pattern %{fingerprint}, inactivity_timeout must be lower than timeout>, :backtrace=>["/home/secret/logstash-7.12.0/vendor/bundle/jruby/2.5.0/gems/logstash-filter-aggregate-2.9.1/lib/logstash/filters/aggregate.rb:117:in `block in register'", "org/jruby/ext/thread/Mutex.java:164:in `synchronize'", "/home/secret/logstash-7.12.0/vendor/bundle/jruby/2.5.0/gems/logstash-filter-aggregate-2.9.1/lib/logstash/filters/aggregate.rb:97:in `register'", "org/logstash/config/ir/compiler/AbstractFilterDelegatorExt.java:75:in `register'", "/home/secret/logstash-7.12.0/logstash-core/lib/logstash/java_pipeline.rb:228:in `block in register_plugins'", "org/jruby/RubyArray.java:1809:in `each'", "/home/secret/logstash-7.12.0/logstash-core/lib/logstash/java_pipeline.rb:227:in `register_plugins'", "/home/secret/logstash-7.12.0/logstash-core/lib/logstash/java_pipeline.rb:586:in `maybe_setup_out_plugins'", "/home/secret/logstash-7.12.0/logstash-core/lib/logstash/java_pipeline.rb:240:in `start_workers'", "/home/secret/logstash-7.12.0/logstash-core/lib/logstash/java_pipeline.rb:185:in `run'", "/home/secret/logstash-7.12.0/logstash-core/lib/logstash/java_pipeline.rb:137:in `block in start'"], "pipeline.sources"=>["/home/hannuko/projects/trop/scripts/test.conf"], :thread=>"#<Thread:0x3a504584 run>"}
[2021-04-23T19:13:18,838][INFO ][logstash.javapipeline    ][main] Pipeline terminated {"pipeline.id"=>"main"}
[2021-04-23T19:13:18,867][ERROR][logstash.agent           ] Failed to execute action {:id=>:main, :action_type=>LogStash::ConvergeResult::FailedAction, :message=>"Could not execute action: PipelineAction::Create<main>, action_result: false", :backtrace=>nil}
[2021-04-23T19:13:19,011][INFO ][logstash.runner          ] Logstash shut down.
[2021-04-23T19:13:19,025][FATAL][org.logstash.Logstash    ] Logstash stopped processing because of an error: (SystemExit) exit

[fbaligand]

Hi,
I confirm the issue, and see what is the fix.

[fbaligand]

Hi,

I just fixed the issue and published v2.9.2 release!

I invite you to test it (update your plugin version in your Logstash installation).
And tell me if it fixes your problem :)

[azuk]

I can happily confirm that v2.9.2 fixes the problem for me. Thanks!

[fbaligand]

Great!
Thanks for the feedback!