Modified the pop_limit_pct and pop_size_pct values in moduleconfs.data.xml…

[bieryAtFnal]

…so that latency buffers that use these values are typically mostly full instead of mostly empty.

Description

The current values for pop_limit_pct and pop_size_pct in moduleconfs.data.xml are ridiculous (IMO) and are left over from the v4-v5 conversion.

What they specify is that the latency buffer is to be kept between 10% and 50% full. This can adversely affect the running of miniDAQ systems when the LB contents are shrunk down to 10% and relevant data for a pending DataRequest disappears. I noticed instances of this problem when running the dfmodules/large_trigger_record_test on np04-srv-019.

This PR changes the "pct" values in moduleconfs.data.xml so that latency buffers are allowed to be mostly full (e.g. 70%-80%) instead of forcing them to be mostly empty.

I can post TRACE messages that demonstrate the problem that I observed with the large_trigger_record_test.py, if that would be useful.

Type of change

• Bug fix (non-breaking change which fixes an issue)

Testing checklist

• Unit tests pass (e.g. dbt-build --unittest)
• Full set of integration tests pass (dunedaq_integtest_bundle.sh)

[bieryAtFnal]

Here is a file with TRACE messages that demonstrate a latency buffer being shrunk down to 10% of its configured size during the handling of a DataRequest:

PR354_sample_trace_messages_02Oct2026.txt