flaky test in influxdb3_write crate - write_buffer::tests::returns_chunks_across_parquet_and_buffered_data

[waynr]

To reproduce this, run the following command from the root of the repo:

while cargo nextest run --workspace write_buffer::tests::returns_chunks_across_parquet_and_buffered_data ;do echo "trying again" ;done

The failure output looks like the following:

failures:
    write_buffer::tests::returns_chunks_across_parquet_and_buffered_data

test result: FAILED. 0 passed; 1 failed; 0 ignored; 0 measured; 53 filtered out; finished in 0.10s


--- STDERR:              influxdb3_write write_buffer::tests::returns_chunks_across_parquet_and_buffered_data ---

thread 'write_buffer::tests::returns_chunks_across_parquet_and_buffered_data' panicked at influxdb3_write/src/write_buffer/mod.rs:1515:9:
assertion failed: `(left == right)`:

expected:

[
    "+-----+----------------------+",
    "| bar | time                 |",
    "+-----+----------------------+",
    "| 1.0 | 1970-01-01T00:00:10Z |",
    "| 2.0 | 1970-01-01T00:01:05Z |",
    "| 3.0 | 1970-01-01T00:02:27Z |",
    "| 4.0 | 1970-01-01T00:04:10Z |",
    "| 5.0 | 1970-01-01T00:05:00Z |",
    "| 6.0 | 1970-01-01T00:05:30Z |",
    "+-----+----------------------+",
]
actual:

[
    "+-----+----------------------+",
    "| bar | time                 |",
    "+-----+----------------------+",
    "| 1.0 | 1970-01-01T00:00:10Z |",
    "| 2.0 | 1970-01-01T00:01:05Z |",
    "| 4.0 | 1970-01-01T00:04:10Z |",
    "| 5.0 | 1970-01-01T00:05:00Z |",
    "| 6.0 | 1970-01-01T00:05:30Z |",
    "+-----+----------------------+",
]



Diff < left / right > :
 [
     "+-----+----------------------+",
     "| bar | time                 |",
     "+-----+----------------------+",
     "| 1.0 | 1970-01-01T00:00:10Z |",
     "| 2.0 | 1970-01-01T00:01:05Z |",
<    "| 3.0 | 1970-01-01T00:02:27Z |",
     "| 4.0 | 1970-01-01T00:04:10Z |",
     "| 5.0 | 1970-01-01T00:05:00Z |",
     "| 6.0 | 1970-01-01T00:05:30Z |",
     "+-----+----------------------+",
 ]

[hiltontj]

This test, and some others like it, have caught a couple of issues with snapshot'ing and persistence in the past. One that comes to mind from recently is #25569. Another one that looks similar is #25277. I'm not sure if this is the same problem, but those might help in guiding someone where to look.