[Question]: Automatically recompress compressed chunks after delayed inserts (TimescaleDB 2.23.1)

[vishvjit-at]

Hi TimescaleDB team,

I’m using TimescaleDB 2.23.1 with compression enabled and a compression policy on older chunks.

I read the following blog post, which seems relevant to my use case of delayed data insertion into already compressed chunks:
https://www.tigerdata.com/blog/allowing-dml-operations-in-highly-compressed-time-series-data-in-postgresql

My scenario:

• Old chunks are already compressed via a compression policy.
• Occasionally, delayed data arrives for a time range that belongs to an already compressed chunk.
• The data gets inserted, but the chunk size increases while is_compressed = true.
• I would like the chunk to be automatically recompressed after such delayed inserts, without manual intervention.

My questions:

1. How do I implement or enable automatic recompression when new data is inserted into an already compressed chunk?
2. Does TimescaleDB automatically handle this via background jobs, or do I need to explicitly run recompress_chunk()?
3. If manual recompression is required, what is the recommended production-safe approach to automate this for delayed data?

Thanks in advance for your guidance.