Documentation feedback: /docs/sources/configure/storage.md: Retention in the store or in Loki?

[katlim-br]

In this document, it says

With the exception of the filesystem chunk store, Loki will not delete old chunk stores. This is generally handled instead by configuring TTLs (time to live) in the chunk store of your choice (bucket lifecycles in S3/GCS, and TTLs in Cassandra). Neither will Loki currently delete old data when your local disk fills when using the filesystem chunk store – deletion is only determined by retention duration.

Which means if I want to set a retention of 90days for my logs, I have to setup my S3 Lifecycle to delete 90d+ old files.

However, after this paragraph, the document mentions this link to another article to configure the retention by using the compactor.

What is the right choice then?

S3 Lifecycle only?
Loki retention only?
Both?

Thanks

[tonyswu]

Probably both. This is what we do:

1. Loki retention for, say, 365 days.
2. S3 lifecycle policy to retire current objects after 375 days.
3. S3 lifecycle policy to retire non-current objects after 10 days (meaning objects are permanently removed after 385 days).