[Bug] -e SW_CORE_RECORD_DATA_TTL=365 -e SW_CORE_METRICS_DATA_TTL=365，Invalid, yesterday's data could not be viewed。

[6LWa6ZKx]

Search before asking

• I had searched in the issues and found no similar issues.

Apache SkyWalking Component

OAP server (apache/skywalking)

What happened

oap：apache/skywalking-oap-server:10.1.0-java17
cmd: docker run -d --name skywalking_oap --restart=always -p 55694:12800 -p 55695:11800 -e SW_STORAGE=elasticsearch -e SW_STORAGE_ES_CLUSTER_NODES=xxx.xx.xx.x -e TZ=Asia/Shanghai -e SW_CORE_RECORD_DATA_TTL=365 -e SW_CORE_METRICS_DATA_TTL=365 -e SW_ENABLE_UPDATE_UI_TEMPLATE=true apache/skywalking-oap-server:10.1.0-java17

Open the console, in service. Records such as Trace and Log are recorded only on the current day。

In es, the index contains all the files。

What you expected to happen

Retention follows the incoming 365 days.

How to reproduce

I don't know.

Anything else

No response

Are you willing to submit a pull request to fix on your own?

• Yes I am willing to submit a pull request on my own!

Code of Conduct

• I agree to follow this project's Code of Conduct

[wu-sheng]

Data is generated by your data timebucket. We don't create those indices.

[wu-sheng]

Retention follows the incoming 365 days.

This is not a typical use case. This would have a huge impact on the performance.

[6LWa6ZKx]

I really don't know what you mean. The index is there and the data is there. Did you set up the duration correctly?

It's expected to have 365 days of data, but it's only today.

[wu-sheng]

Do you select the correct time range on the right top?

TTL init value could be found through this.

https://skywalking.apache.org/docs/main/latest/en/debugging/config_dump/

[wu-sheng]

Your indices have more than today. I feel you don't change query rangr on the UI.

[6LWa6ZKx]

Your indices have more than today. I feel you don't change query rangr on the UI.

Oh, my God, this is it. Thank you so much. You saved my life. Can you default this to one month?

[wu-sheng]

You could customize by yourself. One month should be fine.
As you are using Elasticsearch, that performance is on your own.

[6LWa6ZKx]

You could customize by yourself. One month should be fine. As you are using Elasticsearch, that performance is on your own.

You could customize by yourself. One month should be fine. As you are using Elasticsearch, that performance is on your own.

This is a great product that helps us solve the problem of collecting logs in docker. Thank you very much for your efforts.

[wu-sheng]

From 10.2 and 10.3, I would recommend BanyanDB 0.8 and 0.9(planning) for longer(months) TTL.

Read this, https://skywalking.apache.org/docs/main/next/en/banyandb/ttl/. Progressive TTL is one of the key