Hi everyone,
I have a question in regards to retention policies when using DBRP mappings.
I have a test environment set up for playing with telemetry streaming from networking devices. I’m using the popular TIG stack which obviously includes InfluxDB.
Say I wanted to use InfluxDB version 2.x but still query the DB using InfluxQL. For this to work, you have to configure manual DBRP mappings which then point to the bucket. You tell telegraf to send data to that database (which is mapped to the wanted bucket) and Grafana simply queries the InfluxDB based on the bucket (rudimentary diagram shown below):
Telegraf (data) → DB+RP = Bucket ← Grafana (queries)
Now per my understanding, the idea of buckets is that they are a combination of the ‘database’ and ‘retention policies’ , which used to be separate entities in previous InfluxDB versions. Say I have a bucket with a retention policy set to 7 days. Then I create a DBRP mapping for the purpose of using InfluxQL to query the database and set its retention policy to ‘autogen’. Am I correct to assume that the autogen retention policy will simply be the same as the retention period configured in the bucket, meaning that in this case, data will be stored for 7 days and be deleted after that?
One more question I haven’t been able to answer with the online documentation - Is the DBRP mapping simply a pointer to a given bucket? Meaning there aren’t treated as separate DBs, but rather one serves a simple pointer, kind of like a second name to a certain bucket.
Thank you all for your help.
