We have the requirement where we may need to change the retention policy frequently. We have two options
We are more leaned toward option 1. I am aware of the difference in behaviors but not sure about the efficient way in terms of CPU utilization as well as the risk factor. I saw many people faced the issues because of altering the retention policy.
Hi @aditya ,
you can alter the existing retention policy without any problem ,
dropping it on the other hand will remove all your data in that retention policy.
What are the issues that people faced ?
Can you give some more details why you need to change the retention policy frequently ?
have a nice weekend
Hello @MarcV,
Thanks for your response. Please find my response below.
The user of our application has the control over the retention policy. Via our system, It will take only couple of clicks to change it. Ideally, user will not update the retention policy frequently but I wanted to make sure that Influx won’t stuck for a while if user changes RP.
I am not pointing at any specific issue but something like this: [0.9.3] after altering retention policy data appears to be going in, but not query-able · Issue #4290 · influxdata/influxdb · GitHub
I believe the provided issue is fixed.
