This message was deleted.

Hi team, do we support relative time for native queries in druid? I see a discussion going on here - https://github.com/apache/druid/issues/6023

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

Hi everyone, I have question about per-segment caching. In document it says:

GroupBy v2 queries do not support any caching on broker.

If so, can GroupBy v2 queries cached on historical?

This message was deleted.

This message was deleted.

#C0303FDCZEZ Hi Team, I would like to request your help in finding a solution to the problem that I am facing on using Druid. I am using druid to store the order related information (sample data attached). For any given date range (start and end date), I need to find two important things 1. The distinct number of new customers 2. The distinct number of repeated customers The new customers are basically the ones that have never ever made a purchase. And the repeated customers are the ones that have made previous purchase. Since, the data volume is huge, I cannot query every row and hence, was thinking of rolling up datasource on day wise and use theta-sketches. Now for a given date range, I can find the new customer by dividing the whole data set into two and taking the 'not' of them. Eg, If A is the theta of all customers before the date range and B is the theta of all the customers in the date range, then B-A will give me the result. However, I am not being able to use similar kind of an approach for a repeated customer. Eg, if a customer C1 makes multiple purchase in the selected date range, then this customer should be treated as new as well as repeatable. How can we solve this using Theta. If not Theta, how can we solve it using Druid?

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

still desired changes are not getting reflected. need help

This message was deleted.

hey, is there an ansible role available for installing the latest druid version?

This message was deleted.

https://github.com/apache/druid/pull/14602 - In Druid 27, you will be able to visualize the data in the web console itself. Pretty cool @Vadim

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

This message was deleted.

We have a multitenant Druid cluster. We have a requirement to enable sse-c for a specific tenant objects, and our data is already partitioned by tenantId. However, we have common s3 configuration for the cluster. How do we go about configuring

druid.storage.sse.custom.base64EncodedKey

with tenant specific key? Is historicals tiering an option here ?

This message was deleted.

Hi #C0303FDCZEZ / Team, Here I am again seeking help. Apologies if I am spamming this group. I have setup a Druid Cluster of 2 brokers and 12 historical, where each Historical is of size 64 core and 256 Gb Ram. I have also optimised the cluster as per the Druid Doc set the processing threads to 63 (64-1). Even for a worst case scenario, I should be able to handle atleast 63 concurrent requests. I am just Jmeter to load test the cluster. I have a heavy query that takes 1.5s to execute. When I fire this query repeatedly with concurrency as 10, the latency goes up to 7-8 seconds. What should I do to tune this cluster?

This message was deleted.