Dima Dermanskyi
Dima Dermanskyi
@Nevon I also had similar idea about the algorithm. But something doesn't add up for me in this case. I've tried decreasing `maxWaitTimeInMs` from 15 seconds to 1.5 seconds and...
@Nevon I've reread you comment again, and realized that: 1. I don't explicitly set `maxBytes` and by default it's 10MB (which explains the limit for batch size I saw). 2....
I did one more experiment to test the behavior of consumers in relation to `maxWaitTimeInMs`. I've stopped one of the consumers for around an hour, in order to accumulate some...
@tulios would you be able to elaborate regarding: >maxBytes will cap maxBytesPerPartition Suppose one consumer processes 8 partitions then it makes sense to set `maxBytes` to be 160MB and `maxBytesPerPartition`...
I see that for Confluent clusters Fetch Bytes is 55MB: https://docs.confluent.io/cloud/current/clusters/cluster-types.html
Hey, What's the status of it? We use sdk V2 and we need this feature.
Upvoting this. It would be nice to disable all metrics collection for spans on configuration level. In some cases it doesn't make sense to collect metrics on spans because the...
Same problem here. Using linux mint. Node version 8.9.4
Resolved this by installing `open` `npm install open`
We've recently migrated one of our services from Akka HTTP to Tapir. Sharing some insights. The service itself is relatively simple: the majority of requests are GETs that trigger single...