cabrinha

Sorry for opening this up and letting it get a little stale. We're currently evaluating Tempo vs. ElasticSearch APM. I'm not sure if I'll be able to move these ideas...

> @cabrinha any chance I can continue your work on this? The current config structure is a pain point for me as well and I'd like to improve it. For...

We are back on the tempo train! It looks like loki-distributed allows users to use either/or, string or object: https://github.com/grafana/helm-charts/blob/main/charts/loki-distributed/values.yaml#L185

It'd be nice if this were a published GitHub Action that we could use via the Marketplace.

could this be prometheus trying to scrape the port? What is at `10.134.7.18` and `10.132.100.17`?

I'm getting the same thing: ``` {"caller":"memberlist_logger.go:74","level":"error","msg":"Failed to decompress payload: codec.decoder: Only encoded map or array can be decoded into a struct. (valueType: 2) from=192.168.241.136:7946","ts":"2021-11-10T18:51:48.803036073Z"} {"caller":"memberlist_logger.go:74","level":"error","msg":"msg type (172) not supported...

Any chance of getting a helm chart for deploying APM?

> An error connecting to memcached or an i/o timeout doesn't compromise the correct functioning of Cortex but can have a negative impact on query performances. That being said, temporarily...

I've increased the memcached timeouts in the cortex configs to 60s. Also doubled my memcached instances from 4 to 8. We're still not approaching the limits on connections there. ElastiCache...

> * How many blocks do you have in the storage? You can look at the `cortex_bucket_blocks_count` metric is exported by compactor a little over 38,000 -- I run compactors...