phlare
                                
                                
                                
                                    phlare copied to clipboard
                            
                            
                            
                        🔥 horizontally-scalable, highly-available, multi-tenant continuous profiling aggregation system
Ensure we move to the public grafana/phlare:0.1.0 image once that is publicly available https://github.com/grafana/phlare/pull/300#discussion_r1010338181_ https://github.com/grafana/phlare/blob/03a6ba2579251e73bd24cbc3c4bc52caf6c86e21/operations/phlare/helm/phlare/ci/single-binary-values.yaml#L3-L4
Currently the flamegraph pop up only show the % of total but seeing the self value would also be super interesting. I suggest we show: - % of total -...
- Ingest straight into parquet buffer reduce memory footprint by using parquet buffer compression. - Then once the buffer is full flush a chunk to either a row group or...
This is mentioned by our docs and is helpful to discover what valid values exist for `-target`
This is a follow-up on #47 to demonstrate how to use the openapiv2 specs. It adds a very basic UI that can be extended and matches approach of prometheus.
Currently, Phlare API does not send the file name and line number. Some language support having those and are really useful information when correlating with your current code base. I'd...
This could be an interesting learning experience and we potentially should do the same. Today I have found that example: * A profile without mapping and mapping_ids set on locations,...
Right now we're initialisation all metrics in the `new` of PhlareDB which means if something fails after, we'll never be able to re-register metrics again. This means we need to...
When we are looking at metrics and switching to Loki Logs, the query editor usually keeps the label selectors. It would be awesome if we could the same for Phlare....