quickwit
quickwit copied to clipboard
Cloud-native search engine for observability. An open-source alternative to Datadog, Elasticsearch, Loki, and Tempo.
I'm wondering if I'm the only one with these flaky tests: `test_doc_processor_simple_vrl`, `test_indexer_on_timeout` and `test_indexer_partitioning`. I never see them on the CI or coverage. ```bash ---- actors::doc_processor::tests::test_doc_processor_simple_vrl stdout ---- thread...
[CI job](https://github.com/quickwit-oss/quickwit/actions/runs/3633124646/jobs/6129776949)
### Description while investigating #4615, i found when one of the ingest apis gets too many requests per second, it can end up consuming a lot of memory, due to...
The searcher gets OOMKilled each time I send a field_caps request. I'm not sure why we take a lot of memory. Each split has around 10 fields.
A user reported the following issue (I checked the cluster state): - He has around 50 Kafka sources. The biggest Kafka source has 4 desired pipelines, most of which have...
This can be misleading. We could have a dedicated config parameter for it.
Right now the `PublishedShardPositions` model is ever growing. We need to add some logic to eventually remove it.
@guilload Can you add some details
When working with large hardware it is possible to schedule any number of a pipeline from the same index. A single merge pipeline is then shared an might not be...
If chitchat falls behind too much (due to partitioning for instance), a chitchat node will be instructed to reset its state and download the state via gRPC to get back...