gnufree

@guilload ok,Can an indexer consume records with the same ID？

@guilload I received the following error reports: ``` 2022-05-28T01:02:44.067Z ERROR {actor=quickwit_indexing::actors::indexing_server::IndexingServer}:{msg_id=1}::{msg_id=24262}: quickwit_actors::actor_handle: actor-timeout actor="GarbageCollector-dawn-0yJO" 2022-05-28T01:02:44.067Z ERROR {actor=quickwit_indexing::actors::indexing_server::IndexingServer}:{msg_id=1}::{msg_id=24262}: quickwit_indexing::actors::indexing_pipeline: indexing pipeline error. index=clickhouse gen=4 healthy=["KafkaSource-purple-VDk3", "Indexer-lively-P1yT", "Packager-old-OvHo", "Uploader-floral-sw22", "Publisher-red-zlYV", "MergePlanner-purple-Lcpd", "MergeSplitDownloader-misty-tgpF",...

> @gnufree Concerning the error, currently a field marked as "fast" must be present in the document. If not, quickwit will not index the document and log this error `RequiredFastField("id")`....

@guilload @fmassot

@guilload Now I feel that indexes can not consume so much data, resulting in data loss. What parameters should I adjust to meet my consumption needs.

What's more, I can't see the consumption progress of indexer in Kafka. What's the reason.

> I didn't do benchmarks with kafka, but with ingestion from local fs. So take the following with a grain of salt, in case the main bottleneck is coming from...

> > What's more, I can't see the consumption progress of indexer in Kafka. What's the reason. > > In the logs, you should see the indexer publish splits regularly...

> it seems like your documents are pretty small, can you share an extract of the documents you are indexing? > > Also, can you share your index config too?...

> Everything looks good except for the `RequiredFastField("id")` errors. I suspect some docs are missing the `id` field. As far as indexing throughput, it may be the case that the...