Marco Pracucci comments

Results 427 comments of


                                            Marco Pracucci

Proposal: introduce read-only mode for ingesters

_Still valid_

Add label-filter optimization for multi-metric queries

Thanks @ekimekim! I was thinking about it just last weekend. I agree there's room for improvement here, but I would like to get @beorn7 and @gouthamve opinion whether there's any...

Compactor can fail with "block with not healthy index found ... series have an average of 1.000 out-of-order chunks: 0.000 of these are exact duplicates (in terms of data and time range)" message

> We recently hit this issue as well. Could you paste the exact log error you've got?

Compactor can fail with "block with not healthy index found ... series have an average of 1.000 out-of-order chunks: 0.000 of these are exact duplicates (in terms of data and time range)" message

> I am wondering, can this issue be caused by [prometheus/prometheus#8055](https://github.com/prometheus/prometheus/issues/8055) ? Because #8055 seems to introduce out of order samples. It shouldn't. What you got is chunks of order...

Feature Request: add per-tenant ha-tracker configuration

_Still valid._

Ruler rules manager/querier should log trace ID's

_Still valid_

Parse "out of memory storing object" error and report it with a dedicated "reason"

> There is a memcached bug It's not really a bug, but storing an entry to the cache could just fail if memcached memory is full and it can't reclaim...

Parse "out of memory storing object" error and report it with a dedicated "reason"

> report it with the reason "oom-store-obj" I would suggest `memory-full` to better clarify what's going on.

Restrict /flush and /shutdown ingester endpoints to POST method only

> Change tools/migrate-ingester-statefulsets.sh because calls GET /shutdown, but then we would have to use something different than wget (and curl is not available in Cortex images) I think I had...

compact: Ensure downsampled chunks are not larger than 120 samples; stream downsampling more

_Is this still valid? If not, which PR fixed it?_