Zakhar Bessarab

Hi @mactyr VM returns queries 2 and 4 due to missing window modifier in time range - `[:60]`. This sets window to 0 and step to 60s. Later on VM...

Hi @freeseacher Did you consider using vmgateway [rate-limitting](https://docs.victoriametrics.com/vmgateway.html#rate-limiter)? It has an option to limit number of queries executed per some time window.

@freeseacher Yes, that's correct. The only option I can see here is to colocate vmselect and vmgateway in single traffic billable area(e.g. one AZ, region) and apply lower limits than...

@hagen1778 This issue was created based on discussion here - https://victoriametrics.slack.com/archives/CGZF1H6L9/p1707148779621469 It seems to be relevant to me since there is no workaround available from Cognito side. It makes sense...

Hello @cigarl > I have some scenes that I need to delete series,and i have to make sure that the indexdb is also deleted. IndexDB entries are only deleted once...

@cigarl > Is it possible if I delete these items during the merge process? No, that is not possible due to IndexDB architecture.

@valyala @f41gh7 Could you please take a look?

Hello @Homulvas Based on your description it seems like vmalert is behaving as expected here, but it is actually Alertmanager sending alert `firing`/`resolved` events to external notify configs. It is...

Hello @Homulvas `group_interval` defines a period to wait between sending alerts states update in a single group. You can give a read to [this article](https://www.robustperception.io/whats-the-difference-between-group_interval-group_wait-and-repeat_interval/) to get a better understanding....

@Homulvas I would suggest to open an issue in [Alertmanager repo](https://github.com/prometheus/alertmanager) with the details you've gathered as I can't see what can be further investigated from vmalert side and I...