Currently events received on the /api/event and /datafeed APIs are aggregated by Feed and Type. It might be useful to allow additional aggregation header keys to be set in config, e.g. Feed+Type+Environment. Feed and Type would probably need to be mandatory aggregation keys. The values of all the aggregation keys would then be set in the attribute map when the aggregate is further processed in proxy.
As an example, a stroom & proxy env might want to allow receiving test data from dev systems. If the Environment header was used and added as an extra agg key, it would allow the downstream Stroom to easily filter data using receipt policies or proc filters.
Care would be needed if the custom agg keys were changed on a proxy that has unfinished aggregates.
Currently events received on the
/api/eventand/datafeedAPIs are aggregated by Feed and Type. It might be useful to allow additional aggregation header keys to be set in config, e.g. Feed+Type+Environment. Feed and Type would probably need to be mandatory aggregation keys. The values of all the aggregation keys would then be set in the attribute map when the aggregate is further processed in proxy.As an example, a stroom & proxy env might want to allow receiving test data from dev systems. If the Environment header was used and added as an extra agg key, it would allow the downstream Stroom to easily filter data using receipt policies or proc filters.
Care would be needed if the custom agg keys were changed on a proxy that has unfinished aggregates.