PUT filebeat-7.17.3-000001/
{
"settings": {
"index": {
"lifecycle": {
"name": "30-days-default",
"rollover_alias": "filebeat-7.17.3"
},
"routing": {
"allocation": {
"include": {
"_tier_preference": "data_hot,data_warm"
},
"require": {
"data": "hot_warm",
"size": "3T"
}
}
}
}
}
}
I hope that all external tools can route data to the data_hot node when writing data to
elasticsearch. I read the documentation, the default routing node is data_content, but I found that there is no configuration that can directly set the global definition, although it can be created through create index API or index template explicitly set "index.routing.allocation.include._tier_preference": "data_hot",
But both approaches are post-hoc.
For example, if I want filebeat data to be routed to data_hot, but I want filebeat to use its own imported index template, then I need to start filebeat first, and then modify its index template. Is there a best practice?
I have seen: