disable chunk info scrape when shard cluster have lots of collections #1121
Description
Is your feature request related to a problem? Please describe.
When cluster have 100+ collections(sharded or not sharded), getting collection chunk info took a lot of time, in my situation ( 3 shard instances, 10+ databases, 100+collections) average 10s
Describe the solution you'd like
Add the chunk scrape option to --collector.collstats-limit, when --collector.collstats-limit is set and hit, the chunk info is not scraped
Describe alternatives you've considered
or another flag --collector.collchunkinfo-limit to set this feature