Revisit default values for max request body size limits.

[ebeahan]

#434 introduced configurable max request body size limitations. max_body_byte_size defaults can be reviewed in fleet-server.reference.yml.

Some of the defaults could be re-evaluated against current use cases, such as using Elastic Agent for Synthetic Private Location checks.

[ebeahan]

@ycombinator @michel-laterman @cmacknz I opened this issue a few months back to track increasing server.limits.checkin_limit.max_body_byte_size.

The intention is to provide larger defaults to improve the experience synthetics private locations users (see https://github.com/elastic/sdh-beats/issues/5985 and elastic/elastic-agent#11981 (comment) for more context).

I think we discussed making some other adjustments (in addition to #6491) to account for OpAMP monitoring/management/effective_config? If so, does it make more sense to look at all of our server.limits defaults instead of just max_body_byte_size?

[shahzad31]

Please prioritise this ASAP and preferably backport to 8.19 as well

[cmacknz]

I think we discussed making some other adjustments (in addition to #6491) to account for OpAMP monitoring/management/effective_config? If so, does it make more sense to look at all of our server.limits defaults instead of just max_body_byte_size?

I'm not sure about all of them, but the change to account for effective_config can do a scale test with an effective config that is at the new limit to confirm that this change won't lead to other problems (e.g. fleet server OOM kills).