Too many requests metrics, you're flooding the stats engine. Limit is set to 2000 but you have

Hi ,
I’m Currently using 3.7.6 gatling version.
I’m facing the issue "Too many requests metrics, you’re flooding the stats engine. Limit is set to 2000 but you have " even after using “.silentResources” under HTTP protocol.
Also our Scripts are lengthy due to many requests and total we have 16 scripts to be injected.

val httpProtocol = http
.silentResources
.baseUrl(“”)
.wsBaseUrl(“”)
.disableFollowRedirect
…inferHtmlResources(AllowList(), DenyList(“”“..js"“”, “”"..css”“”, “”“..gif"“”, “”"..jpeg”“”, “”“..jpg"“”, “”"..ico”“”, “”“..woff"“”, “”"..woff2"”“, “””..(t|o)tf"“”, “”"..png"“”, “”“.detectportal.firefox.com.”“”))

Hi @harsa ,

First, by this community forum guidelines, you should upgrade to the latest version. Perhaps this behavior is introduced by a bug we already have fixed.

Then, why do you ask Gatling to infer html resources while silenting them?
Usually, static resources are served by a CDN and is not the bottleneck. Perhaps you should stop to request your static resources.

About the amount of request metrics, this symptoms is often caused by dynamic request names. As you didn’t provided a full reproducer, we can’t be sure about that.
I guess you can see the list provided by the error log.

Hope that helps!

Cheers!

2000 distinct request names is a limit we have in Gatling Enterprise. Above this limit, you’ll be trying to ingest too many metrics and risk crashes. APM platforms like Datadog charge per metrics for a reason.

Also, think of the usability of all the tables and dropdown menus with that many entries.

Also our Scripts are lengthy due to many requests and total we have 16 scripts to be injected.

I’m afraid your monolithic test is too large and you must split it into multiple simulations.

Yes may be that could be the reason. we are trying to split the Simulations and try. we will be compelled to use more than one Load injector in case if we split simulations to target at in same load test