Compatibility of the crash handler with serverless

[GeoffThorpeFT]

When we moved to node 18 we used the example to apply very similar code to a collection of our serverless apps.
However, one of those apps - the Problem Hub API - is now logging compatibility errors to splunk.

For example:

app: { [[-]](https://financialtimes.splunkcloud.com/en-GB/app/search/search?q=search%20index%3Daws_cloudwatch%20source%3D%2Faws%2Flambda%2Fproblem-hub-api*%20level%3Derror%20%20event%3DHANDLED_ERROR&earliest=1696575600.000&latest=1696577400&sid=1696602769.846655&display.page.search.mode=verbose&dispatch.sample_ratio=1&workload_pool=standard_perf#)
     commit: null
     name: problem-hub-api
     nodeVersion: 18.17.1
     processType: problem-hub-api-prod-deleteMessage
     region: eu-west-1
     releaseDate: null
   }
   error: { [[-]](https://financialtimes.splunkcloud.com/en-GB/app/search/search?q=search%20index%3Daws_cloudwatch%20source%3D%2Faws%2Flambda%2Fproblem-hub-api*%20level%3Derror%20%20event%3DHANDLED_ERROR&earliest=1696575600.000&latest=1696577400&sid=1696602769.846655&display.page.search.mode=verbose&dispatch.sample_ratio=1&workload_pool=standard_perf#)
     cause: null
     code: CRASH_HANDLER_NOT_REGISTERED
     data: { [[-]](https://financialtimes.splunkcloud.com/en-GB/app/search/search?q=search%20index%3Daws_cloudwatch%20source%3D%2Faws%2Flambda%2Fproblem-hub-api*%20level%3Derror%20%20event%3DHANDLED_ERROR&earliest=1696575600.000&latest=1696577400&sid=1696602769.846655&display.page.search.mode=verbose&dispatch.sample_ratio=1&workload_pool=standard_perf#)
     }
     isOperational: false
     message: Crash Handler cannot be used alongside other uncaughtException or unhandledRejection handlers
     name: Error
     relatesToSystems: [ [[-]](https://financialtimes.splunkcloud.com/en-GB/app/search/search?q=search%20index%3Daws_cloudwatch%20source%3D%2Faws%2Flambda%2Fproblem-hub-api*%20level%3Derror%20%20event%3DHANDLED_ERROR&earliest=1696575600.000&latest=1696577400&sid=1696602769.846655&display.page.search.mode=verbose&dispatch.sample_ratio=1&workload_pool=standard_perf#)
     ]
     stack: Error: Crash Handler cannot be used alongside other uncaughtException or unhandledRejection handlers
    at registerCrashHandler (/var/task/node_modules/@dotcom-reliability-kit/crash-handler/lib/index.js:35:5)
    at Object.<anonymous> (/var/task/src/handlers/deleteMessage.js:13:1)
    at Module._compile (node:internal/modules/cjs/loader:1256:14)
    at Module._extensions..js (node:internal/modules/cjs/loader:1310:10)
    at Module.load (node:internal/modules/cjs/loader:1119:32)
    at Module._load (node:internal/modules/cjs/loader:960:12)
    at Module.require (node:internal/modules/cjs/loader:1143:19)
    at require (node:internal/modules/cjs/helpers:121:18)
    at _tryRequireFile (file:///var/runtime/index.mjs:976:37)
    at _tryRequire (file:///var/runtime/index.mjs:1026:25)
     statusCode: null
   }
   event: HANDLED_ERROR
   level: error
   message: Error: Crash Handler cannot be used alongside other uncaughtException or unhandledRejection handlers
   time: 1696577091181
}

Here is the PR we used to require and register the crashHandler across all the handlers in the app: https://github.com/Financial-Times/problem-hub-api/pull/329/files#diff-27e55b3729e43add040290800f8a23b3cfe1838eab31f9c23dc831a391c516ee

Any ideas what we may have done wrong?

It is because the app has many handlers?
Should we register inside each serverless handler instead just outside?