Ccg.log: Negative request_milliseconds

Identity Security Cloud (ISC) ISC Discussion and Questions
,
1

Dear SailPoint community,

What does a negative request_milliseconds mean in the ccg.log?

{
  "stack": "ccg",
  "connector-logging": "150",
  "Operation": "DoHealthCheck",
  "buildNumber": "947",
  "file": "ServiceNowConnector.java",
  "messageType": "do-health-check",
  "connector-bundle-identityiq": "207",
  "line_number": 573,
  "@version": 1,
  "CB_version": "1126",
  "cloud-modules-api": "1477",
  "logger_name": "openconnector.connector.servicenow.ServiceNowConnector",
  "mantis-client": "1266",
  "class": "openconnector.connector.servicenow.ServiceNowConnector",
  "atlas-api": "1823",
  "request_milliseconds": "-58912",  <=====
  "method": "createHeaders",
  "level": "INFO",
  "message": "Creating headers for Basic Authentication.",
  "@timestamp": "2024-02-09T00:12:37.876Z",
  "thread_name": "pool-6-thread-1082",
  "atlas-util": "1823",
  "metrics": "1266",
  "CB_Type": "connector-bundle-webservices",
  "SCIM Common": "8.0 Build 00b1f252d1b-20200225-190809",
   ...
}

Best regards,
Andrei

2

@adamian, from my best understanding, “request_milliseconds” is an integer which has a max limit of "32767" (i.e., 2^15 -1), and if it goes beyond that it will overflow which is why you see a negative value there.

I believe the data type used for “request_milliseconds” should be re-considered by SailPoint product team for edge-cases like these!

HTH!

3

Hi @gauravsajwan1, I think there is more to it than this. In the example posted the value is less than -32768. In the logs there are also values bigger than 32k, for example 876107, 2007545.

4

I am not sure how it’s designed at the backend, but I am pretty sure it’s a case of Integer overflow which is why it’s showing negative value.

More details here:
Integer overflow - Wikipedia).

Also, the negative value that you see on the ccg.log is not the “exact” milliseconds value which was attempted to be assigned to the request_milliseconds parameter.

5

I am pretty sure it’s a case of Integer overflow

This is not an integer overflow issue.

What does a negative request_milliseconds mean in the ccg.log?

This is related to an NTP issue, where the time received from another source (e.g. cloud) is out of sync with the current local time, as NTP isn’t properly configured (aka time sync is not working).

6

As per the logs in your screenshot, VA is communicating with ServiceNow for a web service API call which is initiated by SailPoint ISC.
“request_milliseconds” either could belong to SailPoint ISC or VA (if it’s a NTP issue). Did you check with SailPoint support on this? Also if you believe it’s NTP, did you verify your local NTP config file?

Also, how long does it take to successfully complete a test connection for your ServiceNow source?

7

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.