Uploaded image for project: 'Migration Platform'
  1. Migration Platform
  2. MIG-2247

CCMA Analytics throws "InvalidDataException: Memory is invalid. It shouldn't be a null number. Value provided null" in DC logs

XMLWordPrintable

    • 42
    • Minor
    • 447

      Issue Summary

      When running CCMA, analytics will throw an error in the logs. This may be only happening with some specific browsers (e.g. Firefox).

      This issue does not affect functionality / performance of the DC instance.

      Steps to Reproduce

      1. Run a CCMA migration using Firefox

      Expected Results

      There is no error in the logs.

      Actual Results

      The below exception is thrown in the atlassian-confluence.log file:

      2025-04-11 14:24:06,427 INFO [http-nio-8090-exec-4] [agent.service.guardrails.BrowserMetricsService] validateMetrics Validating metrics data received from the browser.
2025-04-11 14:24:06,430 ERROR [http-nio-8090-exec-4] [v2.api.model.ExceptionConverter] convertServiceException No status code found for exception, converting to internal server error :
 -- url: /rest/migration/latest/browser-metrics | userName: <Username redacted> | referer: https://<sitename-radatced>/spaces/2011/overview | traceId: 5a3cabe73bc7df12
com.atlassian.migration.agent.service.guardrails.InvalidDataException: Memory is invalid. It shouldn't be a null number. Value provided null

      Workaround

      Currently there is no known workaround for this behavior. A workaround will be added here when available

            [MIG-2247] CCMA Analytics throws "InvalidDataException: Memory is invalid. It shouldn't be a null number. Value provided null" in DC logs

            oliver.glowa added a comment -

            Hello,

            according to the change by 4cce94d4545f, 3 days ago:

            This issue does not affect functionality / performance of the DC instance

            I would like to know you, that we cannot really agree with this, 'cause we have analyzed the opposite. (on 9.2.6 LTS, DataCenterEdition)

            a) Log files

            Depending on how may Firefox users are working today, the size of the logfiles increase upto 50% per day.  So it is difficult to analyse the log on an incident, 'cause every 2nd/3rd entry is such a polluting text.

            b) Performance

            After we have disabled this addon, we have recognized, that that occurance of following problem has significant reduced:

            org.apache.catalina.valves.StuckThreadDetectionValve.notifyStuckThreadDetected Thread [http-nio-8080-exec-28 url:] (id=[6706]) has been active for [65,969] milliseconds (since [9/18/25, 11:07 AM]) to serve the same request for [https://… and may be stuck...

            Which leads to better system availabilty.

             

            So, we will leave that addon disabled until 2029 .

            Best Regards,
            Oliver

            oliver.glowa added a comment - Hello, according to the change by 4cce94d4545f , 3 days ago: This issue does not affect functionality / performance of the DC instance I would like to know you, that we cannot really agree with this, 'cause we have analyzed the opposite. (on 9.2.6 LTS, DataCenterEdition) a) Log files Depending on how may Firefox users are working today, the size of the logfiles increase upto 50% per day.  So it is difficult to analyse the log on an incident, 'cause every 2nd/3rd entry is such a polluting text. b) Performance After we have disabled this addon, we have recognized, that that occurance of following problem has significant reduced: org.apache.catalina.valves.StuckThreadDetectionValve.notifyStuckThreadDetected Thread [http-nio-8080-exec-28 url:] (id=[6706]) has been active for [65,969] milliseconds (since [9/18/25, 11:07 AM]) to serve the same request for [https: //… and may be stuck... Which leads to better system availabilty.   So, we will leave that addon disabled until 2029 . Best Regards, Oliver

            Abhishekh Shaw (Inactive) added a comment -

            Migration attempt using firefox was successful using local setup. Also as mentioned by Christine, the issue was resolved with the customer.

            Attached migration done from firefox locally.

            https://www.loom.com/share/9dfcf7c7bc6644bb8f9ff18ebf2d0347?sid=8d46e961-f62b-4c2c-bc02-63b5ea9113e5

            Abhishekh Shaw (Inactive) added a comment - Migration attempt using firefox was successful using local setup. Also as mentioned by Christine, the issue was resolved with the customer. Attached migration done from firefox locally. https://www.loom.com/share/9dfcf7c7bc6644bb8f9ff18ebf2d0347?sid=8d46e961-f62b-4c2c-bc02-63b5ea9113e5

            Christine Box added a comment -

            Hello,

            We have seen this issue and managed to resolve this with the customer (this may be an error that reports for multiple reasons, though).

            We saw this issue when there were multiple LDAP user directories, and it was caused by either 1. having a failed synchronisation or 2. having two user directories.

             

            We disabled the second user directory with a failed sync and the CCMA worked.

            Christine Box added a comment - Hello, We have seen this issue and managed to resolve this with the customer (this may be an error that reports for multiple reasons, though). We saw this issue when there were multiple LDAP user directories, and it was caused by either 1. having a failed synchronisation or 2. having two user directories.   We disabled the second user directory with a failed sync and the CCMA worked.

              Unassigned Unassigned
              jrichards@atlassian.com James Richards
              Affected customers:
              7 This affects my team
              Watchers:
              21 Start watching this issue

                Created:
                Updated: