Welcome to the #dominoforever Product Ideas Lab! The place where you can submit product ideas and enhancement request. We encourage you to participate by voting on, commenting on, and creating new ideas. All new ideas will be evaluated jointly by the IBM & HCL Product Management & Engineering teams, and the next steps will be communicated. While not all submitted ideas will be executed upon, community feedback will play a key role in influencing which ideas are and when they will be implemented.

For more information and upcoming events around #dominoforever, please visit our Destination Domino page.


Better NSD crash diagnostic logs collection

We frequently encountered this situation where Domino crashed due to abnormal termination and IBM Support's reply is always the NSD log is insufficient to provide details on the crash and we need to implement NSD -monitor to troubleshoot further.

 

The problem is such crash does not happen often, it is not feasible to implement NSD -monitor for weeks or months waiting for the crash that might not happen again and further more NSD -monitor does not work with Domino running as a service.

 

1. Would appreciate a better NSD in Domino 10, to able to capture the necessary logs without the need to implement additional steps for crash diagnostic.

2. Ability to run manual NSD from Domino console instead of the need to login to OS and trigger NSD over there.

3. Automated NSD log collection if for example lkmgr messages, http waiting for session to complete appears.

  • Avatar32.5fb70cce7410889e661286fd7f1897de Guest
  • Jul 31 2018
  • Attach files
  • Avatar40.8f183f721a2c86cd98fddbbe6dc46ec9
    Guest commented
    31 Jul 13:38

    From my side, I can see a specific scenario to enhance:

    In these Domino server crashes I've seen, there is a too high percentace of these crashes, where a Domino process just dumped code (e.g. http), the process disappears from the OS's process list, and then Domino's ProcessMonitorTask notices that one of it's child PID's is missing.

    In these cases, Domino will issue a Panic, which is the correct behaviour.

    However, there is no useful evidence in the NSD file regarding the missing child process (because it's already gone). In these cases, the root cause of the crash cannot be evaluated, or it is very hard.

    A set of proper crash dump files should allow a skilled technician (or IBM support) to evaluate the crash root cause in almost any case.

    A desirable solution could be to avoid most cases, where a core dump will happen, and instead issue a controlled Panic directly from the problematic Domino process.

    Additionally, it would be good if NSD was able to auto-process core files belonging to the current crash incident, including the processed data in the NSD file.

  • Admin
    Thomas Hampel commented
    08 Aug 16:11
  • Avatar40.8f183f721a2c86cd98fddbbe6dc46ec9
    Guest commented
    08 Aug 18:21

    @Thomas Hampel I‘ve used LND. It acts as a nicer frontend/analysis tool. Hence, it would be nice to have it more accessible, or even as an installable option in the future.

    However, it does not resolve the issues underneath. In my specific problems, and apparently those problems reported by the idea creator, LND does not help.