[SERVER-58475] Collect core dumps for out of memory errors Created: 13/Jul/21  Updated: 18/Nov/23

Status: Backlog
Project: Core Server
Component/s: None
Affects Version/s: None
Fix Version/s: None

Type: Improvement Priority: Major - P3
Reporter: Louis Williams Assignee: [DO NOT ASSIGN] Backlog - DevProd Correctness
Resolution: Unresolved Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Related
Assigned Teams:
Correctness
Participants:
Linked BF Score: 0

 Description   

If a suite fails due to an OOM, the server crashes. It would be useful to have core dumps to diagnose these crashes, as OOMs can be caused by heap corruption, not just memory leaks.



 Comments   
Comment by Alex Neben [ 05/Dec/22 ]

1. How important is this? It looks like the one BF this would help you solve has a score of 0. Are there other common things that this could help with?
2. Would this even work? If something is oom killed then it is dead and I am not sure how we can get a dump if it is dead. Lets say we caught it 1ms before it was killed, I assume getting a core dump would use more memory and then either the program or gdb would get killed.

Based on these two reasons I want to put it on our backlog. Let me know if I don't understand what this is asking or you don't think it belongs on our backlog.

Generated at Thu Feb 08 05:44:37 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.