-
Type:
Bug
-
Resolution: Unresolved
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
Storage Engines
-
StorEng - Defined Pipeline
As part of SPM-3810, we realised that calling the dump() command was crashing when we called it as part of drop(). Stack trace shows failing assertion as below:
[j1:prim] | 2025-02-04T03:22:42.003+00:00 F CONTROL 6384300 [conn30] "Writing fatal message","attr": { message: "Invalid access at address: 0xd8\n"
It looks as though we are trying to access a null pointer somewhere. The stack trace also looks as if it is failing within `tree_walk_internal`:
{"a":"FFFFAD169810","b":"FFFFAD169000","o":"810","s":"__kernel_rt_sigreturn","s+":"0"}
,{"a":"AAAAB97E4D28","b":"AAAAB4B60000","o":"4C84D28","s":"_tree_walk_internal","s+":"48"},{"a":"AAAAB985A8D0","b":"AAAAB4B60000","o":"4CFA8D0","s":"_verbose_dump_cache_apply.constprop.0","s+":"1B4"},
Full patch build linked here. Most failures have the same stack trace as reported above.
- is depended on by
-
SERVER-100390 Add dump after failing drop
-
- Backlog
-