[SERVER-57166] abort() in tcmalloc when allocated memory size is ridiculously large Created: 24/May/21  Updated: 09/Jun/21  Resolved: 09/Jun/21

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

Type: Task Priority: Major - P3
Reporter: Sergey Galtsev (Inactive) Assignee: Sergey Galtsev (Inactive)
Resolution: Duplicate Votes: 1
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original Estimate: Not Specified

Issue Links:
Depends
Duplicate
is duplicated by SERVER-57586 Enforce termination on all allocation... Closed
Related
is related to SERVER-57586 Enforce termination on all allocation... Closed
Sprint: Security 2021-05-31, Security 2021-06-14
Participants:
Linked BF Score: 21

 Description   

In BF-20042 we was a memory corruption issue correlated with tcmalloc: large alloc 3761212780863488 bytes == (nil) @. Memory requested amounts to ~3.7 petabytes. A core dump followed during cache release. It is unclear, where does such large allocation originates, but it clearly is a bug.

In order to make debugging simpler we must fail-fast when it happens, so that a core dump is produced and location of fault is easier to identify.

For the purposes of this ticket, a code change should be limited in functionality to evergreen-only invocations by the use of environment variable.



 Comments   
Comment by Sergey Galtsev (Inactive) [ 09/Jun/21 ]

Duplicate of SERVER-57586

Generated at Thu Feb 08 05:41:09 UTC 2024 using Jira 9.7.1#970001-sha1:2222b88b221c4928ef0de3161136cc90c8356a66.