-
Type: Bug
-
Resolution: Fixed
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: None
-
None
-
Fully Compatible
-
ALL
-
-
Service Arch 2022-10-17, Service Arch 2022-10-31, Service Arch 2022-11-14, Service Arch 2022-11-28, Service Arch 2022-12-12, Service Arch 2023-02-06
-
162
(To fix BF-26477)
The service_executor_bm C++ benchmark creates service worker threads in a loop, measuring the speed of this operation.
Under ASAN this causes an accumulation of mmap-resident AsanThread objects that ASAN uses to track each thread, exhuasting the resource limit after roughly 64ki threads have been made. There are not 64k threads in the system. There are only a few at a time, but the ASAN-allocated resources of dead threads linger.
This may be related to these service worker threads being detached.
It may be some other leak of the pthreads in the code under test.
- causes
-
SERVER-71702 adjust max_conns_override.js expectation
- Closed
- depends on
-
SERVER-70804 Improve SessionWorkflow benchmarks
- Closed
- is depended on by
-
SERVER-68875 Reflow the SessionWorkflow loop
- Closed
- is duplicated by
-
SERVER-71381 ReservedServiceExecutor: actively recover from spawn failure
- Closed
- is related to
-
SERVER-69905 Improve ServiceExecutorSynchronous worker threads
- Backlog