Environment
YugabyteDB Versions:
- 2.14.12.0 and older
- 2.16.7.0 and older
- 2.18.3.0 and older
Issue
Tserver/master crash and the following stack trace is being reported in core dump or tserver.err log file.
*** SIGSEGV (@0x1f40) received by PID 84409 (TID 0x7f4672873700) from PID 8000; stack trace: ***
@ 0x7f523b6f3cd9 (unknown)
@ 0x7f523b6f55f8 _Unwind_Backtrace
@ 0x7f523be152bf __GI___backtrace
@ 0x55d5359892f3 yb::(anonymous namespace)::HandleStackTraceSignal()
@ 0x7f523bd55120 (unknown)
@ 0x7f523b90db09 __lll_lock_wait_private
@ 0x7f523b906ab1 __pthread_create_2_1
We use the backtrace function to collect stack traces of long running operations and to show list of running threads in the UI. But when this function runs in the middle of thread creation or during thread start or thread exit it causes SIGSEGV segmentation faults.
Resolution
Overview
This issue is fixed in YugabyteDB version's 2.14.13.0, 2.16.8.0, 2.18.4.0 and above.
References
Github Issue filed to address this issue- https://github.com/yugabyte/yugabyte-db/issues/18668
Comments
0 comments
Please sign in to leave a comment.