Jason Fehr created IMPALA-14297:
-----------------------------------
Summary: ASAN heap-use-after-free in test_query_log.py
Key: IMPALA-14297
URL: https://issues.apache.org/jira/browse/IMPALA-14297
Project: IMPALA
Issue Type: Bug
Affects Versions: Impala 5.0.0
Reporter: Jason Fehr
heap-use-after-free detected by asan build:
==351307==ERROR: AddressSanitizer: heap-use-after-free on address
0x606000655170 at pc 0x000003774613 bp 0x7fe745df7170 sp 0x7fe745df7168
READ of size 8 at 0x606000655170 thread T234
The custom cluster test log file where this error was seen showed
"--cluster_id=test_query_hist_3". There are two tests that use this cluster id
"test_insert_select" and "test_dml", both in test_query_log.py.
The INFO log did not show any queries being executed. The last few lines show
this:
{noformat}
I20250811 05:26:30.675900 352764 impala-server.cc:3593] Shutdown status:
shutdown grace period left: 0, deadline left: 58s999ms, cancel deadline left:
46s999ms, queries registered on coordinator: 0, queries executing: 0, fragment
instances: 0
I20250811 05:26:30.675997 352764 workload-management-worker.cc:506] Workload
management is shutting down
I20250811 05:26:30.676074 352764 workload-management-worker.cc:517] Workload
management shutdown successful
I20250811 05:26:30.676167 352764 impala-server.cc:3619] Shutdown complete,
going down.
{noformat}
The issue appears to be a heap-use-after-free during workload management
shutdown.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
