Details
-
Bug
-
Status: Open (View Workflow)
-
Major
-
Resolution: Unresolved
-
10.4.24
-
None
-
None
-
None
Description
Reported by one of our customers that, getting invalid value at the query_time in the slow query logs, same is not matching with the value getting in the mysql.slow_log table.
slow_log table entry:
MariaDB [mysql]> SELECT db,`start_time`,`user_host`,`query_time`,`rows_sent`,`rows_examined`,`sql_text` FROM slow_log WHERE
|
-> `sql_text` LIKE "SELECT xxxxxxxxxx%"\G
|
*************************** 1. row ***************************
|
db: yds
|
start_time: 2023-07-13 11:48:07.379140
|
user_host: user2[user2] @ [10.xx.xx.xx]
|
query_time: 838:59:59.325276
|
rows_sent: 288081
|
rows_examined: 535868
|
Same query entry at the slow query log:
cat node1-from_slow.log.file_old | grep -i '18446744073709.324219' |head -n 5
|
# Time: 230713 11:48:07
|
# User@Host:user2[user2] @ [10.xx.xx.xx]
|
# Thread_id: 10917139 Schema: yds QC_hit: No
|
# Query_time: 18446744073709.324219 Lock_time: 0.119901 Rows_sent: 288081 Rows_examined: 535868
|
# Rows_affected: 0 Bytes_sent: 67299080
|
SET timestamp=1689274087;
|
SELECT `xxxxxxxxxxxx`
|
|
We are not clear how to reproduce this case, but it happening very frequently when one or the other queries in their workload.
One such jira ticket for this similar issue is - https://jira.mariadb.org/browse/MDEV-11552, but is old fixed already.