[MDEV-21467] Rows_sent: 6148914691236517205 Created: 2020-01-13  Updated: 2020-02-15

Status: Open
Project: MariaDB Server
Component/s: None
Affects Version/s: 10.3.11
Fix Version/s: None

Type: Bug Priority: Major
Reporter: Vincent Milum Jr Assignee: Unassigned
Resolution: Unresolved Votes: 0
Labels: None
Environment:

CentOS 7



 Description   

I'm investigating some oddities in the slow query log on a client's server. I've come across this:

  1. Query_time: 649.534444 Lock_time: 0.000133 Rows_sent: 6148914691236517205 Rows_examined: 5263394
  2. Rows_affected: 0 Bytes_sent: 1154461212

This oddly high Rows_sent value is happening over and over again within their logs.

6148914691236517205 = 0x5555555555555555

This looks to be some sort of either debug value or default initialization value that isn't being set properly before being read.

Any thoughts? When I get more time later this afternoon, I'll try to filter out some stuff from the logs to hopefully share schema and query information (need to ensure no NDA stuff is shared)



 Comments   
Comment by Elena Stepanova [ 2020-02-14 ]

darkain,
Have you had any luck with narrowing down the schema or whatever you can share?

Comment by Vincent Milum Jr [ 2020-02-14 ]

This is a very active server

Threads: 159 Questions: 3348031299451 Slow queries: 463304 Opens: 9898153 Flush tables: 1 Open tables: 1509 Queries per second avg: 112047.691

The main query in question hitting the slow log is short, but a little complicated. The columns returned are passed through stored functions. There is a subquery. There is JOINs. The WHERE clause uses more custom function calls. I'm going to try to break it out and eliminate each section to see if I can build a test case that still causes the issue.

Generated at Thu Feb 08 09:07:22 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.