[MXS-378] Serious leak that causes MaxScale to crash Created: 2015-09-23  Updated: 2015-09-24  Resolved: 2015-09-24

Status: Closed
Project: MariaDB MaxScale
Component/s: Core
Affects Version/s: 1.2.1
Fix Version/s: 1.2.1

Type: Bug Priority: Blocker
Reporter: Johan Wikman Assignee: Johan Wikman
Resolution: Not a Bug Votes: 0
Labels: None


 Description   

From https://support.mariadb.com/view.php?id=10125

"Over the past several weeks we've been doing load testing. What we've noticed is that the amount of memory consume by maxscale is consistently increasing. See attached screenshot. We've more than doubled the amount of RAM on maxscale from 8G to 20G and while it does take longer to use 20G during the endurance testing maxscale does eventually use the RAM and restart the service. This causes error with our application and invalides the load test. I've also noticed that the connection count is staying pretty low. The RAM does not start to climb until load testing has started, IE when there is not connections maxscale RAM stays flat. Can we please escalate this to the maxscale team? We did not see this behavior when we originally tested. "



 Comments   
Comment by Johan Wikman [ 2015-09-24 ]

The excess memory consumption was due to a confguration issue and not a leak.

Generated at Thu Feb 08 03:58:50 UTC 2024 using Jira 8.20.16#820016-sha1:9d11dbea5f4be3d4cc21f03a88dd11d8c8687422.