Bug #28751 Lots of memory locked in memory causes high kswapd
Submitted: 29 May 2007 17:42 Modified: 11 Jun 2007 22:36
Reporter: Tomas Ulin
Status: Closed
Category:Server: Cluster Severity:S3 (Non-critical)
Version: OS:Any
Assigned to: Tomas Ulin Target Version:

[29 May 2007 17:42] Tomas Ulin 
Description:
We observe high kswapd on some systems when ndbd has lots of memory locked in memory
(close to system limit)

We see that using O_DIRECT writes lowers kswapd

How to repeat:
.

Suggested fix:
optionally enable O_DIRECT writing for LCP, redo log, and backup
(disk data already uses O_DIRECT)

is has been noticed that switching of NUMA removes kswapd as well
[5 Jun 2007 17:33] Bugs System 
A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/28130
[11 Jun 2007 13:39] Bugs System 
Pushed into 5.1.20-beta
[11 Jun 2007 22:36] Jon Stephens 
Thank you for your bug report. This issue has been committed to our source repository of
that product and will be incorporated into the next release.

If necessary, you can access the source repository and build the latest available version,
including the bug fix. More information about accessing the source trees is available at

    http://dev.mysql.com/doc/en/installing-source.html

Documented bugfix in 5.1.20 and telco-6.1.11 changelogs.
[3 Jul 2007 8:47] Jon Stephens 
Also documented for telco-6.2.3 release.