Bug 193696

Summary: Not using all available system memory - swapping too aggressive - high load average (iowait)
Product: Red Hat Enterprise Linux 4 Reporter: David Herselman <bbs2web>
Component: kernelAssignee: Kernel Maintainer List <kernel-maint>
Status: CLOSED DUPLICATE QA Contact: Brian Brock <bbrock>
Severity: high Docs Contact:
Priority: medium    
Version: 4.0   
Target Milestone: ---   
Target Release: ---   
Hardware: i686   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2006-06-01 17:36:40 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 181409    
Attachments:
Description Flags
MRTG graph of available system memory show system not using everything before starting to aggressively swap none

Description David Herselman 2006-05-31 17:33:26 UTC
Description of problem:
  System swaps too aggressively and only utilises +-30% of available system
  memory (reading discussion threads indicates that this only shows up when
  the system has more than 1GB of RAM).

Version-Release number of selected component (if applicable):
  RHEL4.3
  kernel-2.6.9-34.0.1.EL

How reproducible:
  Very

Steps to Reproduce:
1. AMD Athlon 64 3000+ Processor
2. nVidia nForce4 chipset (ASUS A8N5X Motherboard)
3. 1GB of RAM
  
Actual results:
  put load on the server and available system memory is +-750KB. Swap
  utilisation increases and processes increase slowing the system due to
  high iowait (aggressive memory swapping)

Expected results:
  Should only start swapping once system memory starts getting low, no?

Additional info:
  System was previously running RHEL3.7 without problems, MRTG display of
  available system memory shows the problem quite clearly (attached).

  RedHat do have a document called 'Understanding Virtual Memory In Red Hat
  Enterprise Linux 4' but I simply don't understand what I should be changing:
    http://people.redhat.com/nhorman/papers/rhel4_vm.pdf

Comment 1 David Herselman 2006-05-31 17:33:26 UTC
Created attachment 130298 [details]
MRTG graph of available system memory show system not using everything before starting to aggressively swap

Comment 2 Jason Baron 2006-05-31 17:36:58 UTC
hmmm...this looks a lot like bz #188141. Can you please try the U4 beat kernel
and let me know if this fixes it for you. The kernel is 2.6.9-37.
http://people.redhat.com/~jbaron/rhel4/

Comment 3 David Herselman 2006-05-31 17:40:17 UTC
Found additional information from other people experiencing the same issue
in CentOS 4:
  Bug ID:  0001174
  Summary: Extremely high load average and 100% CPU utilization 
  URL:     http://bugs.centos.org/view.php?id=1174

Comment 4 Jason Baron 2006-05-31 17:54:42 UTC
Yes, i'm aware of this. I'm pretty sure the U4 beta kernel mention =ed in
comment #2 will resolve this issue. Also, just to be clear no VM tuning should
be needed. 

Comment 5 David Herselman 2006-05-31 20:24:06 UTC
The kernel-2.6.9-37.EL perfectly resolves the problem we were experiencing.

Many thanks for your prompt assistance!

Comment 6 Jason Baron 2006-06-01 17:36:40 UTC
glad to hear that :)

*** This bug has been marked as a duplicate of 188141 ***