Bug 130357

Summary: High iowait and system load w/ moderate disk io
Product: Red Hat Enterprise Linux 3 Reporter: Jason W. Mitchell <jason+redhat-bugzilla>
Component: kernelAssignee: Doug Ledford <dledford>
Status: CLOSED DUPLICATE QA Contact: Brian Brock <bbrock>
Severity: medium Docs Contact:
Priority: medium    
Version: 3.0CC: petrides, riel
Target Milestone: ---   
Target Release: ---   
Hardware: athlon   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2006-02-21 14:05:10 EST Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: ---
Attachments:
Description Flags
/proc/cpuinfo
none
lspci output none

Description Jason W. Mitchell 2004-08-19 11:43:53 EDT
Description of problem:
- High iowait and cpu load w/ moderate disk io.  
- Similar to bug 129545 and bug121434, but critical difference 
  it that the system lacks RAID completely, but is still SMP.  
- Problem occurs if using either SMP or UP kernels listed below.
- during high iowait and lockup, kswapd is grabbing nearly all
  of the CPU

Version-Release number of selected component (if applicable):
kernel-smp-2.4.21-15.EL
kernel-smp-2.4.21-15.0.4.EL
kernel-smp-2.4.21-18.EL

System:
- Red Hat Enterprise Linux WS release 3 (Taroon Update 2); up2date
  current 2004/08/19
- ASLAB dual Athlon 1900; two IDE disks, separate channels.
- more info will be attached

How reproducible:
tar cf junk.tar /usr
watch top and feel the machine grind to a hault

Steps to Reproduce:
1.
2.
3.
  
Actual results:
machine grinds to hault and becomes unusable, or may crash completely

Expected results:
machine should not grind to hault with moderate disk io, should not
crash from moderate disk io.

Additional info:
- Temporary solution is to use Fedora Core 1: 
  kernel-smp-2.4.22-1.2199.nptl.athlon.rpm.
- tweaking a wide range of /proc/sys/vm/{min,max}-readahead failed
  to solve the problem
Comment 1 Jason W. Mitchell 2004-08-19 11:49:32 EDT
Created attachment 102881 [details]
/proc/cpuinfo
Comment 2 Jason W. Mitchell 2004-08-19 11:51:02 EDT
Created attachment 102882 [details]
lspci output
Comment 3 Doug Ledford 2004-09-02 16:56:03 EDT
This sounds like a dup of 121434 to me.  Marking as such.

*** This bug has been marked as a duplicate of 121434 ***
Comment 4 Red Hat Bugzilla 2006-02-21 14:05:10 EST
Changed to 'CLOSED' state since 'RESOLVED' has been deprecated.