Description of problem:
Light swapping load (less than 20%) can result in oom-kills. This is very
noticable after upgrading the kernel from 4.5 to 4.6 thru 4.8.
Version-Release number of selected component (if applicable):
4.6, 4.7, 4.8
Steps to Reproduce:
1.allocate big mem allocs in 32bit
Premature oom-kills when 32bit servers with limited highmem are upgraded
beyond RHEL/OEL 4.5, to like, 4.7 or 4.8. These servers usually have 6GB and
a large portion (2GB) is allocated to hugepages for a database SGA.
The higemem zone order checks added by patch #2042 induce early oom-kills.
Related bugzilla 234572 which is not open to public.
hence filing this bug.
Reverting the patch # 2042 (linux-2.6.9-vm-excessive-swapout.patch), solves the oom-kills.
Thank you for submitting this issue for consideration in Red Hat Enterprise Linux. The release for which you requested us to review is now End of Life.
Please See https://access.redhat.com/support/policy/updates/errata/
If you would like Red Hat to re-consider your feature request for an active release, please re-open the request via appropriate support channels and provide additional supporting details about the importance of this issue.