This bug has been copied from bug #475567 and has been proposed to be backported to 5.3 z-stream (EUS).
Closing based on following statement: >>> Ok, keep in mind there was some history on this. The reason this >>> originally had the 5.3.z flag is because that's what a prod support >>> customer wanted, however Andy said no, so we went with the hotfix >>> packages on it. The 5.3.z flag was never removed and then someone >>> asked again if we were going to do it and Andy said, only if you >>> isolate a smaller set of patches for 5.3.z. So, Flavio did that (I >>> think ... some confusion here) and the latest plan was to push those >>> into the next 5.3.z. The original requestees have gone away, which >>> made your customer the biggest driving factor behind this request. >>> All that aside, do you think we should drop this request and do a >>> kbase article instead (indicating that TSO needs disabled)? Jiri
*** Bug 502196 has been marked as a duplicate of this bug. ***
Created attachment 350311 [details] Proposed patch Proposed patch for RHEL 5.3.z
Testing kernel builds are available at: http://people.redhat.com/ivecera/rhel-5z-ivtest/
Testing kernel from http://people.redhat.com/ivecera/rhel-5z-ivtest/kernel-2.6.18-128.1.14.el5.ivtest.x86_64.rpm w/ RHEL 5.3
I tried my testing kernel and issue seems to be solved. Joe, could you confirm this also?
Ivan: Sorry, yes, haven't seen any issues w/ this kernel. We ran 72+ hours of traffic "stress" and no issues were seen.