Bug 553719
| Summary: | Very slow rhel5u3 paravirt guest installations with rhel5u4 z-stream kernel | ||||||
|---|---|---|---|---|---|---|---|
| Product: | Red Hat Enterprise Linux 5 | Reporter: | Gurhan Ozen <gozen> | ||||
| Component: | kernel-xen | Assignee: | Xen Maintainance List <xen-maint> | ||||
| Status: | CLOSED DUPLICATE | QA Contact: | Red Hat Kernel QE team <kernel-qe> | ||||
| Severity: | medium | Docs Contact: | |||||
| Priority: | medium | ||||||
| Version: | 5.4.z | CC: | bburns, drjones, emcnabb, jbao, jburke, qcai, rpacheco, xen-maint | ||||
| Target Milestone: | rc | ||||||
| Target Release: | --- | ||||||
| Hardware: | x86_64 | ||||||
| OS: | Linux | ||||||
| Whiteboard: | |||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||
| Doc Text: | Story Points: | --- | |||||
| Clone Of: | Environment: | ||||||
| Last Closed: | 2010-01-12 16:12:35 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: | |||||||
| Attachments: |
|
||||||
Created attachment 382537 [details]
/proc/cpuinfo
This may be the same issue as in bug 545692. We should try an install using phy (blkback) instead of blktap in the configuration and see if the install is faster. This machine wasn't pingable for me, so I couldn't try it myself right now. This is a dup of 545692. I got on the machine and see it has all the same symptoms, including the "blk_tap: invalid kernel buffer -- could not remap it" type logs. I could also install a guest quickly by using a loop device for the backing store, allowing me to use blkback instead of blktap. If I used blktap, it was very, very slow. This machine had a 5.3 host, so I installed a 5.3 guest, so it shows that it's not a regression (at least not a recent one). My guess is that this is a bug that is exposed by a particular hardware profile. I'm capturing the profile of this machine to compare against the other machine. *** This bug has been marked as a duplicate of bug 545692 *** |
Description of problem: This was brought to my attention when the rhts virtualization runs have timed out for the rhel5u3 paravirt guests. It does seem to be on Westmere-EP machines since we haven't yet to see the issue on other machines. These are the findings: with 2.6.18-164.9.1.el5xen kernel: -fast rhel5u4 pv installs, fast rhel5u hvm install, veryyyy slowwww rhel5u3 pv installs ... with 2.6.18-164.el5xen kernel (rhel5u4 release kernel): - no issues... Additionally with 2.6.18-128 kernel: -fast rhel5u3pv kernel Version-Release number of selected component (if applicable): 2.6.18-164.9.1.el5xen How reproducible: Everytime Steps to Reproduce: 1. Install rhel5u4 on a westmere-EP hardware, install 2.6.18-164.9.1.el5xen kernel and then try to install a rhel5u3 paravirt guest.