Bug 699952

Summary: decrease migration_downtime cause migrtion speed very slow
Product: Red Hat Enterprise Linux 6 Reporter: Chao Yang <chayang>
Component: qemu-kvmAssignee: Virtualization Maintenance <virt-maint>
Status: CLOSED NOTABUG QA Contact: Virtualization Bugs <virt-bugs>
Severity: medium Docs Contact:
Priority: medium    
Version: 6.1CC: bcao, juzhang, michen, mkenneth, pbonzini, quintela, virt-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: 2011-04-27 10:10:20 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description Chao Yang 2011-04-27 02:43:08 UTC 
Description of problem:
Hit same issue on RHEL 6.1 host with .
When setting migration downtime to 0.0000001, migration works fine for me, but setting to 0.0000000001, migration speeds slow down to less than 100Kb/s


Version-Release number of selected component (if applicable):
kernel 2.6.32-132.el6.x86_64, qemu-kvm qemu-kvm-0.12.1.2-2.158.el6.x86_64

How reproducible:
100%

Steps to Reproduce:
1. start a guest by following cli:
/usr/libexec/qemu-kvm -M rhel6.1.0 -enable-kvm -m 4096 -smp 4 -name rhel6.1 -uuid `uuidgen` -rtc base=utc,clock=host,driftfix=slew -no-kvm-pit-reinjection -boot c -drive file=/root/RHEL-Server-6.1-64-virtio.raw,if=none,id=drive-virtio0-0-0,media=disk,format=raw,cache=none -device virtio-blk-pci,drive=drive-virtio0-0-0,id=virt0-0-0 -netdev tap,id=hostnet1 -device virtio-net-pci,netdev=hostnet1,id=net1,mac=52:54:40:13:15:53 -usb -device usb-tablet,id=input1 -vnc :6 -monitor stdio -balloon none
2. start another guest in listening mode
 -incoming tcp:0:5880
3. (qemu)migrate_set_downtime 0.0000000001
4. check migration speed

Actual results:
migrate speed is less than 100Kb/s  

Expected results:
Should affect migration speed.

Additional info:
Comment 1 Chao Yang 2011-04-27 02:46:33 UTC 
> 
> Expected results:
> Should affect migration speed.
> 

---> Should not affect migration speed.
Comment 2 Paolo Bonzini 2011-04-27 10:10:20 UTC 
Migration downtime affects migration speed because qemu has more work to do between iterations (checking the dirty bitmap, etc.).  Besides, a downtime of 10^-10 seconds makes little sense when a clock cycle is three times that.
Comment 3 Juan Quintela 2011-05-23 07:39:07 UTC 
Relation here is that we we are never getting few enough pages to migrate in 0.0000001 secondos (that should be one page or something ridiculous like that).  So we try to get to that single one page and stay forever trying (making final speed really small).