Bug 761591

Summary: xen-netfront: wait with gARP notification until backend moves to Connected status [TestOnly]
Product: Red Hat Enterprise Linux 7 Reporter: Laszlo Ersek <lersek>
Component: kernelAssignee: Red Hat Kernel Manager <kernel-mgr>
Status: CLOSED CURRENTRELEASE QA Contact: Virtualization Bugs <virt-bugs>
Severity: medium Docs Contact:
Priority: medium    
Version: 7.0CC: drjones, kernel-mgr, kmcmartin, leiwang, lersek, lkong, pbonzini, wshi, xen-maint
Target Milestone: rcKeywords: EC2, TestOnly
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard: xen
Fixed In Version: v3.3-rc1 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: 713585 Environment:
Last Closed: 2014-06-13 07:35:49 EDT Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Bug Depends On: 866735    
Bug Blocks: 741684    

Comment 9 Lingfei Kong 2014-02-14 02:05:10 EST
I did migration test on RHEL5.10 with RHEL7.0, it got 7-15 packet lost during the migration.
Also I have try to set the net.ipv4.conf.{all,default,IFACE}.arp_notify sysctl to 1 in the guest and did migration test again but the network still lost about 7-15 packet. 
Is it acceptable with lost 7-15 packet? 

Host: kernel-xen-2.6.18-380.el5
Guest: kernel-3.10.0-86.el7

Here are the details:
1. Start nfs in System A and mount the nfs to System B.
  #mount $SystemA_ip:/var/lib/xen/images /var/lib/xen/images

2. Edit "/etc/xen/xend-config.sxp" on two machines as following:
   (xend-relocation-server yes)
   (xend-relocation-port 8002)
   (xend-relocation-address '')
   (xend-relocation-hosts-allow '')
# service xend restart

3. Do migration operations
System A to System B:
--- 10.66.85.222 ping statistics ---
23 packets transmitted, 16 received, 30% packet loss, time 24355ms
rtt min/avg/max/mdev = 0.552/2.155/18.863/4.389 ms

System B  to System A:
--- 10.66.85.222 ping statistics ---
25 packets transmitted, 15 received, 40% packet loss, time 27388ms
rtt min/avg/max/mdev = 0.000/2.422/10.114/3.199 ms

System A to System B:
--- 10.66.85.222 ping statistics ---
20 packets transmitted, 14 received, 30% packet loss, time 22685ms

System B to  System A:
--- 10.66.85.222 ping statistics ---
26 packets transmitted, 16 received, 38% packet loss, time 27730ms
rtt min/avg/max/mdev = 0.000/4.141/49.098/11.691 ms
Comment 10 Andrew Jones 2014-02-14 03:06:06 EST
(In reply to Lingfei Kong from comment #9)
> I did migration test on RHEL5.10 with RHEL7.0, it got 7-15 packet lost
> during the migration.
> Also I have try to set the net.ipv4.conf.{all,default,IFACE}.arp_notify
> sysctl to 1 in the guest and did migration test again but the network still
> lost about 7-15 packet. 
> Is it acceptable with lost 7-15 packet? 

Yeah, that's fine. Thanks.
Comment 11 Lingfei Kong 2014-02-14 03:17:42 EST
Because 7-15 packet lost is acceptable, so change status to VERIFIED.
Comment 12 Ludek Smid 2014-06-13 07:35:49 EDT
This request was resolved in Red Hat Enterprise Linux 7.0.

Contact your manager or support representative in case you have further questions about the request.