Note: This bug is displayed in read-only format because the product is no longer active in Red Hat Bugzilla.
The FDP team is no longer accepting new bugs in Bugzilla. Please report your issues under FDP project in Jira. Thanks.

Bug 1792399

Summary: NFV live migration fails with dpdk "--iova-mode va": Failed to load virtio-net:virtio
Product: Red Hat Enterprise Linux Fast Datapath Reporter: Timothy Redaelli <tredaelli>
Component: openvswitch2.11Assignee: Adrián Moreno <amorenoz>
Status: CLOSED CURRENTRELEASE QA Contact: Pei Zhang <pezhang>
Severity: high Docs Contact:
Priority: high    
Version: FDB 18.04CC: aadam, amorenoz, atragler, chayang, ctrautma, dmarchan, jhsiao, jinzhao, juzhang, kanderso, kfida, ktraynor, maxime.coquelin, nhorman, ovs-qe, pezhang, qding, ralongi, tredaelli
Target Milestone: ---Keywords: Regression
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard:
Fixed In Version: openvswitch2.11-2.11.0-40.el7fdn Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: 1763815
: 1796590 (view as bug list) Environment:
Last Closed: 2020-03-11 18:08:24 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:

Comment 1 Pei Zhang 2020-02-05 11:21:11 UTC 
Steps: Following steps of Description. We tested with Live migration with ovs+vhost-user+dpdk scenarios.

Reproduced with openvswitch2.11-2.11.0-39.el7fdp.x86_64:

# /bin/virsh migrate --verbose --persistent --live rhel7.8 qemu+ssh://10.73.72.196/system
Migration: [100 %]error: internal error: qemu unexpectedly closed the monitor: 2020-02-05T11:10:10.812975Z qemu-kvm: -chardev socket,id=charnet1,path=/tmp/vhostuser0.sock,server: info: QEMU waiting for connection on: disconnected:unix:/tmp/vhostuser0.sock,server
2020-02-05T11:10:10.941050Z qemu-kvm: -chardev socket,id=charnet2,path=/tmp/vhostuser1.sock,server: info: QEMU waiting for connection on: disconnected:unix:/tmp/vhostuser1.sock,server
2020-02-05T11:10:25.970886Z qemu-kvm: VQ 0 size 0x400 < last_avail_idx 0x5e93 - used_idx 0x7d6c
2020-02-05T11:10:25.970918Z qemu-kvm: Failed to load virtio-net:virtio
2020-02-05T11:10:25.970936Z qemu-kvm: error while loading state for instance 0x0 of device '0000:00:06.0:00.0/virtio-net'
2020-02-05T11:10:25.973035Z qemu-kvm: load of migration failed: Operation not permitted

Verified with openvswitch2.11-2.11.0-47.el7fdp.x86_64:

Migration works very well. No any error.

# /bin/virsh migrate --verbose --persistent --live rhel7.8 qemu+ssh://10.73.72.196/system
Migration: [100 %]


Note:
Testpmd cmd in guest, testing with "--iova-mode va".
/usr/bin/testpmd \
-l 1,2,3,4,5 \
-n 4 \
-d /usr/lib64/librte_pmd_virtio.so \
-w 0000:06:00.0 -w 0000:07:00.0 \
--iova-mode va \
-- \
--nb-cores=4 \
-i \
--disable-rss \
--rxd=512 --txd=512 \
--rxq=2 --txq=2


So this bug has been fixed very well.
Comment 2 Pei Zhang 2020-02-07 10:14:58 UTC 
We also tested rhel8 version, this issue has been fixed well too.

openvswitch2.11-2.11.0-47.el8fdp.x86_64 works well
Comment 3 Pei Zhang 2020-02-11 06:37:29 UTC 
Move to 'VERIFIED' as Comment 1 and Comment 2.
Comment 4 Timothy Redaelli 2020-03-11 18:08:24 UTC 
In FDP 20.B