[ovs-discuss] ovs offloading stops suddenly stops working

Manuel Sopena Ballesteros manuel.sb at garvan.org.au
Fri Mar 22 05:18:20 UTC 2019


Dear OVS community,

I am trying to setup OVS offload on a mellanox connectX4-LX through 2 vms suing SR-IOV to the VFs.

For some reason when I start pinging the offload works but after a while I am getting this error message:

[root at zeus-54 ~]# systemctl status openvswitch
● openvswitch.service - LSB: Open vSwitch switch
   Loaded: loaded (/etc/rc.d/init.d/openvswitch; bad; vendor preset: disabled)
   Active: active (running) since Fri 2019-03-22 16:10:34 AEDT; 36s ago
     Docs: man:systemd-sysv-generator(8)
  Process: 36025 ExecStop=/etc/rc.d/init.d/openvswitch stop (code=exited, status=0/SUCCESS)
  Process: 36173 ExecStart=/etc/rc.d/init.d/openvswitch start (code=exited, status=0/SUCCESS)
Main PID: 4423 (code=exited, status=0/SUCCESS)
    Tasks: 61
   CGroup: /system.slice/openvswitch.service
           ├─36197 ovsdb-server: monitoring pid 36198 (healthy)
           ├─36198 ovsdb-server /etc/openvswitch/conf.db -vconsole:emer -vsyslog:err -vfile:info --remote=punix:/var/run/openvswitch/db.sock --private-key=db:Open_vSwitch,SSL,private_key --certificate=db:Open_vSwitch,SSL,certificate --bootstrap-ca-ce...
           ├─36213 ovs-vswitchd: monitoring pid 36214 (healthy)
           └─36214 ovs-vswitchd unix:/var/run/openvswitch/db.sock -vconsole:emer -vsyslog:err -vfile:info --mlockall --no-chdir --log-file=/var/log/openvswitch/ovs-vswitchd.log --pidfile=/var/run/openvswitch/ovs-vswitchd.pid --detach --monitor

Mar 22 16:10:33 zeus-54.localdomain openvswitch[36173]: Starting ovsdb-server [  OK  ]
Mar 22 16:10:33 zeus-54.localdomain ovs-vsctl[36199]: ovs|00001|vsctl|INFO|Called as ovs-vsctl --no-wait -- init -- set Open_vSwitch . db-version=7.16.1
Mar 22 16:10:33 zeus-54.localdomain ovs-vsctl[36204]: ovs|00001|vsctl|INFO|Called as ovs-vsctl --no-wait set Open_vSwitch . ovs-version=2.11.0 "external-ids:system-id=\"585ddcbc-8ee9-435e-9cfa-4df0f34969f5\"" "external-ids:rundir=\"/...em-version=\"7\""
Mar 22 16:10:33 zeus-54.localdomain openvswitch[36173]: Configuring Open vSwitch system IDs [  OK  ]
Mar 22 16:10:34 zeus-54.localdomain openvswitch[36173]: Starting ovs-vswitchd [  OK  ]
Mar 22 16:10:34 zeus-54.localdomain openvswitch[36173]: Enabling remote OVSDB managers [  OK  ]
Mar 22 16:10:34 zeus-54.localdomain ovs-vsctl[36340]: ovs|00001|vsctl|INFO|Called as ovs-vsctl --no-wait set Open_vSwitch . external-ids:hostname=localhost
Mar 22 16:10:34 zeus-54.localdomain systemd[1]: Started LSB: Open vSwitch switch.
Mar 22 16:11:06 zeus-54.localdomain ovs-vswitchd[36214]: ovs|00001|dpif_netlink(handler60)|ERR|failed to offload flow: Invalid argument: ens2f0_0
Mar 22 16:11:06 zeus-54.localdomain ovs-vswitchd[36214]: ovs|00002|dpif_netlink(handler60)|ERR|failed to offload flow: Invalid argument: ens2f0_1
Hint: Some lines were ellipsized, use -l to show in full.

This is an example when I ping between 2 vms on same host:

64 bytes from 10.0.32.201: icmp_seq=501 ttl=64 time=0.142 ms <<<<<<<<<<<<< OVS OFFLOAD WORKING
64 bytes from 10.0.32.201: icmp_seq=502 ttl=64 time=0.140 ms
64 bytes from 10.0.32.201: icmp_seq=503 ttl=64 time=0.140 ms
64 bytes from 10.0.32.201: icmp_seq=504 ttl=64 time=0.140 ms
64 bytes from 10.0.32.201: icmp_seq=505 ttl=64 time=0.136 ms
64 bytes from 10.0.32.201: icmp_seq=506 ttl=64 time=0.164 ms
64 bytes from 10.0.32.201: icmp_seq=507 ttl=64 time=0.136 ms
64 bytes from 10.0.32.201: icmp_seq=508 ttl=64 time=0.431 ms <<<<<<<<<<<<< OVS OFFLOAD STOPS WORKING
64 bytes from 10.0.32.201: icmp_seq=509 ttl=64 time=0.370 ms
64 bytes from 10.0.32.201: icmp_seq=510 ttl=64 time=0.419 ms
64 bytes from 10.0.32.201: icmp_seq=511 ttl=64 time=0.550 ms
64 bytes from 10.0.32.201: icmp_seq=512 ttl=64 time=0.470 ms
64 bytes from 10.0.32.201: icmp_seq=513 ttl=64 time=0.455 ms
64 bytes from 10.0.32.201: icmp_seq=514 ttl=64 time=0.489 ms
64 bytes from 10.0.32.201: icmp_seq=515 ttl=64 time=0.377 ms
64 bytes from 10.0.32.201: icmp_seq=516 ttl=64 time=0.567 ms
64 bytes from 10.0.32.201: icmp_seq=517 ttl=64 time=0.389 ms
64 bytes from 10.0.32.201: icmp_seq=518 ttl=64 time=0.391 ms

I am not an expert on these topics so apologize beforehand for the lack of information provided (I skipped the OVS/kvm configuration for the sake of simplicity but happy to put it here is needed)

Could someone please advise what “Mar 22 16:11:06 zeus-54.localdomain ovs-vswitchd[36214]: ovs|00001|dpif_netlink(handler60)|ERR|failed to offload flow: Invalid argument: ens2f0_0” means?

Thank you very much

Manuel

NOTICE
Please consider the environment before printing this email. This message and any attachments are intended for the addressee named and may contain legally privileged/confidential/copyright information. If you are not the intended recipient, you should not read, use, disclose, copy or distribute this communication. If you have received this message in error please notify us at once by return email and then delete both messages. We accept no liability for the distribution of viruses or similar in electronic communications. This notice should not be removed.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openvswitch.org/pipermail/ovs-discuss/attachments/20190322/63ca76a4/attachment-0001.html>


More information about the discuss mailing list