[ovs-discuss] [bug]ovs-dpdk can't open dpdk port and dpdkvhostuser port added before
Flavio Leitner
fbl at sysclose.org
Thu Feb 25 13:08:50 UTC 2016
On Thu, 25 Feb 2016 06:22:57 +0000
王华夏 <wanghuaxia at jd.com> wrote:
> HI
>
> I am trying to restart ovs ,find error as fellows, all thess bridges and ports I have added before
>
> Bridge br-dpdk
> Port "dpdk0"
> Interface "dpdk0"
> type: dpdk
> error: "could not open network device dpdk0 (Cannot allocate memory)"
> Port br-dpdk
> Interface br-dpdk
> type: internal
> Bridge br-tun
> fail_mode: secure
> Port "vxlan-0a001f8d"
> Interface "vxlan-0a001f8d"
> type: vxlan
> options: {df_default="true", in_key=flow, local_ip="10.0.32.13", out_key=flow, remote_ip="10.0.31.141"}
> Port "vxlan-0a001f7e"
> Interface "vxlan-0a001f7e"
> type: vxlan
> options: {df_default="true", in_key=flow, local_ip="10.0.32.13", out_key=flow, remote_ip="10.0.31.126"}
> Port "vxlan-0a002016"
> Interface "vxlan-0a002016"
> type: vxlan
> options: {df_default="true", in_key=flow, local_ip="10.0.32.13", out_key=flow, remote_ip="10.0.32.22"}
> Port patch-int
> Interface
> patch-int
> type: patch
> options: {peer=patch-tun}
> Port br-tun
> Interface br-tun
> type: internal
> Bridge br-int
> fail_mode: secure
> Port br-int
> Interface br-int
> type: internal
> Port "tap78ec2293-ae"
> tag: 1
> Interface "tap78ec2293-ae"
> type: dpdkvhostuser
> error: "could not open network device tap78ec2293-ae (Unknown error -1)"
> Port "tapb94777b3-5d"
> tag: 1
>
>
> When start the ovs like this ,this problem will occurs:
>
> ovs-vswitchd --dpdk -c 0x1 -n 4 --socket-mem=2048,2048 -- unix:/var/run/openvswitch/db.sock --pidfile --detach -vconsole:emer -vsyslog:err -vfile:info --mlockall --no-chdir --log-file=/var/log/openvswitch/ovs-vswitchd.log –monitor
>
> if change the --socket-mem=2048,2048 to –socket-mem= 2048 , it will have no problem
>
> if it means the numa node1 can not use ?
>
> my ovs version is 2.5 and dpdk version is 2.2 with kernel Linux version 3.10.0-229.20.1.el7.x86_64 (builder at kbuilder.dev.centos.org) (gcc version 4.8.3 20140911 (Red Hat 4.8.3-9) (GCC) ) #1 SMP Tue Nov 3 19:10:07 UTC 2015
>
It depends on how you are allocating the hugepages and also how you are
restarting OVS. If it's crashing, there might be some files still
allocating memory in /dev/hugepages/*.
For instance, you might have allocated more pages in socket#0 and not enough
in socket#1.
--
fbl
More information about the discuss
mailing list