Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 25 May 2022 09:04:37 +0000
From:      bugzilla-noreply@freebsd.org
To:        virtualization@FreeBSD.org
Subject:   [Bug 264198] bridge, ipfw: very slow network on bridge with IPFW, virtio or vtnet on 12.3-RELEASE-p5
Message-ID:  <bug-264198-27103-830AhCYGk3@https.bugs.freebsd.org/bugzilla/>
In-Reply-To: <bug-264198-27103@https.bugs.freebsd.org/bugzilla/>
References:  <bug-264198-27103@https.bugs.freebsd.org/bugzilla/>

next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D264198

--- Comment #4 from Ole <ol@dbconn.net> ---
I could reduce it even more. Seems that the vlan play no role here. I picked
one VM on the hosts. Stopped IPFW on the hosts so network is fine. Then I
crated a bridge and an epair like before:

vtnet0---bridge0---epaira0---epairb0


vm # ifconfig bridge create mtu 1400
vm # ifconfig epair create mtu 1400
vm #  ifconfig bridge0 addm epair0a
vm # ifconfig bridge0 addm vtnet0
vm # ifconfig epair0a up
vm # ifconfig epair0b inet <IP_ADDR>/28 mtu 1400 up
vm # ifconfig bridge0 up
vm # nc -v -l 2222 > /dev/null

again, the only rule is
vm # ipfw list
65535 allow ip from any to any

from another host (in the same network) I send data with nc:
dd if=3D/dev/zero bs=3D1024k count=3D6 | nc -v <IP_ADDR> 2222

result:
6291456 bytes transferred in 70.693446 secs (88996 bytes/sec)

after 'service ipfw stop' on vm-a, it is about 700 times faster )-:
6291456 bytes transferred in 0.100673 secs (62493675 bytes/sec)


So I think the problem is related to IPFW and bridges. Maybe the MTU plays a
role here.

--=20
You are receiving this mail because:
You are the assignee for the bug.=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-264198-27103-830AhCYGk3>