I completely missed that, thanks Hanlin!
For perf tests always use release images.
Regards,
Florin
> On Jan 20, 2022, at 6:27 PM, weizhen9...@163.com wrote:
>
> Yes, I use a debug version.
> Thank you.
>
>
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
Yes, I use a debug version.
Thank you.
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages sent to this group.
View/Reply Online (#20772): https://lists.fd.io/g/vpp-dev/message/20772
Mute This Topic: https://lists.fd.io/mt/88456731/21656
Group Owner: vpp-dev+ow...@lists.fd.io
Unsubscribe:
Latency with vpp is very high for ~50% of the requests so either some config is
still off or something loses packets.
To check if the later is the case, try “show error” and check tcp counters for
out of order enqueues or “show tcp stats” and check retransmission counters.
To make sure we
Hi,
I see that some configs have been changed compared to our private conversation.
Did that influence results?
Nonetheless I’d try with the configs lower. In particular, I noticed you
changed default buffer size, numa heap size and remove multi-seg support which
are not optimal.
Regards,
The version of vpp is:
The config of vpp and nginx is the attachment. The vcl config is attachment.
The type of nic is:
nginx.conf
Description: Binary data
startup.conf
Description: Binary data
vcl_test.conf
Description: Binary data
-=-=-=-=-=-=-=-=-=-=-=-
Links: You receive all messages
Hi,
Numbers are pretty low, but it’s hard to say just from those what might be
wrong. So let’s try to clarify some things:
- what vpp version are you running, how many workers and what’s the config?
- what’s the vcl config? Are you you using something similar to the one here
[1]?
- I’m not