[Flent-users] [tohojo/flent] packet loss stats (#106)
notifications at github.com
Thu Nov 16 14:15:45 EET 2017
Pete Heist <notifications at github.com> writes:
> Cool, the rrul_be_irtt test is working for me. I see what you mean,
> ideally it would just be the same test but you substitute the tool you
> want to use for a particular measurement.
> Maybe this is what you're saying, but could it be that you specify
> what tool you want to use for each particular measurement? In other
> words, the tool for 'TCP throughput' is netperf as of now, and later
> could possibly be iperf. For 'UDP RTT' it can be netperf or irtt. For
> 'ICMP RTT' it can be ping or fping, etc.
Yeah, something like that. For ping we already do this (default to fping
and fall back to regular ping), but those are closer in functionality.
But the idea is the same; the test should not specify "use netperf" but
rather "this is a UDP RR test". Then we pick irtt if it's available and
use netperf as a fallback. Optionally with a command line switch to
force fallback (probably a good idea at least until we get wider
deployment of irtt).
> I'm waving my arms here because there can be more than one measurement
> from a single tool. So maybe it should really be 'UDP RT' because in
> the case of irtt it's doing a UDP round-trip and measuring several
> things from it at once: RTT, OWD, IPDV, etc. Since there isn't a
> one-to-one correspondence between the tool and measurement it returns,
> it's not clear to me yet how it should be specified, still thinking
> about it.
> I guess at some point there will be additional plots for OWD and IPDV.
> Is it a new situation for you that the plots available depend not only
> on the test but the tool used or even its settings?
Meh, irtt's functionality is basically a superset of netperf's UDP_RR.
So automatically picking irtt if available and a fallback is the right
thing to do, I think. I'd just add the plots everywhere, and if OWD data
is not available, those plots would just be empty. Same thing we do for
TCP window stats currently. Longer term, maybe hiding the plots entirely
when there is no data is better, but for now empty plots are fine.
> Personally I don't think it's necessary to check for server-side
> support. If someone is specifying they want to use a particular tool,
> I think they're declaring that it's available, as it is today with
> netperf / netserver.
Yes, but if we do automatic detection with a preference we could get
into the situation where irtt exists on the client but the test is being
run against a server that doesn't have it. This is especially likely to
happen before the *.netperf.bufferbloat.net servers have irtt deployed.
Could you be persuaded to add a 'check_server' action to irtt? Something
that just does the handshake and doesn't run any more tests other than
that. Then we could have Flent call that to verify that irtt is
You are receiving this because you commented.
Reply to this email directly or view it on GitHub:
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Flent-users