Closed dtaht closed 6 years ago
Dave Täht notifications@github.com writes:
while trying to look hard at the per host isolation feature in cake, I did things like start a string of rrul tests slightly staggered in time, on different machines (synced via ntp). when you gather these sort of plots and combine them, you can't see "the action" happening, as flent does not respect the timestamps. So being able to combine plots that have a startpoint at one dataset and an end in another would be useful.
Did you try the --absolute-time parameter to Flent when plotting?
yes.
two plots from different machines, tests started 20 sec apart. --absolute-time doesn't seem to work.
On Thu, Jul 19, 2018 at 11:42 AM Dave Täht notifications@github.com wrote:
yes.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/tohojo/flent/issues/144#issuecomment-406375455, or mute the thread https://github.com/notifications/unsubscribe-auth/AerUv8ZxmNMqQ39Miixf0latT_Wm6xvcks5uINLngaJpZM4VTl5x .
Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org
--
Dave Täht CEO, TekLibre, LLC http://www.teklibre.com Tel: 1-669-226-2619
flent-users notifications@github.com writes:
two plots from different machines, tests started 20 sec apart. --absolute-time doesn't seem to work.
If that was supposed to contain attachments, they didn't make it across...
-Toke
Hmm, turns out there was a bug in the absolute_time handling :)
i pulled. that fix breaks those plots completely in the general case.
(I have no idea how you are holding cake, iproute2, and flent all in your head)
Dave Täht notifications@github.com writes:
i pulled. that fix breaks those plots completely in the general case.
Whaddyamean breaks plots completely? Looks fine on my machine...
-Toke
d@dancer:~/ne m$ flent-gui --absolute-time 500.gz Started Flent 1.2.2-git-09e66b3 using Python 3.5.2. WARNING: Falling back to Qt4 for the GUI. Please consider installing PyQt5.
Initialised matplotlib v1.5.1 on numpy v1.11.0. GUI loaded. Running on PyQt v4.11.4.
and ok, I got around to installing pyqt5 and it does the same thing. Older matplotlib?
Nope; what you're seeing is actually an improvement; it is now partially applying the absolute-time parameter, instead of not applying it at all ;)
the weird thing I'm actually trying to look at is that it takes 10sec for the impact of the 10 new flows to truly hit, then all the other flows come back hard, then 10 sec later, we achieve balance. I've been lying down on the job by not thoroughly auditing the last 6 months worth of AQM changes...
thx! the network behavior I'm looking at itself doesn't make any sense, but at least I can plot it now.
Yay, progress! ;)
while trying to look hard at the per host isolation feature in cake, I did things like start a string of rrul tests slightly staggered in time, on different machines (synced via ntp). when you gather these sort of plots and combine them, you can't see "the action" happening, as flent does not respect the timestamps. So being able to combine plots that have a startpoint at one dataset and an end in another would be useful.
/me ducks