[-- Attachment #1.1: Type: text/plain, Size: 6099 bytes --] Using Flent 1.3.0 on macOS 10.12.6: I notice that the "TCP upload sum" value shown in the upload portion of tests (tcp_nup, rrul, rrul_be) doesn't match the sum of the individual flows. In the first example below, the displayed sum (0.28 Mbps) is considerably lower than the sum of the individual flows (0.77 Mbps - which matches the expected speed of my 7Mbps/768kbps DSL link). I attach the flent.gz and the log files from the first run shown below. (Earlier rrul and rrul_be test runs follow.) What other information could I gather? Thanks. [tcp_nup-2019-09-12T074946.261029.flent-fremont_bufferbloat_net.log](https://github.com/tohojo/flent/files/3605312/tcp_nup-2019-09-12T074946.261029.flent-fremont_bufferbloat_net.log) [tcp_nup-2019-09-12T074946.261029.flent-fremont_bufferbloat_net.flent.gz](https://github.com/tohojo/flent/files/3605313/tcp_nup-2019-09-12T074946.261029.flent-fremont_bufferbloat_net.flent.gz) ``` bash-3.2$ flent -x -L -H flent-fremont.bufferbloat.net -t "flent-fremont.bufferbloat.net" --te=upload_streams=4 --te=ping_hosts=1.1.1.1 tcp_nup Started Flent 1.3.0 using Python 2.7.15. Starting tcp_nup test. Expected run time: 70 seconds. Data file written to ./tcp_nup-2019-09-12T074946.261029.flent-fremont_bufferbloat_net.flent.gz. Summary of tcp_nup test run from 2019-09-12 11:49:46.261029 Title: 'flent-fremont.bufferbloat.net' avg median # data pts Ping (ms) ICMP : 100.34 100.50 ms 348 Ping (ms) ICMP 1.1.1.1 (extra) : 23.40 23.40 ms 349 TCP upload avg : 0.18 0.19 Mbits/s 108 TCP upload sum : 0.28 0.28 Mbits/s 108 TCP upload::1 : 0.22 0.20 Mbits/s 11 TCP upload::2 : 0.16 0.14 Mbits/s 7 TCP upload::3 : 0.17 0.15 Mbits/s 8 TCP upload::4 : 0.22 0.20 Mbits/s 11 ``` ``` Started Flent 1.3.0 using Python 2.7.15. Starting rrul test. Expected run time: 70 seconds. Data file written to ./rrul-2019-09-11T125110.964940.Running_flent_from_Richs-MBP-10337_lan.flent.gz. Summary of rrul test run from 2019-09-11 16:51:10.964940 Title: 'Running_flent_from_Richs-MBP-10337.lan' avg median # data pts Ping (ms) ICMP : 104.75 104.00 ms 348 Ping (ms) ICMP 1.1.1.1 (extra) : 28.96 28.80 ms 346 Ping (ms) UDP BE : 104.78 104.38 ms 261 Ping (ms) UDP BK : 105.18 104.71 ms 273 Ping (ms) UDP EF : 104.18 103.31 ms 247 Ping (ms) avg : 88.23 89.39 ms 353 TCP download BE : 1.53 1.59 Mbits/s 282 TCP download BK : 1.56 1.60 Mbits/s 284 TCP download CS5 : 1.46 1.51 Mbits/s 286 TCP download EF : 1.51 1.58 Mbits/s 282 TCP download avg : 1.51 1.55 Mbits/s 301 TCP download sum : 6.04 6.20 Mbits/s 301 TCP totals : 6.06 6.21 Mbits/s 301 TCP upload BE : 0.09 0.08 Mbits/s 4 TCP upload BK : 0.09 0.07 Mbits/s 3 TCP upload CS5 : 0.11 0.09 Mbits/s 5 TCP upload EF : 0.11 0.09 Mbits/s 5 TCP upload avg : 0.09 0.08 Mbits/s 48 TCP upload sum : 0.14 0.12 Mbits/s 48 ``` ``` Started Flent 1.3.0 using Python 2.7.15. Starting rrul_be test. Expected run time: 70 seconds. Data file written to ./rrul_be-2019-09-11T125231.578848.Running_flent_from_Richs-MBP-10337_lan.flent.gz. Summary of rrul_be test run from 2019-09-11 16:52:31.578848 Title: 'Running_flent_from_Richs-MBP-10337.lan' avg median # data pts Ping (ms) ICMP : 104.88 104.00 ms 350 Ping (ms) ICMP 1.1.1.1 (extra) : 30.32 29.90 ms 349 Ping (ms) UDP BE1 : 105.10 104.38 ms 248 Ping (ms) UDP BE2 : 106.64 106.27 ms 277 Ping (ms) UDP BE3 : 105.71 104.71 ms 279 Ping (ms) avg : 89.55 90.38 ms 353 TCP download BE : 1.45 1.49 Mbits/s 278 TCP download BE2 : 1.52 1.57 Mbits/s 280 TCP download BE3 : 1.66 1.71 Mbits/s 281 TCP download BE4 : 1.36 1.42 Mbits/s 278 TCP download avg : 1.49 1.53 Mbits/s 301 TCP download sum : 5.97 6.13 Mbits/s 301 TCP totals : 6.00 6.15 Mbits/s 301 TCP upload BE : 0.08 0.06 Mbits/s 3 TCP upload BE2 : 0.11 0.09 Mbits/s 5 TCP upload BE3 : 0.11 0.09 Mbits/s 5 TCP upload BE4 : 0.11 0.09 Mbits/s 5 TCP upload avg : 0.09 0.09 Mbits/s 34 TCP upload sum : 0.19 0.23 Mbits/s 34 bash-3.2$ ``` -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/tohojo/flent/issues/182 [-- Attachment #1.2: Type: text/html, Size: 7466 bytes --] [-- Attachment #2: Type: text/plain, Size: 151 bytes --] _______________________________________________ Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org
[-- Attachment #1.1: Type: text/plain, Size: 392 bytes --] Update: dtaht (in https://github.com/tohojo/flent/issues/178#issuecomment-528885331) said of the ping times, "I have seen this statistic be wildly wrong in the summary but correct in the graph." Related? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/tohojo/flent/issues/182#issuecomment-530795911 [-- Attachment #1.2: Type: text/html, Size: 2125 bytes --] [-- Attachment #2: Type: text/plain, Size: 151 bytes --] _______________________________________________ Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org
[-- Attachment #1.1: Type: text/plain, Size: 828 bytes --] Rich Brown <notifications@github.com> writes: > Update: dtaht (in > https://github.com/tohojo/flent/issues/178#issuecomment-528885331) > said of the ping times, "I have seen this statistic be wildly wrong in > the summary but correct in the graph." Yes, this can be the case because of the way the summary values are computed. However, it *should* be the case that what the summary outputs is computed from the aggregate statistics that netperf outputs at the end of the test where possible. However, it seems there was a bug causing all this per-series metadata to be cleared out before the file was saved. Not good! Will post a fix... -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/tohojo/flent/issues/182#issuecomment-531299423 [-- Attachment #1.2: Type: text/html, Size: 2269 bytes --] [-- Attachment #2: Type: text/plain, Size: 151 bytes --] _______________________________________________ Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org
[-- Attachment #1.1: Type: text/plain, Size: 239 bytes --] Closed #182 via ff465203dd7de78d87e0d88717068621b8159c28. -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/tohojo/flent/issues/182#event-2632860804 [-- Attachment #1.2: Type: text/html, Size: 2224 bytes --] [-- Attachment #2: Type: text/plain, Size: 151 bytes --] _______________________________________________ Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org
[-- Attachment #1.1: Type: text/plain, Size: 2803 bytes --] Sorry to re-open, but the new version (Flent 1.9.9-git-ff46520) seems to fix the upload values for the Average - the individual streams sum to the displayed sum. But the Median values still don't look right. Here are two runs of the same command: ``` bash-3.2$ flent -x -L -H flent-fremont.bufferbloat.net -t "flent-fremont.bufferbloat.net" --te=upload_streams=4 --te=ping_hosts=1.1.1.1 tcp_nup Started Flent 1.9.9-git-ff46520 using Python 3.7.4. Starting tcp_nup test. Expected run time: 70 seconds. Data file written to ./tcp_nup-2019-09-13T210706.195971.flent-fremont_bufferbloat_net.flent.gz. Summary of tcp_nup test run from 2019-09-14 01:07:06.195971 Title: 'flent-fremont.bufferbloat.net' avg median # data pts Ping (ms) ICMP : 101.23 101.00 ms 347 Ping (ms) ICMP 1.1.1.1 (extra) : 23.39 23.90 ms 347 TCP upload avg : 0.17 0.18 Mbits/s 354 TCP upload sum : 0.69 0.30 Mbits/s 354 TCP upload::1 : 0.16 0.17 Mbits/s 354 TCP upload::2 : 0.15 0.14 Mbits/s 354 TCP upload::3 : 0.19 0.20 Mbits/s 354 TCP upload::4 : 0.19 0.20 Mbits/s 354 ``` ``` bash-3.2$ flent -x -L -H flent-fremont.bufferbloat.net -t "flent-fremont.bufferbloat.net" --te=upload_streams=4 --te=ping_hosts=1.1.1.1 tcp_nup Started Flent 1.9.9-git-ff46520 using Python 3.7.4. Starting tcp_nup test. Expected run time: 70 seconds. Data file written to ./tcp_nup-2019-09-13T211010.793062.flent-fremont_bufferbloat_net.flent.gz. Summary of tcp_nup test run from 2019-09-14 01:10:10.793062 Title: 'flent-fremont.bufferbloat.net' avg median # data pts Ping (ms) ICMP : 102.19 102.00 ms 347 Ping (ms) ICMP 1.1.1.1 (extra) : 24.91 24.80 ms 341 TCP upload avg : 0.15 0.16 Mbits/s 354 TCP upload sum : 0.62 0.16 Mbits/s 354 TCP upload::1 : 0.12 0.13 Mbits/s 354 TCP upload::2 : 0.18 0.19 Mbits/s 354 TCP upload::3 : 0.18 0.19 Mbits/s 354 TCP upload::4 : 0.14 0.15 Mbits/s 354 bash-3.2$ ``` -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/tohojo/flent/issues/182#issuecomment-531433159 [-- Attachment #1.2: Type: text/html, Size: 4205 bytes --] [-- Attachment #2: Type: text/plain, Size: 151 bytes --] _______________________________________________ Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org
[-- Attachment #1.1: Type: text/plain, Size: 912 bytes --] Rich Brown <notifications@github.com> writes: > Sorry to re-open, but the new version (Flent 1.9.9-git-ff46520) seems > to fix the upload values for the Average - the individual streams sum > to the displayed sum. > > But the Median values still don't look right. Here are two runs of the > same command: Ah, no, the median has the same issue, but there is no per-test output we can use, so we need to calculate it. And since the summary output calculates a per-series median, the sum doesn't match up (which is not too surprising; the median of the total throughput is not necessarily the sum of the medians of each individual flow). We could remove the median from the calculated data series if that is less confusing? -- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/tohojo/flent/issues/182#issuecomment-531489398 [-- Attachment #1.2: Type: text/html, Size: 2369 bytes --] [-- Attachment #2: Type: text/plain, Size: 151 bytes --] _______________________________________________ Flent-users mailing list Flent-users@flent.org http://flent.org/mailman/listinfo/flent-users_flent.org