Gargoyle on Ubiquiti Nano HD?

[pythonic]

It would be interesting to test the speed between openwrt and the phone but I don't know if there's an iperf3 app that I can use.

OpenWrt has iperf3 in it's package collection; you should be able to install it with opkg.

It isn't built for Gargoyle releases but as it only depends on libc it is likely that you can manually install it on v1.14 by downloading the package (iperf3_3.11-1_mipsel_24kc.ipk) from OpenWrt's 22.03 mt7621 packages collection to the router and using gpkg to install the .ipk.

[emmexx]

OpenWrt has iperf3 in it's package collection; you should be able to install it with opkg.

I know. I wrote iperf3 app meaning a phone app, something that acts as an iperf3 server.

[pythonic]

OpenWrt has iperf3 in it's package collection; you should be able to install it with opkg.

I know. I wrote iperf3 app meaning a phone app, something that acts as an iperf3 server.

Oops, sorry, my bad.

However, as I recall running the iperf3 server on OpenWrt (or on your laptop connected to OpenWrt if you wanted to test the chain) with an iperf3 client on the phone you should still be able to get a reasonable guide to performance in both directions as the server can be both data source and/or sink - although you would need a port-forward in place on the router to deal with NAT. In principle at least this should be possible - I don't know whether the phone environment would prevent this working though...

[emmexx]

However, as I recall running the iperf3 server on OpenWrt (or on your laptop connected to OpenWrt if you wanted to test the chain) with an iperf3 client on the phone you should still be able to get a reasonable guide to performance in both directions as the server can be both data source and/or sink - although you would need a port-forward in place on the router to deal with NAT. In principle at least this should be possible - I don't know whether the phone environment would prevent this working though...

I installed this android version of iperf3:
https://github.com/davidBar-On/android- ... me-ov-file

I used my laptop as a server and the phone as the client (I didn't want to mess around with my phone ports and firewall).
In Gargoyle I added a port forward to my laptop for 5201.

Command on my laptop:

Code: Select all

iperf3 -s

Command on my laptop connected to my phone using an usb cable and adb:

Code: Select all

adb shell /data/local/tmp/iperf3.16 -c gargoyle-wan-ip

Results:

Code: Select all

[ ID] Interval           Transfer     Bitrate         Retr  Cwnd
[  5]   0.00-1.00   sec  9.75 MBytes  81.7 Mbits/sec    4    875 KBytes       
[  5]   1.00-2.00   sec  4.38 MBytes  36.7 Mbits/sec    2   1.41 KBytes       
[  5]   2.00-3.00   sec  0.00 Bytes  0.00 bits/sec    1   1.41 KBytes       
[  5]   3.00-4.00   sec   896 KBytes  7.34 Mbits/sec  301    700 KBytes       
[  5]   4.00-5.00   sec  10.9 MBytes  91.3 Mbits/sec    2    621 KBytes       
[  5]   5.00-6.00   sec  11.8 MBytes  98.6 Mbits/sec    0    626 KBytes       
[  5]   6.00-7.00   sec  11.6 MBytes  97.5 Mbits/sec    0    632 KBytes       
[  5]   7.00-8.00   sec  11.6 MBytes  97.5 Mbits/sec    0    708 KBytes       
[  5]   8.00-9.00   sec  11.8 MBytes  98.6 Mbits/sec    0    708 KBytes       
[  5]   9.00-10.00  sec  11.4 MBytes  95.4 Mbits/sec    0    708 KBytes       
- - - - - - - - - - - - - - - - - - - - - - - - -
[ ID] Interval           Transfer     Bitrate         Retr
[  5]   0.00-10.00  sec  84.2 MBytes  70.7 Mbits/sec  310             sender
[  5]   0.00-10.03  sec  84.1 MBytes  70.4 Mbits/sec                  receiver

[pythonic]

That is quite interesting! Except for the interval 1s-4s, in which the throughput collapsed to zero and then rebounded, your phone tethered Gargoyle router was delivering a fairly consistent 90+Mbps to your laptop, which is a bit better than I expected. I don't know enough about Wifi to theorise why it stalled early on though

I think this is enough evidence to suggest that the major performance bottleneck is more likely than not upstream of Gargoyle, and the fact that TTL adjustments seem to help - even if only modestly - also seems to me to point in that direction.

Other than looking for alternative phone plans/SIMs available to you that have less restrictions on how data is used, I can't think of anything else to pursue.

BTW: if your iperf3 client supports all the standard iperf3 options, the -R (or --reverse) option would allow you to test throughput in the upload direction which might also be worth knowing.