[tor-relays] Consensus weight/Advertised bandwidth low on "Gigabit" ISP, despite ISP equipment upgrades

[William Kane]

It can take up to 6 months in my experience until a relay is fully
utilized, and some just never never reach peak bandwidth throughput
for whatever reason.

2020-06-13 5:51 GMT, Neel Chauhan <neel at neelc.org>:
> Hi tor-relays@,
>
> I run a FreeBSD-based Tor relay across two instances on "Wave G", a
> Gigabit ISP in the Seattle metro. You may also know them as
> CondoInternet or CascadeLink, but I joined only this year on a
> Wave-branded service.
>
> These relays have had low consensus weights since I got the service in
> January.
>
> The instances are below:
>
>   *
> https://metrics.torproject.org/rs.html#details/8FABF4D266DF95216F6C646C6D6D4611D3DCF484
>
>   *
> https://metrics.torproject.org/rs.html#details/CE06BA1EA45FD32A79EAF7FE6A3B1919E7FE585B
>
> My server and router are fine, I am in the single digits in terms of CPU
> use on both. The same exact server and router on Verizon FiOS in New
> York never gave me this issue.
>
> There was an underlying ISP performance issue impacting me which led
> consensus weight values to be low, but my ISP has since upgraded their
> equipment in my building. In general, my Internet performance has
> improved by magnitudes.
>
> However, my consensus weight has stayed more or less flat since the
> equipment upgrade, instead of jumping higher. What gives?
>
> How long would it usually take for the bandwidth scanners to measure the
> higher bandwidths?
>
> Should I re-key my relays and start from scratch?
>
> About switching ISPs, I'm not switching to Comcast for obvious
> well-documented reasons, and neither CenturyLink nor Frontier/Ziply
> Fiber serve me, not even copper.
>
> Best,
>
> Neel Chauhan
>
> ===
>
> https://www.neelc.org/
> _______________________________________________
> tor-relays mailing list
> tor-relays at lists.torproject.org
> https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays
>