(Dropped tor-relays)

Dear Network Health Team,

Now that you're working on sbws regularly, can you also answer relay operator questions about low bandwidths?

You know which bugs have been fixed in which sbws versions. And which versions are deployed on which authorities.

I don't really have enough context to answer well.

On 1 Apr 2020, at 08:51, teor teor@riseup.net wrote:

Signed PGP part

I'm cc'ing the network health team, so they are aware of this issue.

...

...

...

On 30 Mar 2020, at 23:00, ha3ks ha3ks@protonmail.com wrote: my relay dropped nearly all it’s bandwidth about a month ago, checking over it I can see 3 of the dir auths have given it a low consensus weighting, would that cause low bandwidth numbers? I checked my ufw in Ubuntu and it’s allowing 443 and 80.

Date: Tue, 31 Mar 2020 09:44:41 +1000 From: teor teor@riseup.net

I have seen similar reports from a few other relay operators.

What is your relay fingerprint?

Have you followed the troubleshooting instructions here: https://trac.torproject.org/projects/tor/wiki/doc/MyRelayIsSlow

On 31 Mar 2020, at 20:46, ha3ks ha3ks@protonmail.com wrote:

The fingerprint is - 6E1DA4C0B0C05FB721B42329C47A20DA22908AEB

I have followed some of this yes, though its a little over my head

I have opened a ticket in sbws to follow up this issue:

sbws measures some relays 100x lower than Torflow https://trac.torproject.org/projects/tor/ticket/33775

It could be related to these other issues:

sbws does not detect changes in descriptor bandwidth values https://trac.torproject.org/projects/tor/ticket/30733

sbws bandwidth scans should require a minimum exit bandwidth https://trac.torproject.org/projects/tor/ticket/33009