Hi,
On 25 Jun 2019, at 09:32, Gunnar Wolf gwolf@gwolf.org wrote:
Signed PGP part teor dijo [Sun, Jun 23, 2019 at 02:54:51PM +1000]:
https://metrics.torproject.org/userstats-relay-country.html?start=2017-06-22...
It seems to be a significant increase in users all over Iran:
https://trac.torproject.org/projects/tor/ticket/30636#comment:8
https://metrics.torproject.org/userstats-relay-country.html?start=2017-06-22...
If all those users are detected to be connecting directly from Iran, and following the text written in that very page you mention:
This graph shows the estimated number of directly-connecting clients; that is, it excludes clients connecting via bridges.
Does this mean that Tor is no longer blocked from within Iran (and users no longer need to connect via bridges)? Any other explanations?
Anyway, I still find the sharp uptake in direct connections quite odd. People connecting via obfs4 (which continues to work just fine) wouldn't migrate to direct connections massively as this shows...
It looks like many clients in Iran can't download the full consensus, so they keep on trying.
And Tor is measuring directory requests, even if the client fails to download the response.
Follow the ticket for more details.
T