Hi, I've added new operator level pages with an interactive graph showing the aggregated guard/exit probability and advertised bandwidth over time across all relays for a given operator. This should allow you to see whether your bandwidth and fractions are growing or decreasing. You can - zoom within the graph to a specific time window (horizontal drag and drop), - change the scale on the y axis (vertical drag and drop) - zoom to a specific area (by selecting a rectangle) or use the preset time frames (week, month, year, ..). To give you a better idea of what these graphs look like, here is an example: https://nusenu.github.io/OrNetStats/www.f3netze.de.html The data source for these graphs is onionoo (torproject.org), it starts at 2019-10-31. The data granularity is about one value per day. kind regards, nusenu -- https://nusenu.github.io
Thank you!!!
On Jan 9, 2021, at 14:39, nusenu <nusenu-lists@riseup.net> wrote:
Hi,
I've added new operator level pages with an interactive graph showing the aggregated guard/exit probability and advertised bandwidth over time across all relays for a given operator.
This should allow you to see whether your bandwidth and fractions are growing or decreasing.
You can - zoom within the graph to a specific time window (horizontal drag and drop), - change the scale on the y axis (vertical drag and drop) - zoom to a specific area (by selecting a rectangle) or use the preset time frames (week, month, year, ..).
To give you a better idea of what these graphs look like, here is an example:
https://nusenu.github.io/OrNetStats/www.f3netze.de.html
The data source for these graphs is onionoo (torproject.org), it starts at 2019-10-31. The data granularity is about one value per day.
kind regards, nusenu
_______________________________________________ tor-relays mailing list tor-relays@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays
On 1/9/21 9:38 PM, nusenu wrote:
I've added new operator level pages with an interactive graph showing the aggregated guard/exit probability and advertised bandwidth over time across all relays for a given operator. cool idea,
canthose graphs being linked from eg https://yui.cat/relay/63BF46A63F9C21FD315CD061B3EAA3EB05283A0A.html ? -- Toralf
Toralf Förster:
On 1/9/21 9:38 PM, nusenu wrote:
I've added new operator level pages with an interactive graph showing the aggregated guard/exit probability and advertised bandwidth over time across all relays for a given operator. cool idea,
canthose graphs being linked from eg https://yui.cat/relay/63BF46A63F9C21FD315CD061B3EAA3EB05283A0A.html ?
Yes, that should be easy. The best way to get that done is to either submit a patch or feature request here: https://github.com/tempname1024/allium/issues kind regards, nusenu -- https://nusenu.github.io
On 1/9/21 2:42 PM, nusenu wrote:
Toralf Förster:
cool idea,
canthose graphs being linked from eg https://yui.cat/relay/63BF46A63F9C21FD315CD061B3EAA3EB05283A0A.html ?
Yes, that should be easy. The best way to get that done is to either submit a patch or feature request here: https://github.com/tempname1024/allium/issues
kind regards, nusenu
Sweet graphs, thanks Nusenu! I opened an issue[1] to address support for the ContactInfo-Information-Sharing-Specification which includes outbound links to graphs of verified operators/domains at OrNetStats, and rendering of provided attributes on per-relay pages. [1] https://github.com/tempname1024/allium/issues/9 -- Jordan https://yui.cat/
On 09.01.2021 21:38, nusenu wrote:
I've added new operator level pages with an interactive graph showing the aggregated guard/exit probability and advertised bandwidth over time across all relays for a given operator.
This should allow you to see whether your bandwidth and fractions are growing or decreasing. Nice, thanks.
I entered the DNS TXT records for my relays. 2 questions: - Shall we also add the fingerprints from the bridges in 'proof'? - Can I leave my email address as it is for "foo bar", see below? I don't want to expect the average user to fish something obfuscated out of the string. Yes, I am old and conservative ;-) https://nusenu.github.io/ContactInfo-Information-Sharing-Specification/ An example ContactInfo string as defined by this specification could look like this: foo bar email:tor[]example.com url:https://example.com proof:uri-rsa ciissversion:2 ^^^^^^^ abuse@for-privacy.net email:admin[]for-privacy.net url:for-privacy.net proof:dns-rsa keybase:boldsuck ciissversion:2 ^^^^^^^ General question: Is it possible to introduce a new option in the configuration file for the ContactInfo string? (ContactInfo + ContactInfoString)? Maybe when the IPv6 development is done. Because of contact info. There are a couple of relays that have given these lists addresses: https://metrics.torproject.org/rs.html#search/contact:tor-relays@lists.torpr... -- ╰_╯ Ciao Marco! Debian GNU/Linux It's free software and it gives you freedom!
On 1/10/21 7:08 PM, lists@for-privacy.net wrote:
I don't want to expect the average user to fish something obfuscated out of the string. Yes, I am old and conservative ;-)
/me too - therefore I tend to not add detailed technical information about the relays. (BTW that doesn't work at all with the current ciissversion if relays are hosted at different hoster at all, or?) -- Toralf
On 10.01.2021 19:19, Toralf Förster wrote:
(BTW that doesn't work at all with the current ciissversion if relays are hosted at different hoster at all, or?)
I wanted to set up a httpd first. Luckily I left that. I wrote the fingerprints in the DNS. Would like to have as little software or services as possible on the relays. -- ╰_╯ Ciao Marco! Debian GNU/Linux It's free software and it gives you freedom!
Toralf Förster:
(BTW that doesn't work at all with the current ciissversion if relays are hosted at different hoster at all, or?)
I didn't understand you there, would you like to elaborate? Using multiple hosters is good and recommended and CIISS does not limit that, it supports as many hosters as you like. kind regards, nusenu -- https://nusenu.github.io
Hi Am 10.01.2021 um 19:08 schrieb lists@for-privacy.net:
On 09.01.2021 21:38, nusenu wrote: - Shall we also add the fingerprints from the bridges in 'proof'?
https://gitlab.torproject.org/tpo/core/torspec/-/blob/master/proposals/326-t... /.well-known/tor-relay/rsa-fingerprint.txt - The file contains one or more Tor relay RSA SHA1 fingerprints operated by the entity in control of this website. - Each line contains one fingerprint. - The file may contain comments (starting with #). - Non-comment lines must be exactly 40 characters long and consist of the following characters [a-fA-F0-9]. - Fingerprints are not case-sensitive. - Each fingerprint MUST appear at most once. - The file MUST not be larger than one MByte. - The file MUST NOT contain fingerprints of Tor bridges (or hashes of bridge fingerprints). - The content MUST be a media type of "text/plain". :: No bridge fingerprints :: -- Cheers, Felix
On 10.01.2021 21:37, Felix wrote:
Am 10.01.2021 um 19:08 schrieb lists@for-privacy.net:
- Shall we also add the fingerprints from the bridges in 'proof'?
https://gitlab.torproject.org/tpo/core/torspec/-/blob/master/proposals/326-t...
:: No bridge fingerprints ::
OK thanks. Possibly @nusenu can mention that here: https://nusenu.github.io/ContactInfo-Information-Sharing-Specification/ at 'dns-rsa' -- ╰_╯ Ciao Marco! Debian GNU/Linux It's free software and it gives you freedom!
- Can I leave my email address as it is [...]?
The use of "[]" is optional ("SHOULD"), so this is valid: email:contact@example.com Also the example you mentioned is valid.
I wanted to set up a httpd first. Luckily I left that. I wrote the fingerprints in the DNS. Would like to have as little software or services as possible on the relays.
just to clarify: there is no requirement to run additional software on the relay. the url can point to an arbitrary domain that you manage. kind regards, nusenu
On 11.01.2021 18:46, nusenu wrote:
- Can I leave my email address as it is [...]?
The use of "[]" is optional ("SHOULD"), so this is valid:
email:contact@example.com
Also the example you mentioned is valid.
OK thanks.
I wanted to set up a httpd first. Luckily I left that. I wrote the fingerprints in the DNS. Would like to have as little software or services as possible on the relays.
just to clarify: there is no requirement to run additional software on the relay. the url can point to an arbitrary domain that you manage.
OK. Unfortunately I can only redirect in private webspace. And that is "MUST NOT". My DNS-TXT records are slowly spreading across the DNS world. In a few days my 'proof' should show up on OrNetStats. -- ╰_╯ Ciao Marco! Debian GNU/Linux It's free software and it gives you freedom!
OK. Unfortunately I can only redirect in private webspace. And that is "MUST NOT". My DNS-TXT records are slowly spreading across the DNS world. In a few days my 'proof' should show up on OrNetStats.
The DNS record check will be added in a few days, at that point your relays and a few more other groups will show up on the page if it finds your DNSSEC signed records.
Minor complaint apart from the good job you've done already: The overlap on the top right side, as pointed out in the attachment, bugs the hell out of me. Otherwise, a solid idea and improvement. William 2021-01-13 21:40 GMT, nusenu <nusenu-lists@riseup.net>:
OK. Unfortunately I can only redirect in private webspace. And that is "MUST NOT". My DNS-TXT records are slowly spreading across the DNS world. In a few days my 'proof' should show up on OrNetStats.
The DNS record check will be added in a few days, at that point your relays and a few more other groups will show up on the page if it finds your DNSSEC signed records. _______________________________________________ tor-relays mailing list tor-relays@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-relays
William Kane:
Minor complaint apart from the good job you've done already: The overlap on the top right side, as pointed out in the attachment, bugs the hell out of me.
Otherwise, a solid idea and improvement.
thank you for your feedback, it should be fixed now.
2021-01-13 21:40 GMT, nusenu <nusenu-lists@riseup.net>:
OK. Unfortunately I can only redirect in private webspace. And that is "MUST NOT". My DNS-TXT records are slowly spreading across the DNS world. In a few days my 'proof' should show up on OrNetStats.
The DNS record check will be added in a few days, at that point your relays and a few more other groups will show up on the page if it finds your DNSSEC signed records.
The DNS check has been added and your (for-privacy.net) and others did verify properly are now also on the list. Interesting graph you produced there :) next up: using RIPEstat as data source for IP to autonomous system mapping. -- https://nusenu.github.io
On 17.01.2021 00:57, nusenu wrote:
The DNS check has been added and your (for-privacy.net) and others did verify properly are now also on the list. Interesting graph you produced there :)
At myLoc (Servdiscount) I have 2 Tor-instances on each server blade and the provider limits the bandwidth¹ from 1,000 Mbit/s to 200 Mbit/s when I use too much traffic. :-( ¹(for the rest of the month) I would like to rent a 1/8 rack at some point: https://www.myloc.de/en/colocation/colocation-offer.html The exits run unlimited. Thanks to Francisco.
next up: using RIPEstat as data source for IP to autonomous system mapping.
There's a lot of work going into OrNetStats and OrNetRadar. Thanks a lot for this. Small note: The "Missing AS name" at OrNetRadar with IP's from Frantec/BuyVM. People have not yet given Francisco any or incorrect personal details. -- ╰_╯ Ciao Marco! Debian GNU/Linux It's free software and it gives you freedom!
Maybe nitpicking but IMO It is irritating, that the 0% value (zero) of the left y-axis doesn't match the 0 GBit/s at the right y-axis (per default).
I find it actually useful to have distinct scaling and offsets for the two y axis, because this way the available space is used more efficiently, no? After learning that these are not starting at 0, things should be clear. What doesn't make sense for me: negative ranges, which I'll try to address soon. thanks for your input, nusenu
participants (7)
-
Felix -
John Ricketts -
Jordan -
lists＠for-privacy.net -
nusenu -
Toralf Förster -
William Kane