Dear fellow Onion Service operators,
I'm curios how you monitor the availability of your onion services?
I implemented a solution with icinga2 check commands. It basically queries the onion service via torsocks and check_tcp command [1].
object CheckCommand "check_tor_onion" { command = [ "torsocks", PluginDir + "/check_tcp" ] timeout = 6m arguments = { "-H" = "$hostname$" "-p" = "$port$" "-t" = "$timeout$" } }
This solution is kind of flaky and produces a lot of mail noise, even if I run it with large timeout value (360s), check attempts and retry intervals.
Recently I came across the tool called hsprober [2], which looks like a more compelling option. though it requires a prometheus setup.
Since it is a network, where connections constantly looses connections, I would like to know how you treat this flakyness. Regardless of which software stack you use, I'm interested in your concept for monitoring your onion services from the outside (user side) and from the inside (server side). And also which tools do you use.
Thank you for any answers and opinions. Have a happy new year.
shadow
[1] - https://www.monitoring-plugins.org/doc/man/check_tcp.html [2] - https://git.autistici.org/ale/hsprober
‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ On Saturday, December 28, 2019 4:50 PM, shadow shadow@systemli.org wrote:
Dear fellow Onion Service operators,
I'm curios how you monitor the availability of your onion services?
As part of our work with Ablative Hosting we offer https://onionwatch.email
It's free to use and the sourcecode is on Github; https://github.com/BrassHornCommunications/OnionWatch
Lacking a decent keyboard at the moment, but I use this:
https://github.com/alecmuffett/real-world-onion-sites/blob/master/rwos-db.py
... it's a work in progress and I have yet to implement garbage collection for the database, but it's not too bad.
-a
Shell, netcat, poll descriptor and service. You don't need some grand architecture for simple shit.
I'm sure you could use an ELK stack with an HTTP Heartbeat monitor behind a Tor socks proxy.
On Dec 30, 2019, at 9:16 PM, grarpamp grarpamp@gmail.com wrote:
Shell, netcat, poll descriptor and service. You don't need some grand architecture for simple shit. _______________________________________________ tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
Shell, netcat, poll descriptor and service. You don't need some grand architecture for simple shit.
And then you can write a webserver in awk and attach it to inetd via openssl and set up a cronjob to scrape and publish the result to your monitoring dashboard via sort, uniq, col and more awk. And then do report generation in nroff. Simple.
-a
2E9C7EC9D2AC9A2FB6702BD247939A6F http://zimmermann.mayfirst.org/pks/lookup?op=hget&search=2E9C7EC9D2AC9A2FB6702BD247939A6F
On Wed., 1 Jan. 2020, 6:11 am grarpamp, grarpamp@gmail.com wrote:
And then... Simple.
Yes exactly, it's called oniontop, v2 scales to 100k on the console with sqlite, v3 puts pretty graphs in /tmp with rrdtool ;) _______________________________________________ tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
I think that is not an easy option or it does not apply to setups with a large amount of onion services.
Or at least not with a little bit of more explanation.
Happy new year,
shadow
On 31.12.19 03:15, grarpamp wrote:
Shell, netcat, poll descriptor and service. You don't need some grand architecture for simple shit. _______________________________________________ tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
Thank you for your answers,
I will checkout the different solutions and see what fits for my usecase.
Alecs piece of software was a good first starting point for and shows that it is a complexer problem.
I hope you all are doing well and have a happy new year,
shadow
On 28.12.19 17:50, shadow wrote:
Dear fellow Onion Service operators,
I'm curios how you monitor the availability of your onion services?
I implemented a solution with icinga2 check commands. It basically queries the onion service via torsocks and check_tcp command [1].
object CheckCommand "check_tor_onion" { command = [ "torsocks", PluginDir + "/check_tcp" ] timeout = 6m arguments = { "-H" = "$hostname$" "-p" = "$port$" "-t" = "$timeout$" } }
This solution is kind of flaky and produces a lot of mail noise, even if I run it with large timeout value (360s), check attempts and retry intervals.
Recently I came across the tool called hsprober [2], which looks like a more compelling option. though it requires a prometheus setup.
Since it is a network, where connections constantly looses connections, I would like to know how you treat this flakyness. Regardless of which software stack you use, I'm interested in your concept for monitoring your onion services from the outside (user side) and from the inside (server side). And also which tools do you use.
Thank you for any answers and opinions. Have a happy new year.
shadow
[1] - https://www.monitoring-plugins.org/doc/man/check_tcp.html [2] - https://git.autistici.org/ale/hsprober
tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
Hey all,
following up on this, we wrote a small prometheus-onion-exporter in go:
https://github.com/systemli/prometheus-onion-service-exporter
Maybe it helps you monitoring your onion sites.
cheers shadow
On 31.12.19 18:15, shadow wrote:
Thank you for your answers,
I will checkout the different solutions and see what fits for my usecase.
Alecs piece of software was a good first starting point for and shows that it is a complexer problem.
I hope you all are doing well and have a happy new year,
shadow
On 28.12.19 17:50, shadow wrote:
Dear fellow Onion Service operators,
I'm curios how you monitor the availability of your onion services?
I implemented a solution with icinga2 check commands. It basically queries the onion service via torsocks and check_tcp command [1].
object CheckCommand "check_tor_onion" { command = [ "torsocks", PluginDir + "/check_tcp" ] timeout = 6m arguments = { "-H" = "$hostname$" "-p" = "$port$" "-t" = "$timeout$" } }
This solution is kind of flaky and produces a lot of mail noise, even if I run it with large timeout value (360s), check attempts and retry intervals.
Recently I came across the tool called hsprober [2], which looks like a more compelling option. though it requires a prometheus setup.
Since it is a network, where connections constantly looses connections, I would like to know how you treat this flakyness. Regardless of which software stack you use, I'm interested in your concept for monitoring your onion services from the outside (user side) and from the inside (server side). And also which tools do you use.
Thank you for any answers and opinions. Have a happy new year.
shadow
[1] - https://www.monitoring-plugins.org/doc/man/check_tcp.html [2] - https://git.autistici.org/ale/hsprober
tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
I like the concept of a tor search engine but it is way to complicated for me, I am in my eighties, and I am not a geek, so not knowing what this onion stuff is I will say goodby Mike.
-----Original Message----- From: shadow Sent: Monday, August 16, 2021 12:05 AM To: tor-onions@lists.torproject.org Subject: Re: [tor-onions] Onion Service Monitoring
Hey all,
following up on this, we wrote a small prometheus-onion-exporter in go:
https://github.com/systemli/prometheus-onion-service-exporter
Maybe it helps you monitoring your onion sites.
cheers shadow
On 31.12.19 18:15, shadow wrote:
Thank you for your answers,
I will checkout the different solutions and see what fits for my usecase.
Alecs piece of software was a good first starting point for and shows that it is a complexer problem.
I hope you all are doing well and have a happy new year,
shadow
On 28.12.19 17:50, shadow wrote:
Dear fellow Onion Service operators,
I'm curios how you monitor the availability of your onion services?
I implemented a solution with icinga2 check commands. It basically queries the onion service via torsocks and check_tcp command [1].
object CheckCommand "check_tor_onion" { command = [ "torsocks", PluginDir + "/check_tcp" ] timeout = 6m arguments = { "-H" = "$hostname$" "-p" = "$port$" "-t" = "$timeout$" } }
This solution is kind of flaky and produces a lot of mail noise, even if I run it with large timeout value (360s), check attempts and retry intervals.
Recently I came across the tool called hsprober [2], which looks like a more compelling option. though it requires a prometheus setup.
Since it is a network, where connections constantly looses connections, I would like to know how you treat this flakyness. Regardless of which software stack you use, I'm interested in your concept for monitoring your onion services from the outside (user side) and from the inside (server side). And also which tools do you use.
Thank you for any answers and opinions. Have a happy new year.
shadow
[1] - https://www.monitoring-plugins.org/doc/man/check_tcp.html [2] - https://git.autistici.org/ale/hsprober
tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
tor-onions mailing list tor-onions@lists.torproject.org https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-onions
tor-onions@lists.torproject.org
-
Alec Muffett -
Dylan Issa -
Gareth Llewellyn -
grarpamp -
Jassy Knight -
madhatton＠shaw.ca -
shadow