[tor-bugs] #28647 [Core Tor/sbws]: Update INSTALL.rst and DEPLOY.rst based on Torflow's documentation

Tor Bug Tracker & Wiki blackhole at torproject.org
Fri Nov 30 11:44:57 UTC 2018


#28647: Update INSTALL.rst and DEPLOY.rst based on Torflow's documentation
---------------------------+-----------------------------------
 Reporter:  teor           |          Owner:  (none)
     Type:  defect         |         Status:  new
 Priority:  Medium         |      Milestone:  sbws: 1.0.x-final
Component:  Core Tor/sbws  |        Version:
 Severity:  Normal         |     Resolution:
 Keywords:                 |  Actual Points:
Parent ID:                 |         Points:
 Reviewer:                 |        Sponsor:
---------------------------+-----------------------------------

Comment (by juga):

 Replying to [comment:3 teor]:
 > Replying to [comment:2 juga]:
 > > Replying to [ticket:28647 teor]:
 > > > Are these things required for sbws:
 > > > * ~~500Mbit-1Gbit of upstream~~ documented in INSTALL.rst
 > >
 > > It currently says `at least 20MB/s (160 Mbit/s).`
 >
 > Why did you pick this number?

 That's what the testing sbws server has and seemed to work well.

 > 100 Mbit/s is the most common available bandwidth.
 >
 > > I think this should be changed to `between 500Mbit/s or 1Gbit/s`. The
 scanner used by longclaw is 1Gbit/s
 >
 > Why is 1 Gbit/s required?
 > Torflow and sbws scaling is relative, so high bandwidths give better
 accuracy, but they're not required.

 At some point we might relay more on the measurements than the relay's
 observed bandwidth?
 >
 > Before we set a requirement, we should make sure it is actually
 available to bandwidth authority operators.

 What do you think is the minimum required?
 I'll ask then to the bwauth operators what's available to them.
 >
 > > > * a fixed IP address
 >
 > Is this needed?

 actually, i don't think so

 > > > * SSL is needed to avoid HTTP content caches at the various exit
 nodes
 > >
 > > When the configuration file it's parsed, accepts both http and https.
 Should this be changed to allow only https and give an error when it's
 http?
 >
 > I think we should require HTTPS, because it stops caches and other forms
 of cheating.

 ok

 > > > * Self-signed certs are OK
 > > > * The server will consume around 12-15Gbytes/day
 > > > * A script to create the file
 > >
 > > Should the script be included in sbws or just written in the
 documentation?.
 >
 > I think it's enough to include it in the documentation.
 > It's a very short script.

 ok

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/28647#comment:4>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online


More information about the tor-bugs mailing list