On Thu, Jun 5, 2014 at 3:50 PM, Iain R. Learmonth irl@fsfe.org wrote:
I'm wondering if anyone knows anything about JANET's (the UK academic ISP) views on running a Tor relay?
It's probably not worth putting together a proposal for running one in a UK University if the ISP is going to have problems with it.
Has anyone tried approaching these networks themselves to see about running relays there? Their bandwidth for sponsored things is often free. In the US you might try internet2.edu and all its various connecting regional networks.
On a second further approach, once all the risks and pro-reasons have been discussed and accepted, the consortia backing of many Uni's within would be bulletproof.
On Jun 7, 2014 3:27 PM, "grarpamp" grarpamp@gmail.com wrote:
Has anyone tried approaching these networks themselves to see about running relays there? Their bandwidth for sponsored things is often free. In the US you might try internet2.edu and all its various connecting regional networks.
I'm at a member institution for Internet2, and the buy-in process put us in a research VLAN "outside" the university network. I'd be very interested in hearing from people at other member institutions about coordinating management of risk such that our service is more supportable and robust.
C, there is also a tor-relays-universities list. Forwarding there to keep the initial chat primed.
Once you have buy in from legal, chairs, security, upstream, etc this can be a very strong position, often better than pay 'contract' of random ISP host. I have seen such 'outside' nets used for these not strictly mission things, such I suggest it in this thread. Different approach depends on if you can find and house a legitimate paper producing research purpose, or if you simply will run it for supporting freedom point of view.
Worth mention is that both internet2 and nanog have mailing lists where queries and propositions could be sent. Cold contacts at regionals are not hard to find.
---------- Forwarded message ---------- From: Cristóbal Palmer cmpalmer@ibiblio.org Date: Mon, Jun 9, 2014 at 5:56 PM Subject: Re: [tor-relays] Running relays at consortia networks [was: JANET/edu] To: tor-relays@lists.torproject.org
On Jun 7, 2014 3:27 PM, "grarpamp" grarpamp@gmail.com wrote:
Has anyone tried approaching these networks themselves to see about running relays there? Their bandwidth for sponsored things is often free. In the US you might try internet2.edu and all its various connecting regional networks.
I'm at a member institution for Internet2, and the buy-in process put us in a research VLAN "outside" the university network. I'd be very interested in hearing from people at other member institutions about coordinating management of risk such that our service is more supportable and robust.
On Mon, Jun 9, 2014 at 8:09 PM, grarpamp grarpamp@gmail.com wrote:
C, there is also a tor-relays-universities list. Forwarding there to keep the initial chat primed.
Once you have buy in from legal, chairs, security, upstream, etc this can be a very strong position, often better than pay 'contract' of random ISP host. I have seen such 'outside' nets used for these not strictly mission things, such I suggest it in this thread. Different approach depends on if you can find and house a legitimate paper producing research purpose, or if you simply will run it for supporting freedom point of view.
Our experience is similar; after an initial period of discomfort (both I and my supervisor had "yes, the IDS will continue to get false positives on this host, it is not actually infected with a virus / ridiculously out of date on security patches / running Windows XP" form letters on tap for a while), the university has been very supportive.
We *were* asked to make sure that the exit policy denied access to library resources that the university has a contractual obligation to make available only to affiliates, but they gave us a list of IPs to blacklist, so that was no problem. You might want to mention up front that you can do that if they need you to.
zw
tor-relays@lists.torproject.org
-
Cristóbal Palmer -
grarpamp -
Zack Weinberg