June 2014 - tor-relays - lists.torproject.org

Lost hsdir flag
by kingqueen 21 Jun '14

21 Jun '14

hello, I don't know why, my relay nicknamed kingqueenbtnftw used to have the hsdir flag but now doesn't. I think from what I can gather, hsdir is based purely on the time since last boot / restart Tor service, rather than how long the relay has been up in total? That would explain it. Thanks kingqueen --- This email is free from viruses and malware because avast! Antivirus protection is active. http://www.avast.com

2 2

Traffic limitation
by johhher 20 Jun '14

20 Jun '14

Hi, I'm running a Tor relay on a cheap Linux vserver with high bandwidth. I have a traffic limitation of 500Gb per month and was just wondering what would be the best configuration for the network. Currently I've limited the traffic to a daily maximum to prevent it from hibernating a long time at the end of the month. Is this the right way to do it? Or would be better to limit it on a monthly basis. Thanks for an answer! Johhher

6 5

Yahoo and AOL emails removed from tor-relays due to bounces
by Andrew Lewman 20 Jun '14

20 Jun '14

It seems Yahoo and AOL email servers have decided to refuse all messages from this tor-relays list. Other lists aren't affected, yet. We're seeing this message from Yahoo and AOL servers: "Status: 5.2.1 Remote-MTA: dns; mailin-04.mx.aol.com Diagnostic-Code: smtp; 521 5.2.1 : (DMARC) This message failed DMARC Evaluation and is being refused due to provided DMARC Policy" There is no fix for this yet as it breaks mailing list software globally. This bug explains the situation, https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=752084 If you use yahoo or aol for email, unfortunately you can't join nor receive email from our lists. I suspect this will become true for all of our email lists soon. -- Andrew pgp 0x6B4D6475 https://www.torproject.org/ +1-781-948-1982

3 2

CentOS of the air
by I 19 Jun '14

19 Jun '14

"At about 02:00 UTC June 18th 2014, changes at the domain registrar caused the CentOS.org domain to get marked as Inactive. This caused widespread outages, with all services in centos.org including: - - Mirror Network ( internal centos.org as well as external ) - - All mirrorlist - - All isolist - - Website ( including bugs.centos.org and wiki.centos.org - - All internal support infrastructure Services were restored at the registrar just before 9:00 AM UTC June 18th, however depending on dns path some users might still see issues for upto 6 to 8 hrs from that point.

1 0

tor-arm question
by Adam Griffin 19 Jun '14

19 Jun '14

Hi, Just started a new tor relay on my raspberry pi yesterday and I'm monitoring with tor-arm over ssh. I'm currently listed on atlas.torproject.org with flags 'Fast', 'Running', 'V2Dir', and 'Valid' with > 17hrs uptime. However, tor-arm is showing a blank for uptime and about once a minute, the CLI goes blank and then flickers back. (the flickering happens at the same rate for local and remote SSH) Also, about once every couple hours I get a 'Relay unresponsive' and then 10 seconds later 'Relay resumed'. Anything to do to remedy these issues?

3 5

Help keeping Tor relay up
by Evelyne Fong 19 Jun '14

19 Jun '14

I've managed to get my relay running long enough to run the bandwidth check and have it succeed. I'll check Atlus and it'll say my relay's up and running. However when I check again about an hour later, it's saying that "Running: false". I've set up the relay recently, so I'm not expecting a lot of traffic. However, I'm confused as to if my relay is suddenly failing for no reason or is Atlus saying my relay is no longer running because there's no traffic on it. Any help would be appreciated. Thanks.

3 3

Debian relay Puppet module
by Alexander Fortin 19 Jun '14

19 Jun '14

Hi folks, I’ve recently joined the Tor Project and I have been running a non exit relay for a few days. I’m also a Puppet user and, more in general, I try to make deploying applications on the servers I administer as easy as possibile, Tor included. I think Tor documentation to install on a Debian server is quite good, but I still prefer to have Puppet doing that for me, and I’m quite sure every Puppet user out there would think the same. This is the work-in-progress version of the module I’m currently using to manage my relay: https://github.com/shaftoe/puppet-tor/tree/fixes I’m open to any feedback on how to improve it and eventually publish it to the Puppet forge (make it easier for other Puppet users to use it); maybe adding some decent disclaimer on top of the README? I can also drop the whole idea if someone convince me is a bad one... -- Alexander Fortin http://about.me/alexanderfortin

9 22

Tor-relay - IP
by Rui Branco 18 Jun '14

18 Jun '14

hello, I'm a new user of the Tor-relay and I have many doubts. Sorry about that. In the atlas project why the tor-relays ip are discovered? isn't it dangerous? -- Best Regards, Rui Branco

4 3

normal for a new relay node to be under-utilized?
by Kali Tor 18 Jun '14

18 Jun '14

Is it normal for a new relay node to carry almost no traffic even after 24 hrs? From Atlast Uptime: 1 day 8 hours 30 minutes and 28 seconds Advertised Bandwidth: 59.71 KB/s From arm Bandwidth (limit: 16 Mb/s, burst: 32 Mb/s, measured: 160.0 b/s) Upload (0.0 b/sec - avg: 738.8 b/sec, total: 10.8 MB) Download (0.0 b/sec - avg: 3.6 Kb/sec, total: 52.2 MB) torrcRelayBandwidthRate 2048 KB RelayBandwidthBurst 4096 KB AccountingMax 100 GB AccountingStart month 1 03:00 DirPort 9030

3 2

Advertised Bandwidth increasing
by kingqueen 18 Jun '14

18 Jun '14

Hi I'm running the new, imaginatively-nicknamed kingqueenbtnftw relay. I have read around a lot but something I don't get. I thought advertised bandwidth was what I'd put in torrc whereas consensus weight was what some uber-machines tested my relay at, with the idea that this can foil disruption attacks by false relays pretending to give high bandwidth. SO I expected consensus weight to gradually increase. What I didn't expect was for the advertised bandwidth to increase by step changes (as shown on atlas and globe). Evidently I've got something wrong. Could somebody please enlighten this poor nub? By the way, I appreciate that consensus weight is a dimensions number but I understand that it originated as being based on kibibytes per second? If this is still the case, it's currently significantly more than my advertised bandwidth (which is less than that set in TORRC) but appears below it in the graphs? Thanks to anyone who educates me. I have full intent on both the T-shirts on offer - the standard Tor one for uptime and bandwidth, and the EFF one for still being online in a year with over 1MBps transfer :-) Hi to all who know me Kingqueen

2 2