[metrics-team] Problem with RelaySearch or the relay data

teor teor2345 at gmail.com
Mon Dec 11 22:09:51 UTC 2017


Hi,

> On 12 Dec 2017, at 06:38, Andreas Piesk <a.piesk at mailbox.org> wrote:
> 
> Hello list,
> 
> Relay Search shows wrong data of one of my relays. I'm not sure if the data is wrong or if relay search does the wrong thing or the directory server doesn't like te relay.
> 
> The relay's fingerprint is 43AD3376735B4A227920AD06D2D9B6970B3C75DF and it is running 3.1.9 on Linux.
> 
> Relay Search reports:
> 
> Downtime
>   2 days 23 hours 20 minutes and 15 seconds
> Last Seen
>   2017-12-08 20:00:00Tor 0.3.1.9 on Linux
> Additional Flags
>   Not Recommended IPv6 ORPort
> Last Restarted
>   2017-12-10 23:12:29
> Platform
>   Tor 0.3.1.9 on Linux
> 
> It got the "Not Recommended" flag because of "This relay is running an outdated Tor version and should be updated to a recent release of Tor that may contain important fixes.".
> 
> Something is not right:
> 
> 1. the relay IS running and i see V2, V3 and V4 connections.

The relay is reachable over IPv4, but not IPv6.

The 5 directory authorities that check ReachableIPv6 have marked your relay
as not Running. That means the IPv6 address is wrong or unreachable.
(The 3 that only check IPv4 marked it as Running, but they're not a majority.
And 1 authority did not vote.)

Also, the bandwidth authorities are still voting for your relay.

This page lists all the votes for your relay:
https://consensus-health.torproject.org/consensus-health-2017-12-11-17-00.html#43AD3376735B4A227920AD06D2D9B6970B3C75DF

We're working on having better support for IPv6 across Relay Search and
consensus health.

> 2. if it was last seen 2017-12-08 20:00:00,

It was last seen in the consensus.

> where does the restart date 2017-12-10 23:12:29 come from?

The restart date comes from the descriptor.

> 3. 3.1.9 is not outdated as Relay Search reports

I'm not sure if you received nusenu's email:

> this has likely todo with this bug:
> https://trac.torproject.org/projects/tor/ticket/22488

> How to fix it? Unfortunately there is no way to delete the relay and start from scratch,is it?
> I don't want generate a new key and leave this relay as a zombie in the database, i would like to keep thois relay.

Please check your relay's IPv6 address!


Here's what I do:

From the relay:
ping6 ipv6.google.com

From another IPv6 machine:
ping6 (the relay's IPv6 address)

Or you can try using telnet to the ORPort.

T
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.torproject.org/pipermail/metrics-team/attachments/20171212/5c8c494b/attachment.html>


More information about the metrics-team mailing list