Opened 2 years ago

Closed 23 months ago

#24160 closed defect (fixed)

Wrong info about bridges?

Reported by: eduperez Owned by: metrics-team
Priority: Medium Milestone:
Component: Metrics/Relay Search Version:
Severity: Normal Keywords:
Cc: Actual Points:
Parent ID: Points:
Reviewer: Sponsor:

Description

I am currently running a bridge, and have tried to check it at atlas.torproject.org, but the info there seems to be wrong.

I have copied the contents of the "hashed-fingerprint" file into the search box at ATLAS, and the info shown there seems to belong to my node: "nickname", "first seen", "last restarted", and "platform" all match my node. However, "OR address", and "advertised bandwidth" are off:

  • I have configured my node as "ORPort 9001", and effectively port 9001 is open and listening; but ATLAS shows "IPv4:52563" at "OR address".
  • I have also configured "RelayBandwidthRate 512 KBytes" and "RelayBandwidthBurst 1024 KBytes", but ATLAS shows "55.09 KiB/s" at "advertised bandwith".

Is this a known issue with ATLAS? Or perhaps I am misinterpreting the page? Many thanks!

Child Tickets

Change History (6)

comment:1 in reply to:  description Changed 2 years ago by karsten

Replying to eduperez:

I am currently running a bridge, and have tried to check it at atlas.torproject.org, but the info there seems to be wrong.

I have copied the contents of the "hashed-fingerprint" file into the search box at ATLAS, and the info shown there seems to belong to my node: "nickname", "first seen", "last restarted", and "platform" all match my node. However, "OR address", and "advertised bandwidth" are off:

  • I have configured my node as "ORPort 9001", and effectively port 9001 is open and listening; but ATLAS shows "IPv4:52563" at "OR address".

The reason is that we're sanitizing TCP ports in order to make it harder to discover bridges with unusual ports. See https://metrics.torproject.org/bridge-descriptors.html#tcp-port

I wonder if we should take out bridge OR address and ports from Atlas entirely. They were added before we started sanitizing TCP ports, and then they made some sense. But that may not be the case anymore.

  • I have also configured "RelayBandwidthRate 512 KBytes" and "RelayBandwidthBurst 1024 KBytes", but ATLAS shows "55.09 KiB/s" at "advertised bandwith".

The advertised bandwidth does not only depend on your configuration but also on what your relays reports. See https://metrics.torproject.org/glossary.html#advertised-bandwidth

Is this a known issue with ATLAS? Or perhaps I am misinterpreting the page? Many thanks!

Not a code issue, but possibly a documentation issue. I'll leave it up to irl, the Atlas maintainer, to say whether there's something to fix in the documentation or presentation. Thanks for the report!

comment:2 Changed 2 years ago by eduperez

Ok, it all makes sense now; thanks for the info!

comment:3 Changed 2 years ago by cypherpunks

Resolution: not a bug
Status: newclosed

comment:4 Changed 2 years ago by karsten

Resolution: not a bug
Status: closedreopened

Hang on, let's hear from irl whether he'd like to change the documentation or presentation to avoid similar confusions in the future.

comment:5 Changed 2 years ago by irl

Not my top priority, but please keep the ticket around and I'll look into updating the explanations around it. I like that Atlas can explain concepts to relay operators as well as just provide data and I'd like to achieve that.

comment:6 Changed 23 months ago by irl

Resolution: fixed
Status: reopenedclosed

Fixed in c371ef4 and 8b45726.

Note: See TracTickets for help on using tickets.