On 22 Mar 2017, at 07:46, Tyler Johnson tylrcjhnsn@gmail.com wrote:
Hello!
I've enabled bandwidth accounting on my tor relay, but encountering issues when the relay attempts to wake from hibernation....
Mar 20 23:47:25 tcj Tor[74346]: Bandwidth soft limit reached; commencing hibernation. No new connections will be accepted Mar 20 23:52:30 tcj Tor[74346]: Heartbeat: Tor's uptime is 10 days 12:00 hours, with 150 circuits open. I've sent 90.20 GB and received 90.30 GB. We are currently hibernating. Mar 20 23:52:30 tcj Tor[74346]: Circuit handshake stats since last time: 2745/2745 TAP, 15557/15557 NTor. Mar 20 23:52:30 tcj Tor[74346]: Since startup, we have initiated 0 v1 connections, 0 v2 connections, 0 v3 connections, and 113565 v4 connections; and received 1100 v1 connections,130 v2 connections, 41 v3 connections, and 249967 v4 connections. Mar 21 00:00:00 tcj Tor[74346]: Configured hibernation. This interval began at 2017-03-21 00:00:00; the scheduled wake-up time was 2017-03-21 00:00:00; we expect to exhaust our quota for this interval around 2017-03-22 00:00:00; the next interval begins at 2017-03-22 00:00:00 (all times local) Mar 21 00:00:00 tcj Tor[74346]: Hibernation period ended. Resuming normal activity. Mar 21 00:00:00 tcj Tor[74346]: Opening OR listener on 185.92.223.3:8080 Mar 21 00:00:00 tcj Tor[74346]: Could not bind to 185.92.223.3:8080: Address already in use. Is Tor already running? Mar 21 00:01:00 tcj Tor[74346]: Opening OR listener on 185.92.223.3:8080 Mar 21 00:01:00 tcj Tor[74346]: Could not bind to 185.92.223.3:8080: Address already in use. Is Tor already running? Mar 21 00:02:01 tcj Tor[74346]: Opening OR listener on 185.92.223.3:8080 Mar 21 00:02:01 tcj Tor[74346]: Could not bind to 185.92.223.3:8080: Address already in use. Is Tor already running?
It will continue in this manner until I manually intervene and restart the relay instance.
This seems to be an internal state management issue. I have opened ticket #21810 to track it. https://trac.torproject.org/projects/tor/ticket/21810
I would like to confirm tor's behaviour when this happens:
Is there something listening on 185.92.223.3:8080? Is it this tor process, or another tor process? If it is this tor process, is tor still working as a relay despite the warnings?
snippet from torrc:
## Set a maximum of 30 gigabytes each way per period. AccountingMax 30 GBytes
## Each period starts daily at midnight (AccountingMax is per day) AccountingStart day 00:00
## Required: what port to advertise for incoming Tor connections. ORPort 185.92.223.3:8080
## The IP address or full DNS name for incoming connections to your ## relay. Leave commented out and Tor will guess. Address 185.92.223.3
## If you have multiple network interfaces, you can specify one for ## outgoing traffic to use. OutboundBindAddress 185.92.223.3
Tor version and OS: $ tor --version Tor version 0.2.9.10 (git-1f6c8eda0073f464). $ uname -a OpenBSD tcj.me 6.1 GENERIC#224 amd64
We sometimes have bugs on OpenBSD because it's one of our rarer platforms.
Do you happen to know if SO_RERUSEADDR has unusual semantics on OpenBSD? (Or is it disabled entirely?)
Has anyone else experienced this issue?
This code has had similar issues in the past, particularly on rarer platforms.
I have only one outbound interface, so is setting the OutboundBindAddress unnecessary and possibly causing the problem?
No, OutboundBindAddress is completely unrelated. OutboundBindAddress is used for connections initiated by your relay. ORPort is used for connections accepted by your relay.
T
-- Tim Wilson-Brown (teor)
teor2345 at gmail dot com PGP C855 6CED 5D90 A0C5 29F6 4D43 450C BA7F 968F 094B ricochet:ekmygaiu4rzgsk6n xmpp: teor at torproject dot org ------------------------------------------------------------------------