[tor-relays] SOLVED: Failing because we have 4063 connections already // Number of file descriptors

2017-12-15 Thread Ralph Seichter
On 15.12.2017 11:12, Toralf Förster wrote: > # cat /etc/conf.d/tor > # > # Set the file limit > rc_ulimit="-n 3" Ah, thanks a lot! Limits were implied by openrc-run/start-stop-daemon, overriding my limits.conf entries. Turns out that /etc/conf.d/tor got deleted, although I have no idea how

Re: [tor-relays] Is your IPv6 relay not Running?

2017-12-15 Thread Toralf Förster
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 12/15/2017 11:46 AM, teor wrote: > >> On 15 Dec 2017, at 21:09, Toralf Förster wrote: >> >>> On 12/11/2017 11:20 PM, teor wrote: >>> >>> We're working on having better support for IPv6 across Relay Search and >>>

Re: [tor-relays] Is your IPv6 relay not Running?

2017-12-15 Thread teor
> On 15 Dec 2017, at 21:09, Toralf Förster wrote: > >> On 12/11/2017 11:20 PM, teor wrote: >> >> We're working on having better support for IPv6 across Relay Search and >> consensus health. > > At my 2 relays (1AF72E8906 and D11D1187776) I have both ipv4 and ipv6 >

Re: [tor-relays] exit relay

2017-12-15 Thread Gary Smith
Hello. If it is not set in your torrc, try looking in the torrc-default file /usr/share/tor/tor-service-defaults-torrc If you forget where this file is, look in the log where for two lines near the top for "loading torrc" or something If nothing is there, write EXITRELAY 0 yourself, right at

Re: [tor-relays] Failing because we have 4063 connections already // Number of file descriptors

2017-12-15 Thread r1610091651
On Fri, 15 Dec 2017, 11:13 r1610091651, wrote: > That depends on how tor is started and have different origins. What i know: > * if started by systemd: the limit can be specified in the service > descripton file /lib/systemd/system/tor@default.service: => >

Re: [tor-relays] Failing because we have 4063 connections already // Number of file descriptors

2017-12-15 Thread r1610091651
That depends on how tor is started and have different origins. What i know: * if started by systemd: the limit can be specified in the service descripton file /lib/systemd/system/tor@default.service: => LimitNOFILE= * if started by init: usually pam/security will be applied => using

Re: [tor-relays] Failing because we have 4063 connections already // Number of file descriptors

2017-12-15 Thread Toralf Förster
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 12/15/2017 10:38 AM, Ralph Seichter wrote: > My relay uses Gentoo Linux kernel version If you run a Gentoo system then take a look at this file : # cat /etc/conf.d/tor # # Set the file limit rc_ulimit="-n 3" - -- Toralf PGP C4EACDDE

Re: [tor-relays] Is your IPv6 relay not Running?

2017-12-15 Thread Toralf Förster
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 12/11/2017 11:20 PM, teor wrote: > > We're working on having better support for IPv6 across Relay Search and > consensus health. At my 2 relays (1AF72E8906 and D11D1187776) I have both ipv4 and ipv6 activated. The load is about 1.5 TByte/day.

Re: [tor-relays] Failing because we have 4063 connections already // Number of file descriptors

2017-12-15 Thread Ralph Seichter
On 15.12.2017 10:45, r1610091651 wrote: > could be that your tuning is not being picked up by the distro. Looks like you are right: # cat /proc/3534/limits Limit Soft Limit Hard Limit Units Max cpu time unlimited unlimited seconds Max file size

Re: [tor-relays] Failing because we have 4063 connections already // Number of file descriptors

2017-12-15 Thread r1610091651
Hi Please verify the effective limit used for your tor process: cat /proc//limits with process id of the tor process. could be that your tuning is not being picked up by the distro. Regards On Fri, 15 Dec 2017 at 10:39 Ralph Seichter wrote: > Since a couple of days

[tor-relays] Failing because we have 4063 connections already // Number of file descriptors

2017-12-15 Thread Ralph Seichter
Since a couple of days ago, one of my relay nodes keeps logging messages like this: Tor[3534]: Failing because we have 4063 connections already. Please read doc/TUNING for guidance. [over 1601 similar message(s) suppressed in last 21600 seconds] I found

Re: [tor-relays] Fwd: someone is livestreaming a bad exit

2017-12-15 Thread grarpamp
> This guy does not seem to understand why his “experimentation” was dangerous. What's more dangerous than some youtube stunt would be foolishly failing to understand that perhaps half the nodes out there could easily be secret experiments, even mass sybil operations, dangerous to the users and