On Sat, Oct 12, 2019 at 12:08:02PM +1100, Damien Gardner wrote:
That was a fun morning! Managed to get that old VM up to Ubuntu 16.04, which you still have a repo for, and got tor upgraded on it :)
Awesome! Looks good, thanks.
Though looks like the node may have been blacklisted due to the old tor version?
Oct 12 12:02:10 minbar Tor[12066]: http status 400 ("Fingerprint is marked rejected -- if you think this is a mistake please set a valid email address in ContactInfo and send an email to bad-relays@lists.torproject.org mentioning your fingerprint(s)?") response from dirserver '204.13.164.118:80'. Please correct.
Yes, this is exactly what happened. We moved forward with the plan to remove the old relays: https://blog.torproject.org/removing-end-life-relays-network
Is that what these errors mean? If that's all it is, are you able to un-blacklist it please
Yes, I've started the process of unblacklisting the fingerprint. It will need a threshold of the directory authorities to pull the update, so it could be a couple of days until it takes effect, but hopefully it will be relatively soon.
You can go to the bottom of https://consensus-health.torproject.org/#relayinfo and paste in your fingerprint or nickname to see progress.
Hoping this is new enough? If not, I can try taking it up to ubuntu 18.04, but that may be problematic as it's living on a VERY old xen host :D
Hm. It looks based on https://ubuntu.com/about/release-cycle that you have another year and a bit until Ubuntu 16.04 goes dead.
So, that's plenty of time to plan ahead, especially if you start soon. :)
Regarding bandwidth, I don't have any of the RelayBandwidth* setting configured in torrc - so had always just assumed it would be open slather. Do I need to specifically set these to 0? As long as it doesn't go nuts (aka saturate the gig ethernet into the host or affect paying customers on the segment), I don't really mind too much now much traffic it uses.
Ok, sounds good, we'll see where it goes from here.
(It looks like the relay is located in Australia? That would explain why it isn't growing in usage as much as relays closer to the middle of the internet -- Tor tries to shift load toward relays that can handle it better than their peers, and the huge latencies to and from Australia make it harder to look good compared to the other relays.)
How often does the node status page update? ( https://metrics.torproject.org/rs.html#details/72657B96BF99B28FBD9F2B036A590...) It's been a few hours now since the timestamp at the bottom was updated, when I convert from UTC to local :)
I'm not sure how often exactly, but pretty often. Usually often enough that it is not the bottleneck at least. Once some directory authorities update things should start looking better.
Thanks! --Roger