[network-health] How is a relay's uptime determined?

[Silvia/Hiro]

On 19/12/22 18:49, arf20 wrote:
> Hello there,
> I run a tor middle relay, and last night I restarted the service due 
> to an update.
> Then, looking up the relay in the Tor site, it looks like I lost all 
> the uptime and flags.
> A bit annoying since I have had the server down for several minutes 
> without consequences in other times. I was 5 days away from the t-shirt :(
> I was just wondering under what circumstances the relay's uptime is 
> reset, if possible.
> Thanks.
>
Hello!

The uptime is set on onionoo 
(https://gitlab.torproject.org/tpo/network-health/metrics/onionoo/-/blob/master/src/main/java/org/torproject/metrics/onionoo/updater/NodeDetailsStatusUpdater.java#L225) 
and depends on what is published on the latest descriptor.

The flags are instead parsed from the latest consensus 
(https://gitlab.torproject.org/tpo/network-health/metrics/onionoo/-/blob/master/src/main/java/org/torproject/metrics/onionoo/updater/NodeDetailsStatusUpdater.java#L259)

I wonder if your relay was not included in a consesus. You may try to 
check the consensus and descriptors published around the time you 
restarted your relay.

cheers,

-hiro


> --
> arf20
>
> _______________________________________________
> network-health mailing list
> network-health at lists.torproject.org
> https://lists.torproject.org/cgi-bin/mailman/listinfo/network-health