On 8/8/26 01:17, Linus Nordberg via tor-consensus-health wrote: > [email protected] wrote > Sat, 08 Aug 2026 04:05:21 +0000 (UTC): > >> ERROR: The following directory authorities are not reporting bandwidth >> scanner results: maatuska, bastet > > My sbws keeps consuming all available RAM and swap, together with a lot > of CPU. I have turned it off for now and disabled bw voting for > maatuska.
So what changed for this to happen to you, Stefani and me? All the memory including swap maxes out over a few hours. The two py processes eat a lot of CPU also. The BSDs recycle memory only when they need it, so I'm not too concerned... but it's very much worth putting my sbws in the same bucket as Linus and Stefani IMHO. My sbws box is pretty lame in terms of specs, but it's in the recent past that this resource consumption started happening. I regressed back to the pre-patch version that juga has applied, and it continues. I would assume it's tied to the network and not sbws itself. Could something has changed on the network to affect measurements? I have some more things to look at on my end... g _______________________________________________ tor-consensus-health mailing list -- [email protected] To unsubscribe send an email to [email protected]
