[tor-relays] amount of unmeasured relays continuously rising since 2 weeks

micah anderson micah at riseup.net
Wed May 20 15:58:25 UTC 2015


Roger Dingledine <arma at mit.edu> writes:

> On Mon, May 18, 2015 at 09:02:27PM +0000, nusenu wrote:
>> now even DocTor starts to complain
>> 
>> https://lists.torproject.org/pipermail/tor-consensus-health/2015-May/005772.html

>From what I've seen, there were only two messages from DocTor about
this (Mon May 18 08:06:49 UTC 2015[0], and Tue May 19 10:06:13 UTC
2015[1).

> Yep. moria1's bwauth seems to be working fine, so I'm the only one
> paying attention to the thread here, and also the only one who doesn't
> have anything to fix. :(

I'm seeing this thread as well, but its not apparently obvious that
there is something to fix.

nusenu <nusenu at openmailbox.org> writes:
>sounds alarming if people operating tor's key infrastructure do not
>care (or have not enough time to care)

I don't think that is a fair assessment, I dont think anyone running a
bwauth is failing to care, or doesn't prioritize the proper running of
the service. There were only two complaints from DocTor, and it has
stopped. Two isolated warning usually isn't a cause for alarm, or
suggestions that bwauths are not caring.

There are two possibilities here:

1. there is an actual problem on the existing bwauths that needs to be
fixed: if that is true, it is not obvious what the problem is. The
bandwidth for longclaw seems to be working fine, and the bwauths
processes seem to be doing what they are supposed to be doing. If there
is something amiss with longclaw's bwauths that I am missing, and can do
something to resolve, please let me know. 

2. none of the bwauths are down, but rather we've reached a tipping
point and the bwauths cannot measure the entire network in the
measurement run. if this is true, then this is indicative of a trend and
we will be seeing more of these alerts from DocTor in the future, and
this is not a case of bwauths failing to care for critical
tor infrastructure

micah

0. https://lists.torproject.org/pipermail/tor-consensus-health/2015-May/005772.html
1. https://lists.torproject.org/pipermail/tor-consensus-health/2015-May/005773.html



More information about the tor-relays mailing list