New round of contacting operators for DNS issues and badexiting problematic relays (02/08/2021)
We are good here this week. No request related issues in my last scan that were not already mentioned and dealt with in #24 (closed).
We have some accumulation of circuit failures with TIMEOUT reason for
the following relays (those are persistent and show up during manual
testing, too):
3E50CBCA98A20F637BC4551FD4F132D062DB9A51
A2534EF23390CAE079B1586F0FDF9CE11F556062
A2DB293FFC5A76A718863BF1AEDBC8DFB1CB1097
CA94704217260E7483DA88719CACD7A94C564D5C
D6D6B6614C9EF2DAD13AC0C94487AD8ED3B6877F
E43A346CB81DDF364B6FF68235AFADBA0E8692B8
FD60E8CC44458386A3666DB6F20B3EE37AA9F6AD
CF97B121E511B80125ED8DFF27CA403A480CB20AThe first group belongs to a family related to Torservers and all of those relays have an elaborate exit policy (while members of that family not having issues do have a different exit policy). Not sure if that's related.
The other relay exhibits the same measurement issues but does not belong to the same family.
Needless to say, using those relays in Tor Browser works fine. So, maybe that points to connectivity issues related to the guard relay I am using.
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information