[Author Prev][Author Next][Thread Prev][Thread Next][Author Index][Thread Index]
Re: [tor-bugs] #8532 [Tor]: Make TestingTorNetwork produce consensuses quicker
#8532: Make TestingTorNetwork produce consensuses quicker
---------------------------------------+------------------------------------
Reporter: ln5 | Owner:
Type: defect | Status: new
Priority: normal | Milestone: Tor: 0.2.5.x-final
Component: Tor | Version:
Keywords: SponsorF20131031 tor-auth | Parent:
Points: | Actualpoints:
---------------------------------------+------------------------------------
Comment(by ln5):
In order for clients and relays to be able to postpone the first
consensus download for about 10s (using
Testing*ConsensusDownloadSchedule), run_scheduled_events() needs to
call update_networkstatus_downloads() more often than once per minute
(CHECK_DESCRIPTOR_INTERVAL).
Would letting it run every second be scary in any way? I'm thinking
about things like hosing the tor doing this and also about hammering
other tors for network status documents.
In update_networkstatus_downloads(), dir auths call
update_v2_networkstatus_cache_downloads() which currently won't do
anything with a higher frequency than ten minutes
(AUTHORITY_NS_CACHE_INTERVAL). That looks safe to me.
Then, every kind of tor invokes
update_consensus_networkstatus_downloads() which seems to have a few
guards -- time_to_download_next_consensus[i],
download_status_is_ready() and
connection_dir_get_by_purpose_and_resource().
To conclude, I don't see that it would be harmful to have
update_networkstatus_downloads() being invoked every second but would
really like to hear from other people on this.
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/8532#comment:3>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
_______________________________________________
tor-bugs mailing list
tor-bugs@xxxxxxxxxxxxxxxxxxxx
https://lists.torproject.org/cgi-bin/mailman/listinfo/tor-bugs