Opened 6 years ago

Closed 6 years ago

Last modified 5 years ago

#3803 closed defect (fixed)

connection_buckets_decrement() does lots of stuff that bufferevents need to do someplace

Reported by: nickm Owned by: nickm
Priority: Medium Milestone: Tor: 0.2.3.x-final
Component: Core Tor/Tor Version:
Severity: Keywords: tor-client
Cc: Actual Points:
Parent ID: #3561 Points:
Reviewer: Sponsor:

Description

With bufferevents, connection_buckets_decrement is a no-op. But it is supposed to do lots of stuff (particularly wrt rephist) that somebody should really be doing. I should check that out and verify that this stuff is actually getting called somewhere, or make it get called if it isn't.

Child Tickets

Change History (6)

comment:1 Changed 6 years ago by nickm

Owner: set to nickm
Status: newaccepted

comment:2 Changed 6 years ago by nickm

Status: acceptedneeds_review

See branch bug3803 in my public repo. It may also fix #3824 and #3826.

comment:3 Changed 6 years ago by Sebastian

changelog typo: "we're" -> "we've"

Otherwise, looks ok. One question: Inside record_num_bytes_transferred(), we don't bail out anymore if num_read or num_written is larger than INT_MAX. I took a quick look at the rephist code, and that seemed ok. Still, was that change intentional?

comment:4 Changed 6 years ago by nickm

Resolution: fixed
Status: needs_reviewclosed

Hm. It was intentional, but at this point, I think I'd rather actually do it redundantly. Made the fixes you suggested and merged.

comment:5 Changed 5 years ago by nickm

Keywords: tor-client added

comment:6 Changed 5 years ago by nickm

Component: Tor ClientTor
Note: See TracTickets for help on using tickets.