nihilist/tor
1
0
Fork 0
mirror of https://gitlab.torproject.org/tpo/core/tor.git synced 2024-11-27 22:03:31 +01:00
Code Issues Packages Projects Releases Wiki Activity

Launch early reachability tests once a minute, not once a second.

Browse Source 
This fixes bug 40083, which was introduced in 9f31908a in
0.2.8.1-alpha.
This commit is contained in:
Nick Mathewson 2020-08-04 12:59:49 -04:00
parent c4742b89b2
commit 62361cc698
2 changed files with 9 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
changes/bug40083 Normal file
View File

@ -0,0 +1,5 @@
o Minor bugfixes (relay, self-testing):
- When starting up as a relay, if we haven't been able to verify that
we're reachable, only launch reachability tests at most once a minute.
Previously, we had been launching tests up to once a second, which
was needlessly noisy. Fixes bug 40083; bugfix on 0.2.8.1-alpha.

5
src/core/mainloop/mainloop.c
View File

@ -2280,6 +2280,9 @@ check_for_reachability_bw_callback(time_t now, const or_options_t *options)
{
/* XXXX This whole thing was stuck in the middle of what is now
* XXXX check_descriptor_callback. I'm not sure it's right. */
/** How often should we consider launching reachability tests in our first
* TIMEOUT_UNTIL_UNREACHABILITY_COMPLAINT seconds? */
#define EARLY_CHECK_REACHABILITY_INTERVAL (60)
static int dirport_reachability_count = 0;
/* also, check religiously for reachability, if it's within the first
@ -2291,7 +2294,7 @@ check_for_reachability_bw_callback(time_t now, const or_options_t *options)
router_do_reachability_checks(1, dirport_reachability_count==0);
if (++dirport_reachability_count > 5)
dirport_reachability_count = 0;
return 1;
return EARLY_CHECK_REACHABILITY_INTERVAL;
} else {
/* If we haven't checked for 12 hours and our bandwidth estimate is
* low, do another bandwidth test. This is especially important for