Opened 3 years ago
Last modified 2 years ago
#20505 new defect
make min_bw configurable in bwauths
Reported by: | teor | Owned by: | |
---|---|---|---|
Priority: | Medium | Milestone: | |
Component: | Core Tor/Torflow | Version: | |
Severity: | Normal | Keywords: | |
Cc: | Actual Points: | ||
Parent ID: | Points: | ||
Reviewer: | Sponsor: |
Description
In a testing network, there isn't much traffic. So skipping nodes with an observed bandwidth less than 1 MByte/second means that it's easy to skip a large fraction of the network.
It would be helpful if min_bw were a configurable option in bwauthority.cfg. If it is an option, we could consider making the unmeasured node scanner measure all nodes, regardless of bandwidth.
Child Tickets
Attachments (1)
Change History (9)
Changed 3 years ago by
Attachment: | 0003-Tune-scanning-for-small-networks.patch added |
---|
comment:1 follow-up: 2 Changed 3 years ago by
This doesn't look like something we would want to apply to the main repo. Where would you want it to live? In a separate branch? I'm doubtful it would be helpful to re-architect things sufficiently that this was easily parameterized...
comment:2 Changed 3 years ago by
Replying to tom:
This doesn't look like something we would want to apply to the main repo. Where would you want it to live? In a separate branch? I'm doubtful it would be helpful to re-architect things sufficiently that this was easily parameterized...
I'm not sure - this is an architectural question:
Core tor has an option that sets a whole lot of parameters useful for test networks.
Perhaps we could do the same with the bandwidth scanner. But that means the resulting changes are scattered throughout the code.
Maybe a separate branch is better, but then it easily gets outdated.
Ideally, we could make these parameters configurable, and distribute a testing config (or a set of testing defaults).
Perhaps documenting this issue and providing a patch is the best strategy?
comment:3 Changed 2 years ago by
Priorities and Severities in torflow are meaningless, setting them all to Medium/Normal.
comment:4 Changed 2 years ago by
Owner: | aagbsn deleted |
---|---|
Status: | new → assigned |
aagbsn was the default owner, unassigning
comment:5 Changed 2 years ago by
I applied the patch but that wasn't enough for bwauth to work in a small test netork. It seemed to go OK at first, but then scanner.1 went into logspam mode with these lines:
./bwauthority.py:71: SyntaxWarning: name 'p' is assigned to before global declaration global p NOTICE[Sat Nov 11 16:13:10 2017]:TorFlow Version: master 75abb44bf73b3a061063145ebdb97029afe91fb7 NOTICE[Sat Nov 11 16:13:10 2017]:TorCtl Version: detached c8fcb25b079d52a20cafc7f7adf178e90ab76338 NOTICE[Sat Nov 11 16:13:10 2017]:Child Process Spawned... NOTICE[Sat Nov 11 16:14:13 2017]:Starting slice for percentiles 0-100 NOTICE[Sat Nov 11 16:14:15 2017]:Ran out of choices in ExactUniformGenerator. Incrementing nodes NOTICE[Sat Nov 11 16:14:15 2017]:Ran out of routers during buildpath.. NOTICE[Sat Nov 11 16:14:15 2017]:Ran out of choices in ExactUniformGenerator. Incrementing nodes NOTICE[Sat Nov 11 16:14:15 2017]:Ran out of routers during buildpath..
comment:6 Changed 2 years ago by
Would you mind opening a separate ticket with these log lines?
And describing the size of your network (authorities, relays, exits)?
I'll see if I have any local patches that might help.
If we ever want to be able to run bwauthority in chutney, we'll need to fix issues like this.
comment:8 Changed 2 years ago by
Status: | assigned → new |
---|
Mark all tickets that are assigned to nobody as "new".
Workaround for small networks