Closed
Milestone
expired on Jul 1, 2021
sbws: 1.1.x-final
This release of sbws will be the one that will be generaly deployed.
Unstarted Issues (open and unassigned)
0
Ongoing Issues (open and assigned)
0
Completed Issues (closed)
105
- Clarify Torflow aggregation in docs
- Revert to calculate network stream and filtered bandwidth averages without relay types
- Re-format the code to following the team Python guidelines
- Return from `measure_relay` when a helper is not found
- Revert to SocksPort auto configuration
- Check the info level logs and revert to use info level by default
- Calculate network stream and filtered bandwidth averages per relay type
- Document better the differences between sbws and torflow scaling
- Update tor and Python versions in the tests and requirements
- Teach sbws to use SNI or move off of Fastly for bwauth files
- Compare bandwidth values across relays
- Find out which relays are in torflow but not sbws after generating bandwidth file with 28 days of measurements
- Log in which possition a relay is being measured and whether it's an exit
- Configure longclaw's sbws with a 2nd Web server (destination) that is not a CDN
- Update the required disk space after #40017
- Add an script to unzip `datadir` files when upgrading sbws to a version including #40017
- Find out why some non-exits are being measured/used as exits and fix it
- When generating the bandwidth file, add the relays that are not in the consensus but are in the own last bandwidth file
- Stop trying to measure relays not in the consenus
- Fix Sphinx warnings when creating documentation
- Identify and reduce circuit failures with some relays
- Simplify building circuits
- Update bandwidth file version after fixing #40023
- Figure out why it seems it measures less relays over time
- Error decoding state.dat because is malformed
- Consider keeping data from the last 30 days instead of 5 last days
- Run tool to detect which relays are not scanned in a stable vm
- Bandwidth file spec v1.5 adds Tor version but sbws adds Tor version with bandwidth file spec as v1.4
- Update documentation on what/where to check regarding bandwidth values/files
- Avoid calculating mean or median with empty lists
- Linter error in release.py after merging #29294
- Work out what is wrong with the patch in #40006
- MR 19 breaks integration tests
- Integrate tests into chutney
- Create a .gitlab-ci.yml similar to .travis.yml to automatically test sbws with Gitlab runners
- Some relays do not get measured
- Update Debian package
- Create new release
- Create a page in the wiki with a timeline of bandwith authority events
- Change the references to Trac to Gitlab in the documentation
- Test net keys expired
- Maybe vote on the relays that have few or close measurements
- Check that relay_recent_measurement_attempt_count and relay_recent_priority_list_count are correc
- Compare sbws and Torflow
- Scale exactly as torflow does?
- For relays that change ip, only the measurements with the last ip are kept
- Relays without descriptors are not scaled, but still added to the bwlines without vote=0
- relay measures low / high on same network and same server
- Provide better logging for unit test runs in sbws
- sbws measures some relays 100x lower than Torflow
- Migrate to gitlab.tpo
- Fix typos in sbws comments/code
- Correct the relays to keep after retrieving new consensuses
- Document which branches developers should base their patches on
- Document that bwauths should checkout stable versions when installing sbws from git
- Is sbws weighting some relays too high?
- Fix lint error after merging #30733
- Check changes related to descriptors in a bandwidth file created by a bwauth before next release
- Test changes in descriptors
- Allow to connect to an external control port
- sbws stuck thinking a destination is dead
- sbws bandwidth scans should require a minimum exit bandwidth
- SBWS maatuska descriptor information appears stale
- sbws consensus timestamp updates incorrectly use the current time
- race condition: new_relays_dict is modified during iteration
- Maybe monitoring values in the state file should be reset when sbws is restarted
- Include the commit hash in the sbws version
- Unclear check for skipping scaling due to missing bandwidths or missing descriptors
- Work out which relays are ignored by all sbws instances
- sbws does not detect changes in descriptor bandwidth values
- Make sbws vote for all measured relays, even if they are not Running / not in the consensus
- Does "success" record relay measurements older than 5 days?
- Work out why relay_in_recent_consensus_count is 13 days for some relays
- Work out why 90% of sbws measurements fail
- Do not scale relays' bandwidth when there is not descriptor
- Document that there is now stable branch
- Minor fixes from observed behaviour in the public network
- Missing diagnostic keys in relay lines, but the data is in the header
- sbws should show relays for diagnostics, even when MIN_REPORT has not been reached
- Document the new bandwidth file keys added in sbws 1.1.0
- Create debian package for last 1.1.0 version
- Create 1.1.0 release
- Create a blog post about work done and future work
- Document next steps and/or tickets remaining
- Document that authorities are not measured
- sbws reports fewer relays than torflow
- Add KeyValues to monitor relays that are not measured
- Recover a destination when it stops failing
- Include scanner country and Web server country in the bandwidth file header
- Help 2-3 dirauths to deploy sbws
- Measure first relays with the lowest number of attempts first.
- Work out how long it takes sbws to measure the network
- sbws keeps the number of AsyncResults less than the number of threads
- sbws AsyncResults have no timeout
- Try enabling predicted circuits in sbws
- sbws stops accumulating, silently
- When sbws stops making progress, log a warning
- After several days, most of the circuits timeout
- Report relays that sbws wants to test, but the test doesn't work
- Report relays that sbws sees in the consensus, but never chooses to test
- Report excluded results in a relay's bandwidth line
- Work out how sbws can report excluded relays in the bandwidth file
- Monitor relays that are not measured by each sbws instance
- Disable measurement timing rules on sbws by default
- Monitor dirauth running sbws in production
Loading
Loading
Loading