Fixes #1852 - Unmap addresses from routers that become unreachable in topology changes.

[ted-ross]

NOTE: This is a draft PR. I'm putting this here as a suggested way to solve the problem in the raised issue.

This needs more testing and examination before it can be accepted.

[github-advanced-security]

This pull request sets up GitHub code scanning for this repository. Once the scans have completed and the checks have passed, the analysis results for this pull request branch will appear on this overview. Once you merge this pull request, the 'Security' tab will show more code scanning analysis results (for example, for the default branch). Depending on your configuration and choice of analysis tool, future pull requests will be annotated with code scanning analysis results. For more information about GitHub code scanning, check out the documentation.

[gabordozsa]

@ted-ross Would it make sense to unmap the addresses in two phases in order to give a chance to a lost link to recover? I.e. mark the unreachable nodes first and call unmap a few ticks later if they are still unreachable. I'm thinking of a router losing the access to a large number of other routers due to a peer is being restarted.

[ted-ross]

@ted-ross Would it make sense to unmap the addresses in two phases in order to give a chance to a lost link to recover? I.e. mark the unreachable nodes first and call unmap a few ticks later if they are still unreachable. I'm thinking of a router losing the access to a large number of other routers due to a peer is being restarted.

This is a good suggestion. My main concern with this PR is that it might cause a lot of churn if there was a momentary loss of connection. This is why the original functionality was in place, but 60 seconds is arguably much too long of an interval to allow disconnects to get fixed.

[ted-ross]

Further to @gabordozsa 's suggestion...

We could add a new configuration interval called removeUnreachableHoldoffSeconds (or similar) to indicate how many seconds the router should wait after declaring a peer-router "unreachable" until its addresses are flushed. A good default might be 1 or 2 seconds, but it could be increased for larger networks or networks with many addresses.

[ted-ross]

This PR needs a test to validate the new functionality.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 67.2%. Comparing base (60cd6ec) to head (79b7c2f).
⚠️ Report is 10 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##            main   #1854     +/-   ##
=======================================
- Coverage   70.5%   67.2%   -3.4%     
=======================================
  Files        209     204      -5     
  Lines      47838   47716    -122     
  Branches    5156    5326    +170     
=======================================
- Hits       33765   32081   -1684     
- Misses     11608   13405   +1797     
+ Partials    2465    2230    -235     

Flag	Coverage Δ
pysystemtests	77.3% <ø> (+0.9%)	⬆️
systemtests	61.9% <ø> (-5.6%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
unittests	67.2% <84.5%> (-3.4%)	⬇️
systemtests	67.2% <84.5%> (-3.4%)	⬇️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[ted-ross]

The changes to the PR are very good. I inserted a couple of questions for discussion before we approve this.