-
Notifications
You must be signed in to change notification settings - Fork 12
Release v0.5.0 #135
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Release v0.5.0 #135
Conversation
Fix for br-lan iface.
Add more reflectors in APAC region
Closes Missing curl command when running setup #93
Review, re-organise, and re-write the documentation Co-authored-by: Daniel Lakeland <dlakelan@street-artists.org> Incorporating review changes and comments from - @Lochnair - @trippleflux - @richb-hanover
add contact details for developers
…torate into develop/main not sure why this is even necessary
modularisation - part 1 - move utility functions unrelated to SQM and CAKE to another module - set the correct path to pickup the new modules
modularisation phase 2 - settings Priority order of settings - OpenWRT uci /etc/config/sqm-autorate - command line - environment variables
several changes were made direct on the testing branch merge these back into develop/main so that they do not get lost
- further changes were performed direct on the release branch - merge to prevent further divergence
- uci: output.suppress_statistics - cli: --suppress-statistics - env: SQMA_SUPPRESS_STATISTICS
WAN interface takedown tx/rx stats file handling bug fix (issue #101, 105)
- incorporated most changes by @richb-hanover - check directly for CAKE being fully installed - do not abort on unexpected distro name - ask user
- to use a single '-', the option name must only be one character - align documentation with current code - resolve issues found during testing - update default configuration
re-organise setup script
Fail-Safe Fix for GH issue #99 per @jeverley. Closes #99. Closes Missing curl command when running setup #93 Removed bad reflectors. Per @jeverley's feedback, there is no need for a pattern match here. This appears to only be needed for ingress. Removing egress reference. Fixed incorrectly named variable. Attempt at br-lan fix. Daniel Lakeland add some lines and text to bandwidth diagram add an example run bw graph for use in readme Nils Andreas Svee set names on threads for debugging Add more reflectors in APAC region jeverley Incorporate changes discussed in 112 in clean rebase to develop/main Rich Brown Update getstats.sh Chas-IOT fix use of argparse and align documentation (#133) re-organise setup script - incorporated most changes by @richb-hanover - check directly for CAKE being fully installed - do not abort on unexpected distro name - ask user Fix #131 - crash when environment variable not set (#132) documentation updates fix formatting automate enable_verbose_baseline_output quieten statistics with new setting - uci: output.suppress_statistics - cli: --suppress-statistics - env: SQMA_SUPPRESS_STATISTICS quieten logging make setup and service ready for modularisation improved language and layout in julia/datavis.html (#72) Modularisation stage 2 - settings (#123) Modularisation stage 1 - requires and utilities (#122) prevent pollution from windows dev environment (#129) relicense all code to MPLv2 add contact details for developers edit and re-organise the README (#113) edit and re-organise the README (#113) Fix download locations for move to new organisation revision of the docs to align with settings
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Running for 13+ hours now without any runtime errors. I vote for "ship it!"
|
@chas-iot , @Fail-Safe @Lochnair @jeverley I'm delighted that it's working for you. But you're one of the developers :-) In #137, I'm advocating that we use as criteria:
This minimizes confusion for forum testers who try out the new version. (It would be really confusing to announce 0.5.0, and then have to release 0.5.1 right away because of a silly typo...) It's your call. I promise to install the new branch in the next couple hours and will report tomorrow morning, no matter what. Thanks. |
|
@richb-hanover we are learning and one of the things that we need to achieve is to document how a 'm.o.p.' can install from one of our testing or development branches so that we can improve our processes. Hence my question to you in the discussion thread. What I'm worried about for this release is less the functional running of the main sqm-autorate.lua and more the installation and initial startup ( now there's a hostage to fortune 🤣 ). The initial review with @Fail-Safe led to a significant change to the release notes to cover the various ways that sqm-autorate can run. |
OK. I believe it installs and runs OK. Other comments:
|
|
This release has had 24 hours of review, has run on at least 3 different routers, and at least 2 for more than 24 hours. While it is not perfect, I believe that it is an improvement. |
If you run sqm-autorate via
service sqm-autorate enable && service sqm-autorate startthen nothing changes, skip to Changes. If you use an alternate method and it fails, then see Starting sqm-autorateChanges
lua sqm-autorate.lua -hfor helplua sqm-autorate.lua -sto see the internal settingsStarting sqm-autorate
sqm-autorate is now split into several modules. This requires some changes to the way the program is started so that
luacan find and load the modules.service sqm-autorate startetc. then no change.lua /usr/lib/sqm-autorate/sqm-autorate.lua, then trycd /usr/lib/sqm-autorate/ ; lua sqm-autorate./usr/lib/sqm-autorate/sqm-autorate.luathen typecd /usr/lib/sqm-autorate/ ; ./sqm-autorate.luaLUA_PATHto add/usr/lib/sqm-autorate/?.luaCommits
Fail-Safe
Daniel Lakeland
Nils Andreas Svee
jeverley
Rich Brown
Chas-IOT