Skip to content
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

Chiara/cherrypicking master #855

Merged
merged 13 commits into from
Mar 18, 2024
Merged

Chiara/cherrypicking master #855

merged 13 commits into from
Mar 18, 2024

Conversation

ch-iara
Copy link
Contributor

@ch-iara ch-iara commented Mar 15, 2024

PAYMENTS_ALTHEA test currently failing for tx parsing

jkilpatr and others added 11 commits March 13, 2024 11:04
This patch adds a simple parser for /proc/net/dev that allows the caller
to get the data usage counter for any interface. This is useful for
various applciations for example we could display usage per port on the
lan interfaces page.
We switched to Telnyx for SMS registration functions a while ago, so we
don't actually have api keys to run the previous code. Also the phone
number parsing logic has been refined to be slightly less complicated.
This patch removes cruft around startup args for Rita and modifies the
default configuration file behavior to generate the config if one is not
found.

Previously I didn't think generating a config was valuable as a feature.
But recently I've see strange behavior on some devices with limited
storage where the config was no longer present.
This is a real unforced error of a bug, obviously the char
representation of the octets can exceed u32 match because of the way it
is translated into more significant bits when parsed.

This is probably having a negative impact on exit enforcement in
production.
Previously it was possible for this to panic on some invalid utf8 cases.
This patch represents a complete overhaul of the automated rescue
systems in rita

1) everywhere where we used to try and crash actix to force a restart,
   we instead do a system reboot. Logs show these rescue functions are
   invoked pretty often, but killing the actix thread does not actually
   stop Rita like was originally hoped leaving routers in a bad state

2) The ping check that's used to reboot client routers when they can't
   reach the internet has been overhauld to ping a lot more destinations
   and combined with the hAP overload check

For (1) I've been looking for the reason routers would simply stop opening
tunnels and operating correctly. Reviewing this broken loop rescue
behavior seems like a smoking gun for that sort of bug.
These continue to be difficult to maintain, especially considering we
don't really use any mips target routers anymore as the processing power
is too low.
@jkilpatr jkilpatr merged commit 49654f1 into beta21 Mar 18, 2024
28 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants