-
-
Notifications
You must be signed in to change notification settings - Fork 3k
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
feat: transport config #7479
feat: transport config #7479
Conversation
41c19b9
to
8c6c312
Compare
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.
A few typos, and added some questions/comments, but looks good. ❤️ docs
test/sharness/t0125-twonode.sh
Outdated
iptb run -- ipfs config --json Swarm.Transports.Security TLS false && | ||
iptb run -- ipfs config --json Swarm.Transports.Security Secio false |
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.
Any recommendations on how to keep the sharness tests up to date as the default configs change? Is it just grepping through the sharness tests for the configuration options related to the ones you're changing (e.g. Swarm.Transports.Security if any of the security transports change)?
I suspect we're not going to be adding new security transports anytime soon and so this is mostly hypothetical, but the prior "override" semantics made it easier to keep these tests correct when new security protocols were added. IMO the new one (this PR) is much nicer to interact with in code and in the config file editing though.
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.
This is a problem, but I think we can revisit it later. We may want a second DisableDefaults
flag.
Ok, so I've "fixed" the config commands to mostly do what we need. They're still buggy but that's an issue for a future release. |
Digging into the type checking, it looks like it was a failed attempt to auto-decode the input into the correct type (e.g., detect if something should be a bool, etc.). However:
|
13291a9
to
8644485
Compare
This way, users can disable transports (especially QUIC), and set muxer/security transport priorities.
This bug was pointed out in #1440 (comment) but was ignored for some reason.
8644485
to
3ce725d
Compare
Add a config section for configuring transports. This way, users can disable transports (especially QUIC), and set muxer/security transport priorities.