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

feat(errors): add Error::is_shutdown #2745

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 5 additions & 2 deletions src/error.rs
Original file line number Diff line number Diff line change
@@ -56,7 +56,6 @@ pub(super) enum Kind {
/// The body write was aborted.
BodyWriteAborted,
/// Error calling AsyncWrite::shutdown()
#[cfg(feature = "http1")]
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why change the feature-gate here? Would it be better to clarify the helper?

Shutdown,

/// A general error from h2.
@@ -201,6 +200,11 @@ impl Error {
self.find_source::<TimedOut>().is_some()
}

/// Returns true if the error was caused by a shutdown.
pub fn is_shutdown(&self) -> bool {
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

e.g.

Suggested change
pub fn is_shutdown(&self) -> bool {
#[cfg(feature = "http1")]
pub fn is_http1_shutdown(&self) -> bool {

Either that, or this should probably also handle h2 goaways?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hm, that does make me think the name could confuse people. It's not a graceful shutdown, it was an error that occurred when calling AsyncWrite::shutdown.

It originally was just meant to provide additional context on top of the IO error. Is that context useful on its own?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't understand what you mean. There is nothing we can do about those errors in the wild (most often they are a "broken pipe" error), so we don't want to log them at all, so we need to discriminate them to ignore them.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So, it does seem like the context is useful to you. Now the question is, is the name confusing, such that someone might think it's a "graceful shutdown" error received over HTTP/2? Is that a bad thing? Should we hijack the name to make both cases be a is_shutdown error?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

h2 errors currently are all represented as Kind::H2 right? Maybe I should rename that method to is_http1_shutdown then.

matches!(self.inner.kind, Kind::Shutdown)
}

/// Consumes the error, returning its cause.
pub fn into_cause(self) -> Option<Box<dyn StdError + Send + Sync>> {
self.inner.cause
@@ -440,7 +444,6 @@ impl Error {
#[cfg(any(feature = "http1", feature = "http2"))]
Kind::BodyWrite => "error writing a body to connection",
Kind::BodyWriteAborted => "body write aborted",
#[cfg(feature = "http1")]
Kind::Shutdown => "error shutting down connection",
#[cfg(feature = "http2")]
Kind::Http2 => "http2 error",