fix(ping): Fix panic in WASM caused by retrying on dial upgrade errors

Cargo.toml

@@ -93,7 +93,7 @@ libp2p-mplex = { version = "0.41.0", path = "muxers/mplex" }
 libp2p-muxer-test-harness = { path = "muxers/test-harness" }
 libp2p-noise = { version = "0.44.0", path = "transports/noise" }
 libp2p-perf = { version = "0.3.1", path = "protocols/perf" }
-libp2p-ping = { version = "0.44.1", path = "protocols/ping" }
+libp2p-ping = { version = "0.44.2", path = "protocols/ping" }
 libp2p-plaintext = { version = "0.41.0", path = "transports/plaintext" }
 libp2p-pnet = { version = "0.24.0", path = "transports/pnet" }
 libp2p-quic = { version = "0.10.3", path = "transports/quic" }

protocols/ping/CHANGELOG.md

@@ -1,12 +1,17 @@
-## 0.44.1 - unreleased
+## 0.44.2
+
+- Use `web-time` instead of `instant`.
+  See [PR 5347](https://github.com/libp2p/rust-libp2p/pull/5347).
+
+- Fixes panic in WASM caused by retrying on dail errors.
+  See [PR 5447](https://github.com/libp2p/rust-libp2p/pull/5447).
+
+## 0.44.1
 
 - Impose `Sync` on `ping::Failure::Other`.
   `ping::Event` can now be shared between threads.
   See [PR 5250]
 
-- Use `web-time` instead of `instant`.
-  See [PR 5347](https://github.com/libp2p/rust-libp2p/pull/5347).
-
 [PR 5250]: https://github.com/libp2p/rust-libp2p/pull/5250
 
 ## 0.44.0

protocols/ping/Cargo.toml

@@ -3,7 +3,7 @@ name = "libp2p-ping"
 edition = "2021"
 rust-version = { workspace = true }
 description = "Ping protocol for libp2p"
-version = "0.44.1"
+version = "0.44.2"
 authors = ["Parity Technologies <[email protected]>"]
 license = "MIT"
 repository = "https://github.com/libp2p/rust-libp2p"

protocols/ping/src/handler.rs

@@ -184,6 +184,18 @@ impl Handler {
     ) {
         self.outbound = None; // Request a new substream on the next `poll`.
 
+        // Timer is already polled and expired before substream request is initiated
+        // and will be polled again later on in our `poll` because we reset `self.outbound`.
+        //
+        // `futures-timer` allows an expired timer to be polled again and returns
+        // immidietly a `Poll::Ready`. However in its WASM implementation there is
+        // a bug that causes the expired timer to panic.
+        //
+        // This is a workaround until a proper fix is merged and released.
+        //
+        // See async-rs/futures-timer#74 and libp2p/rust-libp2p#5447 for more info.
+        self.interval.reset(Duration::new(0, 0));
+
         let error = match error {
             StreamUpgradeError::NegotiationFailed => {
                 debug_assert_eq!(self.state, State::Active);