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.

Sign up for GitHub

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

consensus: fix AddrInUse error that can occur during simtest crash/re… #20818

Merged
merged 1 commit into from
Jan 8, 2025
Merged

consensus: fix AddrInUse error that can occur during simtest crash/re… #20818

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
44 changes: 31 additions & 13 deletions consensus/core/src/network/tonic_network.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,7 +6,7 @@ use std::{
net::{SocketAddr, SocketAddrV4, SocketAddrV6},
pin::Pin,
sync::Arc,
time::Duration,
time::{Duration, Instant},
};

use async_trait::async_trait;
Expand Down Expand Up @@ -775,18 +775,36 @@ impl<S: NetworkService> NetworkManager<S> for TonicManager {
}
}

let server = sui_http::Builder::new()
.config(
sui_http::Config::default()
.initial_connection_window_size(64 << 20)
.initial_stream_window_size(32 << 20)
.http2_keepalive_interval(Some(config.keepalive_interval))
.http2_keepalive_timeout(Some(config.keepalive_interval))
.accept_http1(false),
)
.tls_config(tls_server_config)
.serve(own_address, consensus_service)
.unwrap();
let http_config = sui_http::Config::default()
.initial_connection_window_size(64 << 20)
.initial_stream_window_size(32 << 20)
.http2_keepalive_interval(Some(config.keepalive_interval))
.http2_keepalive_timeout(Some(config.keepalive_interval))
.accept_http1(false);

// Create server
//
// During simtest crash/restart tests there may be an older instance of consensus running
// that is bound to the TCP port of `own_address` that hasn't finished relinquishing
// control of the port yet. So instead of crashing when the address is inuse, we will retry
// for a short/reasonable period of time before giving up.
let deadline = Instant::now() + Duration::from_secs(20);
let server = loop {
match sui_http::Builder::new()
.config(http_config.clone())
.tls_config(tls_server_config.clone())
.serve(own_address, consensus_service.clone())
{
Ok(server) => break server,
Err(err) => {
warn!("Error starting consensus server: {err:?}");
if Instant::now() > deadline {
panic!("Failed to start consensus server within required deadline");
}
tokio::time::sleep(Duration::from_secs(1)).await;
}
}
};

info!("Server started at: {own_address}");
self.server = Some(server);
Expand Down
1 change: 1 addition & 0 deletions crates/sui-http/src/config.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,6 +5,7 @@ use std::time::Duration;

const DEFAULT_HTTP2_KEEPALIVE_TIMEOUT_SECS: u64 = 20;

#[derive(Debug, Clone)]
pub struct Config {
init_stream_window_size: Option<u32>,
init_connection_window_size: Option<u32>,
Expand Down
Loading