atproto relay implementation in zig
zlay.waow.tech
fix: add read timeout to metrics server to prevent connection stalls
the single-threaded metrics server on :3001 blocks on accept/read with
no timeout. if any client connects but doesn't complete the request
(e.g. during startup storm), the thread blocks forever and all
subsequent scrapes fail. sets SO_RCVTIMEO=5s on accepted connections.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
+3
+3
src/main.zig
+3
src/main.zig
···
59
59
log.debug("metrics accept error: {s}", .{@errorName(err)});
60
60
continue;
61
61
};
62
+
// 5s read timeout — prevents stale connections from blocking the single-threaded server
63
+
const timeout = std.posix.timeval{ .sec = 5, .usec = 0 };
64
+
std.posix.setsockopt(conn.stream.handle, std.posix.SOL.SOCKET, std.posix.SO.RCVTIMEO, std.mem.asBytes(&timeout)) catch {};
62
65
handleMetricsConn(conn.stream, self.stats, self.validator, self.data_dir, self.persist, self.bc);
63
66
}
64
67
}