mirror of
https://github.com/minio/minio.git
synced 2024-12-26 23:25:54 -05:00
91805bcab6
allow non-inlined on disk to be inlined via an unversioned ReadVersion() call, we only need ReadXL() to resolve objects with multiple versions only. The choice of this block makes it to be dynamic and chosen by the user via `mc admin config set` Other bonus things - Start measuring internode TTFB performance. - Set TCP_NODELAY, TCP_CORK for low latency
135 lines
5.0 KiB
Go
135 lines
5.0 KiB
Go
//go:build linux
|
|
// +build linux
|
|
|
|
// Copyright (c) 2015-2023 MinIO, Inc.
|
|
//
|
|
// This file is part of MinIO Object Storage stack
|
|
//
|
|
// This program is free software: you can redistribute it and/or modify
|
|
// it under the terms of the GNU Affero General Public License as published by
|
|
// the Free Software Foundation, either version 3 of the License, or
|
|
// (at your option) any later version.
|
|
//
|
|
// This program is distributed in the hope that it will be useful
|
|
// but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
// GNU Affero General Public License for more details.
|
|
//
|
|
// You should have received a copy of the GNU Affero General Public License
|
|
// along with this program. If not, see <http://www.gnu.org/licenses/>.
|
|
|
|
package http
|
|
|
|
import (
|
|
"context"
|
|
"net"
|
|
"syscall"
|
|
"time"
|
|
|
|
"github.com/minio/minio/internal/deadlineconn"
|
|
"golang.org/x/sys/unix"
|
|
)
|
|
|
|
func setTCPParametersFn(opts TCPOptions) func(network, address string, c syscall.RawConn) error {
|
|
return func(network, address string, c syscall.RawConn) error {
|
|
c.Control(func(fdPtr uintptr) {
|
|
// got socket file descriptor to set parameters.
|
|
fd := int(fdPtr)
|
|
|
|
_ = unix.SetsockoptInt(fd, unix.SOL_SOCKET, unix.SO_REUSEADDR, 1)
|
|
|
|
_ = unix.SetsockoptInt(fd, unix.SOL_SOCKET, unix.SO_REUSEPORT, 1)
|
|
|
|
// Enable custom socket send/recv buffers.
|
|
if opts.SendBufSize > 0 {
|
|
_ = unix.SetsockoptInt(fd, unix.SOL_SOCKET, unix.SO_SNDBUF, opts.SendBufSize)
|
|
}
|
|
|
|
if opts.RecvBufSize > 0 {
|
|
_ = unix.SetsockoptInt(fd, unix.SOL_SOCKET, unix.SO_RCVBUF, opts.RecvBufSize)
|
|
}
|
|
|
|
if opts.NoDelay {
|
|
_ = syscall.SetsockoptInt(fd, syscall.IPPROTO_TCP, unix.TCP_NODELAY, 1)
|
|
_ = syscall.SetsockoptInt(fd, syscall.SOL_TCP, unix.TCP_CORK, 0)
|
|
}
|
|
|
|
// Enable TCP open
|
|
// https://lwn.net/Articles/508865/ - 32k queue size.
|
|
_ = syscall.SetsockoptInt(fd, syscall.SOL_TCP, unix.TCP_FASTOPEN, 32*1024)
|
|
|
|
// Enable TCP fast connect
|
|
// TCPFastOpenConnect sets the underlying socket to use
|
|
// the TCP fast open connect. This feature is supported
|
|
// since Linux 4.11.
|
|
_ = syscall.SetsockoptInt(fd, syscall.IPPROTO_TCP, unix.TCP_FASTOPEN_CONNECT, 1)
|
|
|
|
// Enable TCP quick ACK, John Nagle says
|
|
// "Set TCP_QUICKACK. If you find a case where that makes things worse, let me know."
|
|
_ = syscall.SetsockoptInt(fd, syscall.IPPROTO_TCP, unix.TCP_QUICKACK, 1)
|
|
|
|
/// Enable keep-alive
|
|
{
|
|
_ = unix.SetsockoptInt(fd, unix.SOL_SOCKET, unix.SO_KEEPALIVE, 1)
|
|
|
|
// The time (in seconds) the connection needs to remain idle before
|
|
// TCP starts sending keepalive probes
|
|
_ = syscall.SetsockoptInt(fd, syscall.IPPROTO_TCP, syscall.TCP_KEEPIDLE, 15)
|
|
|
|
// Number of probes.
|
|
// ~ cat /proc/sys/net/ipv4/tcp_keepalive_probes (defaults to 9, we reduce it to 5)
|
|
_ = syscall.SetsockoptInt(fd, syscall.IPPROTO_TCP, syscall.TCP_KEEPCNT, 5)
|
|
|
|
// Wait time after successful probe in seconds.
|
|
// ~ cat /proc/sys/net/ipv4/tcp_keepalive_intvl (defaults to 75 secs, we reduce it to 15 secs)
|
|
_ = syscall.SetsockoptInt(fd, syscall.IPPROTO_TCP, syscall.TCP_KEEPINTVL, 15)
|
|
}
|
|
|
|
// Set tcp user timeout in addition to the keep-alive - tcp-keepalive is not enough to close a socket
|
|
// with dead end because tcp-keepalive is not fired when there is data in the socket buffer.
|
|
// https://blog.cloudflare.com/when-tcp-sockets-refuse-to-die/
|
|
// This is a sensitive configuration, it is better to set it to high values, > 60 secs since it can
|
|
// affect clients reading data with a very slow pace (disappropriate with socket buffer sizes)
|
|
if opts.UserTimeout > 0 {
|
|
_ = syscall.SetsockoptInt(fd, syscall.IPPROTO_TCP, unix.TCP_USER_TIMEOUT, opts.UserTimeout)
|
|
}
|
|
|
|
if opts.Interface != "" {
|
|
if h, _, err := net.SplitHostPort(address); err == nil {
|
|
address = h
|
|
}
|
|
// Create socket on specific vrf device.
|
|
// To catch all kinds of special cases this filters specifically for loopback networks.
|
|
if ip := net.ParseIP(address); ip != nil && !ip.IsLoopback() {
|
|
_ = syscall.SetsockoptString(fd, syscall.SOL_SOCKET, syscall.SO_BINDTODEVICE, opts.Interface)
|
|
}
|
|
}
|
|
})
|
|
return nil
|
|
}
|
|
}
|
|
|
|
// DialContext is a function to make custom Dial for internode communications
|
|
type DialContext func(ctx context.Context, network, address string) (net.Conn, error)
|
|
|
|
// NewInternodeDialContext setups a custom dialer for internode communication
|
|
func NewInternodeDialContext(dialTimeout time.Duration, opts TCPOptions) DialContext {
|
|
return func(ctx context.Context, network, addr string) (net.Conn, error) {
|
|
dialer := &net.Dialer{
|
|
Timeout: dialTimeout,
|
|
Control: setTCPParametersFn(opts),
|
|
}
|
|
conn, err := dialer.DialContext(ctx, network, addr)
|
|
if err != nil {
|
|
return nil, err
|
|
}
|
|
if opts.DriveOPTimeout != nil {
|
|
// Read deadlines are sufficient for now as per various
|
|
// scenarios of hung node detection, we may add Write deadlines
|
|
// if needed later on.
|
|
return deadlineconn.New(conn).WithReadDeadline(opts.DriveOPTimeout()), nil
|
|
}
|
|
return conn, nil
|
|
}
|
|
}
|