Files
zeekay b61441f7d7 Volume: node-local persistent volumes, under our own name
A fork of Rancher's local-path-provisioner at v0.0.36. The behaviour is
unchanged and deliberately so — a directory on the node a pod lands on is
the right thing for a disk. What changes is identity.

The provisioner answers to hanzo.ai/volume rather than rancher.io/local-path.
That is not branding: two provisioners sharing one name race for the same
claims, and a stock k3s carries the upstream one, so a cluster running both
would settle our storage by chance. Data lives at /var/lib/hanzo/volume, so
reading a Hanzo cluster does not mean learning another vendor's directory.

Apache-2.0 as published, unaltered, with Rancher's attribution in NOTICE.

The README says what this is not: durability. A volume here dies with its
node. That is hanzoai/replicate's question, one layer up, and keeping the
two apart is what lets the disk stay this simple.
2026-09-07 17:30:01 -07:00

104 lines
2.9 KiB
Go

//go:build !windows
// +build !windows
package dns
import (
"net"
"golang.org/x/net/ipv4"
"golang.org/x/net/ipv6"
)
// This is the required size of the OOB buffer to pass to ReadMsgUDP.
var udpOOBSize = func() int {
// We can't know whether we'll get an IPv4 control message or an
// IPv6 control message ahead of time. To get around this, we size
// the buffer equal to the largest of the two.
oob4 := ipv4.NewControlMessage(ipv4.FlagDst | ipv4.FlagInterface)
oob6 := ipv6.NewControlMessage(ipv6.FlagDst | ipv6.FlagInterface)
if len(oob4) > len(oob6) {
return len(oob4)
}
return len(oob6)
}()
// SessionUDP holds the remote address and the associated
// out-of-band data.
type SessionUDP struct {
raddr *net.UDPAddr
context []byte
}
// RemoteAddr returns the remote network address.
func (s *SessionUDP) RemoteAddr() net.Addr { return s.raddr }
// ReadFromSessionUDP acts just like net.UDPConn.ReadFrom(), but returns a session object instead of a
// net.UDPAddr.
func ReadFromSessionUDP(conn *net.UDPConn, b []byte) (int, *SessionUDP, error) {
oob := make([]byte, udpOOBSize)
n, oobn, _, raddr, err := conn.ReadMsgUDP(b, oob)
if err != nil {
return n, nil, err
}
return n, &SessionUDP{raddr, oob[:oobn]}, err
}
// WriteToSessionUDP acts just like net.UDPConn.WriteTo(), but uses a *SessionUDP instead of a net.Addr.
func WriteToSessionUDP(conn *net.UDPConn, b []byte, session *SessionUDP) (int, error) {
oob := correctSource(session.context)
n, _, err := conn.WriteMsgUDP(b, oob, session.raddr)
return n, err
}
func setUDPSocketOptions(conn *net.UDPConn) error {
// Try setting the flags for both families and ignore the errors unless they
// both error.
err6 := ipv6.NewPacketConn(conn).SetControlMessage(ipv6.FlagDst|ipv6.FlagInterface, true)
err4 := ipv4.NewPacketConn(conn).SetControlMessage(ipv4.FlagDst|ipv4.FlagInterface, true)
if err6 != nil && err4 != nil {
return err4
}
return nil
}
// parseDstFromOOB takes oob data and returns the destination IP.
func parseDstFromOOB(oob []byte) net.IP {
// Start with IPv6 and then fallback to IPv4
// TODO(fastest963): Figure out a way to prefer one or the other. Looking at
// the lvl of the header for a 0 or 41 isn't cross-platform.
cm6 := new(ipv6.ControlMessage)
if cm6.Parse(oob) == nil && cm6.Dst != nil {
return cm6.Dst
}
cm4 := new(ipv4.ControlMessage)
if cm4.Parse(oob) == nil && cm4.Dst != nil {
return cm4.Dst
}
return nil
}
// correctSource takes oob data and returns new oob data with the Src equal to the Dst
func correctSource(oob []byte) []byte {
dst := parseDstFromOOB(oob)
if dst == nil {
return nil
}
// If the dst is definitely an IPv6, then use ipv6's ControlMessage to
// respond otherwise use ipv4's because ipv6's marshal ignores ipv4
// addresses.
if dst.To4() == nil {
cm := new(ipv6.ControlMessage)
cm.Src = dst
oob = cm.Marshal()
} else {
cm := new(ipv4.ControlMessage)
cm.Src = dst
oob = cm.Marshal()
}
return oob
}