moby/libnetwork/ns/init_linux.go
Cory Snider 7fc29c1435 libnetwork/osl: clean up Linux InvokeFunc()
Aside from unconditionally unlocking the OS thread even if restoring the
thread's network namespace fails, func (*networkNamespace).InvokeFunc()
correctly implements invoking a function inside a network namespace.
This is far from obvious, however. func InitOSContext() does much of the
heavy lifting but in a bizarre fashion: it restores the initial network
namespace before it is changed in the first place, and the cleanup
function it returns does not restore the network namespace at all! The
InvokeFunc() implementation has to restore the network namespace
explicitly by deferring a call to ns.SetNamespace().

func InitOSContext() is a leaky abstraction taped to a footgun. On the
one hand, it defensively resets the current thread's network namespace,
which has the potential to fix up the thread state if other buggy code
had failed to maintain the invariant that an OS thread must be locked to
a goroutine unless it is interchangeable with a "clean" thread as
spawned by the Go runtime. On the other hand, it _facilitates_ writing
buggy code which fails to maintain the aforementioned invariant because
the cleanup function it returns unlocks the thread from the goroutine
unconditionally while neglecting to restore the thread's network
namespace! It is quite scary to need a function which fixes up threads'
network namespaces after the fact as an arbitrary number of goroutines
could have been scheduled onto a "dirty" thread and run non-libnetwork
code before the thread's namespace is fixed up. Any number of
(not-so-)subtle misbehaviours could result if an unfortunate goroutine
is scheduled onto a "dirty" thread. The whole repository has been
audited to ensure that the aforementioned invariant is never violated,
making after-the-fact fixing up of thread network namespaces redundant.
Make InitOSContext() a no-op on Linux and inline the thread-locking into
the function (singular) which previously relied on it to do so.

func ns.SetNamespace() is of similarly dubious utility. It intermixes
capturing the initial network namespace and restoring the thread's
network namespace, which could result in threads getting put into the
wrong network namespace if the wrong thread is the first to call it.
Delete it entirely; functions which need to manipulate a thread's
network namespace are better served by being explicit about capturing
and restoring the thread's namespace.

Rewrite InvokeFunc() to invoke the closure inside a goroutine to enable
a graceful and safe recovery if the thread's network namespace could not
be restored. Avoid any potential race conditions due to changing the
main thread's network namespace by preventing the aforementioned
goroutines from being eligible to be scheduled onto the main thread.

Signed-off-by: Cory Snider <csnider@mirantis.com>
2022-10-25 13:35:44 -04:00

108 lines
2.8 KiB
Go

package ns
import (
"fmt"
"os/exec"
"strings"
"sync"
"syscall"
"time"
"github.com/sirupsen/logrus"
"github.com/vishvananda/netlink"
"github.com/vishvananda/netns"
)
var (
initNs netns.NsHandle
initNl *netlink.Handle
initOnce sync.Once
// NetlinkSocketsTimeout represents the default timeout duration for the sockets
NetlinkSocketsTimeout = 3 * time.Second
)
// Init initializes a new network namespace
func Init() {
var err error
initNs, err = netns.Get()
if err != nil {
logrus.Errorf("could not get initial namespace: %v", err)
}
initNl, err = netlink.NewHandle(getSupportedNlFamilies()...)
if err != nil {
logrus.Errorf("could not create netlink handle on initial namespace: %v", err)
}
err = initNl.SetSocketTimeout(NetlinkSocketsTimeout)
if err != nil {
logrus.Warnf("Failed to set the timeout on the default netlink handle sockets: %v", err)
}
}
// ParseHandlerInt transforms the namespace handler into an integer
func ParseHandlerInt() int {
return int(getHandler())
}
// GetHandler returns the namespace handler
func getHandler() netns.NsHandle {
initOnce.Do(Init)
return initNs
}
// NlHandle returns the netlink handler
func NlHandle() *netlink.Handle {
initOnce.Do(Init)
return initNl
}
func getSupportedNlFamilies() []int {
fams := []int{syscall.NETLINK_ROUTE}
// NETLINK_XFRM test
if err := checkXfrmSocket(); err != nil {
logrus.Warnf("Could not load necessary modules for IPSEC rules: %v", err)
} else {
fams = append(fams, syscall.NETLINK_XFRM)
}
// NETLINK_NETFILTER test
if err := loadNfConntrackModules(); err != nil {
if checkNfSocket() != nil {
logrus.Warnf("Could not load necessary modules for Conntrack: %v", err)
} else {
fams = append(fams, syscall.NETLINK_NETFILTER)
}
} else {
fams = append(fams, syscall.NETLINK_NETFILTER)
}
return fams
}
// API check on required xfrm modules (xfrm_user, xfrm_algo)
func checkXfrmSocket() error {
fd, err := syscall.Socket(syscall.AF_NETLINK, syscall.SOCK_RAW, syscall.NETLINK_XFRM)
if err != nil {
return err
}
syscall.Close(fd)
return nil
}
func loadNfConntrackModules() error {
if out, err := exec.Command("modprobe", "-va", "nf_conntrack").CombinedOutput(); err != nil {
return fmt.Errorf("Running modprobe nf_conntrack failed with message: `%s`, error: %v", strings.TrimSpace(string(out)), err)
}
if out, err := exec.Command("modprobe", "-va", "nf_conntrack_netlink").CombinedOutput(); err != nil {
return fmt.Errorf("Running modprobe nf_conntrack_netlink failed with message: `%s`, error: %v", strings.TrimSpace(string(out)), err)
}
return nil
}
// API check on required nf_conntrack* modules (nf_conntrack, nf_conntrack_netlink)
func checkNfSocket() error {
fd, err := syscall.Socket(syscall.AF_NETLINK, syscall.SOCK_RAW, syscall.NETLINK_NETFILTER)
if err != nil {
return err
}
syscall.Close(fd)
return nil
}