ipfs-cluster/api/rest/restapi.go
Hector Sanjuan 33d9cdd3c4 Feat: emancipate Consensus from the Cluster component
This commit promotes the Consensus component (and Raft) to become a fully
independent thing like other components, passed to NewCluster during
initialization. Cluster (main component) no longer creates the consensus
layer internally. This has triggered a number of breaking changes
that I will explain below.

Motivation: Future work will require the possibility of running Cluster
with a consensus layer that is not Raft. The "consensus" layer is in charge
of maintaining two things:
  * The current cluster peerset, as required by the implementation
  * The current cluster pinset (shared state)

While the pinset maintenance has always been in the consensus layer, the
peerset maintenance was handled by the main component (starting by the "peers"
key in the configuration) AND the Raft component (internally)
and this generated lots of confusion: if the user edited the peers in the
configuration they would be greeted with an error.

The bootstrap process (adding a peer to an existing cluster) and configuration
key also complicated many things, since the main component did it, but only
when the consensus was initialized and in single peer mode.

In all this we also mixed the peerstore (list of peer addresses in the libp2p
host) with the peerset, when they need not to be linked.

By initializing the consensus layer before calling NewCluster, all the
difficulties in maintaining the current implementation in the same way
have come to light. Thus, the following changes have been introduced:

* Remove "peers" and "bootstrap" keys from the configuration: we no longer
edit or save the configuration files. This was a very bad practice, requiring
write permissions by the process to the file containing the private key and
additionally made things like Puppet deployments of cluster difficult as
configuration would mutate from its initial version. Needless to say all the
maintenance associated to making sure peers and bootstrap had correct values
when peers are bootstrapped or removed. A loud and detailed error message has
been added when staring cluster with an old config, along with instructions on
how to move forward.

* Introduce a PeerstoreFile ("peerstore") which stores peer addresses: in
ipfs, the peerstore is not persisted because it can be re-built from the
network bootstrappers and the DHT. Cluster should probably also allow
discoverability of peers addresses (when not bootstrapping, as in that case
we have it), but in the meantime, we will read and persist the peerstore
addresses for cluster peers in this file, different from the configuration.
Note that dns multiaddresses are now fully supported and no IPs are saved
when we have DNS multiaddresses for a peer.

* The former "peer_manager" code is now a pstoremgr module, providing utilities
to parse, add, list and generally maintain the libp2p host peerstore, including
operations on the PeerstoreFile. This "pstoremgr" can now also be extended to
perform address autodiscovery and other things indepedently from Cluster.

* Create and initialize Raft outside of the main Cluster component: since we
can now launch Raft independently from Cluster, we have more degrees of
freedom. A new "staging" option when creating the object allows a raft peer to
be launched in Staging mode, waiting to be added to a running consensus, and
thus, not electing itself as leader or doing anything like we were doing
before. This additionally allows us to track when the peer has become a
Voter, which only happens when it's caught up with the state, something that
was wonky previously.

* The raft configuration now includes an InitPeerset key, which allows to
provide a peerset for new peers and which is ignored when staging==true. The
whole Raft initialization code is way cleaner and stronger now.

* Cluster peer bootsrapping is now an ipfs-cluster-service feature. The
--bootstrap flag works as before (additionally allowing comma-separated-list
of entries). What bootstrap does, is to initialize Raft with staging == true,
and then call Join in the main cluster component. Only when the Raft peer
transitions to Voter, consensus becomes ready, and cluster becomes Ready.
This is cleaner, works better and is less complex than before (supporting
both flags and config values). We also backup and clean the state whenever
we are boostrapping, automatically

* ipfs-cluster-service no longer runs the daemon. Starting cluster needs
now "ipfs-cluster-service daemon". The daemon specific flags (bootstrap,
alloc) are now flags for the daemon subcommand. Here we mimic ipfs ("ipfs"
does not start the daemon but print help) and pave the path for merging both
service and ctl in the future.

While this brings some breaking changes, it significantly reduces the
complexity of the configuration, the code and most importantly, the
documentation. It should be easier now to explain the user what is the
right way to launch a cluster peer, and more difficult to make mistakes.

As a side effect, the PR also:

* Fixes #381 - peers with dynamic addresses
* Fixes #371 - peers should be Raft configuration option
* Fixes #378 - waitForUpdates may return before state fully synced
* Fixes #235 - config option shadowing (no cfg saves, no need to shadow)

License: MIT
Signed-off-by: Hector Sanjuan <code@hector.link>
2018-05-07 07:39:41 +02:00

833 lines
18 KiB
Go

// Package rest implements an IPFS Cluster API component. It provides
// a REST-ish API to interact with Cluster.
//
// rest exposes the HTTP API in two ways. The first is through a regular
// HTTP(s) listener. The second is by tunneling HTTP through a libp2p
// stream (thus getting an encrypted channel without the need to setup
// TLS). Both ways can be used at the same time, or disabled.
package rest
import (
"context"
"crypto/tls"
"encoding/json"
"errors"
"fmt"
"net"
"net/http"
"strconv"
"strings"
"sync"
types "github.com/ipfs/ipfs-cluster/api"
mux "github.com/gorilla/mux"
rpc "github.com/hsanjuan/go-libp2p-gorpc"
gostream "github.com/hsanjuan/go-libp2p-gostream"
p2phttp "github.com/hsanjuan/go-libp2p-http"
cid "github.com/ipfs/go-cid"
logging "github.com/ipfs/go-log"
libp2p "github.com/libp2p/go-libp2p"
host "github.com/libp2p/go-libp2p-host"
peer "github.com/libp2p/go-libp2p-peer"
ma "github.com/multiformats/go-multiaddr"
manet "github.com/multiformats/go-multiaddr-net"
)
var logger = logging.Logger("restapi")
// Common errors
var (
// ErrNoEndpointEnabled is returned when the API is created but
// no HTTPListenAddr, nor libp2p configuration fields, nor a libp2p
// Host are provided.
ErrNoEndpointsEnabled = errors.New("neither the libp2p nor the HTTP endpoints are enabled")
// ErrHTTPEndpointNotEnabled is returned when trying to perform
// operations that rely on the HTTPEndpoint but it is disabled.
ErrHTTPEndpointNotEnabled = errors.New("the HTTP endpoint is not enabled")
)
// API implements an API and aims to provides
// a RESTful HTTP API for Cluster.
type API struct {
ctx context.Context
cancel func()
config *Config
rpcClient *rpc.Client
rpcReady chan struct{}
router *mux.Router
server *http.Server
host host.Host
httpListener net.Listener
libp2pListener net.Listener
shutdownLock sync.Mutex
shutdown bool
wg sync.WaitGroup
}
type route struct {
Name string
Method string
Pattern string
HandlerFunc http.HandlerFunc
}
type peerAddBody struct {
PeerMultiaddr string `json:"peer_multiaddress"`
}
// NewAPI creates a new REST API component with the given configuration.
func NewAPI(cfg *Config) (*API, error) {
return NewAPIWithHost(cfg, nil)
}
// NewAPIWithHost creates a new REST API component and enables
// the libp2p-http endpoint using the given Host, if not nil.
func NewAPIWithHost(cfg *Config, h host.Host) (*API, error) {
err := cfg.Validate()
if err != nil {
return nil, err
}
router := mux.NewRouter().StrictSlash(true)
s := &http.Server{
ReadTimeout: cfg.ReadTimeout,
ReadHeaderTimeout: cfg.ReadHeaderTimeout,
WriteTimeout: cfg.WriteTimeout,
IdleTimeout: cfg.IdleTimeout,
Handler: router,
}
s.SetKeepAlivesEnabled(true) // A reminder that this can be changed
ctx, cancel := context.WithCancel(context.Background())
api := &API{
ctx: ctx,
cancel: cancel,
config: cfg,
server: s,
host: h,
rpcReady: make(chan struct{}, 2),
}
api.addRoutes(router)
// Set up api.httpListener if enabled
err = api.setupHTTP()
if err != nil {
return nil, err
}
// Set up api.libp2pListener if enabled
err = api.setupLibp2p(ctx)
if err != nil {
return nil, err
}
if api.httpListener == nil && api.libp2pListener == nil {
return nil, ErrNoEndpointsEnabled
}
api.run()
return api, nil
}
func (api *API) setupHTTP() error {
if api.config.HTTPListenAddr == nil {
return nil
}
n, addr, err := manet.DialArgs(api.config.HTTPListenAddr)
if err != nil {
return err
}
var l net.Listener
if api.config.TLS != nil {
l, err = tls.Listen(n, addr, api.config.TLS)
} else {
l, err = net.Listen(n, addr)
}
if err != nil {
return err
}
api.httpListener = l
return nil
}
func (api *API) setupLibp2p(ctx context.Context) error {
// Make new host. Override any provided existing one
// if we have config for a custom one.
if api.config.Libp2pListenAddr != nil {
h, err := libp2p.New(
ctx,
libp2p.Identity(api.config.PrivateKey),
libp2p.ListenAddrs([]ma.Multiaddr{api.config.Libp2pListenAddr}...),
)
if err != nil {
return err
}
api.host = h
}
if api.host == nil {
return nil
}
l, err := gostream.Listen(api.host, p2phttp.P2PProtocol)
if err != nil {
return err
}
api.libp2pListener = l
return nil
}
// HTTPAddress returns the HTTP(s) listening address
// in host:port format. Useful when configured to start
// on a random port (0). Returns error when the HTTP endpoint
// is not enabled.
func (api *API) HTTPAddress() (string, error) {
if api.httpListener == nil {
return "", ErrHTTPEndpointNotEnabled
}
return api.httpListener.Addr().String(), nil
}
// Host returns the libp2p Host used by the API, if any.
// The result is either the host provided during initialization,
// a default Host created with options from the configuration object,
// or nil.
func (api *API) Host() host.Host {
return api.host
}
func (api *API) addRoutes(router *mux.Router) {
for _, route := range api.routes() {
if api.config.BasicAuthCreds != nil {
route.HandlerFunc = basicAuth(route.HandlerFunc, api.config.BasicAuthCreds)
}
router.
Methods(route.Method).
Path(route.Pattern).
Name(route.Name).
Handler(route.HandlerFunc)
}
api.router = router
}
func basicAuth(h http.HandlerFunc, credentials map[string]string) http.HandlerFunc {
return func(w http.ResponseWriter, r *http.Request) {
w.Header().Set("WWW-Authenticate", `Basic realm="Restricted"`)
username, password, ok := r.BasicAuth()
if !ok {
resp, err := unauthorizedResp()
if err != nil {
logger.Error(err)
return
}
http.Error(w, resp, 401)
return
}
authorized := false
for u, p := range credentials {
if u == username && p == password {
authorized = true
}
}
if !authorized {
resp, err := unauthorizedResp()
if err != nil {
logger.Error(err)
return
}
http.Error(w, resp, 401)
return
}
h.ServeHTTP(w, r)
}
}
func unauthorizedResp() (string, error) {
apiError := types.Error{
Code: 401,
Message: "Unauthorized",
}
resp, err := json.Marshal(apiError)
return string(resp), err
}
func (api *API) routes() []route {
return []route{
{
"ID",
"GET",
"/id",
api.idHandler,
},
{
"Version",
"GET",
"/version",
api.versionHandler,
},
{
"Peers",
"GET",
"/peers",
api.peerListHandler,
},
{
"PeerAdd",
"POST",
"/peers",
api.peerAddHandler,
},
{
"PeerRemove",
"DELETE",
"/peers/{peer}",
api.peerRemoveHandler,
},
{
"Allocations",
"GET",
"/allocations",
api.allocationsHandler,
},
{
"Allocation",
"GET",
"/allocations/{hash}",
api.allocationHandler,
},
{
"StatusAll",
"GET",
"/pins",
api.statusAllHandler,
},
{
"SyncAll",
"POST",
"/pins/sync",
api.syncAllHandler,
},
{
"RecoverAll",
"POST",
"/pins/recover",
api.recoverAllHandler,
},
{
"Status",
"GET",
"/pins/{hash}",
api.statusHandler,
},
{
"Pin",
"POST",
"/pins/{hash}",
api.pinHandler,
},
{
"Unpin",
"DELETE",
"/pins/{hash}",
api.unpinHandler,
},
{
"Sync",
"POST",
"/pins/{hash}/sync",
api.syncHandler,
},
{
"Recover",
"POST",
"/pins/{hash}/recover",
api.recoverHandler,
},
{
"ConnectionGraph",
"GET",
"/health/graph",
api.graphHandler,
},
}
}
func (api *API) run() {
if api.httpListener != nil {
api.wg.Add(1)
go api.runHTTPServer()
}
if api.libp2pListener != nil {
api.wg.Add(1)
go api.runLibp2pServer()
}
}
// runs in goroutine from run()
func (api *API) runHTTPServer() {
defer api.wg.Done()
<-api.rpcReady
logger.Infof("REST API (HTTP): %s", api.config.HTTPListenAddr)
err := api.server.Serve(api.httpListener)
if err != nil && !strings.Contains(err.Error(), "closed network connection") {
logger.Error(err)
}
}
// runs in goroutine from run()
func (api *API) runLibp2pServer() {
defer api.wg.Done()
<-api.rpcReady
listenMsg := ""
for _, a := range api.host.Addrs() {
listenMsg += fmt.Sprintf(" %s/ipfs/%s\n", a, api.host.ID().Pretty())
}
logger.Infof("REST API (libp2p-http): ENABLED. Listening on:\n%s\n", listenMsg)
err := api.server.Serve(api.libp2pListener)
if err != nil && !strings.Contains(err.Error(), "context canceled") {
logger.Error(err)
}
}
// Shutdown stops any API listeners.
func (api *API) Shutdown() error {
api.shutdownLock.Lock()
defer api.shutdownLock.Unlock()
if api.shutdown {
logger.Debug("already shutdown")
return nil
}
logger.Info("stopping Cluster API")
api.cancel()
close(api.rpcReady)
// Cancel any outstanding ops
api.server.SetKeepAlivesEnabled(false)
if api.httpListener != nil {
api.httpListener.Close()
}
if api.libp2pListener != nil {
api.libp2pListener.Close()
}
// This means we created the host
if api.config.Libp2pListenAddr != nil {
api.host.Close()
}
api.wg.Wait()
api.shutdown = true
return nil
}
// SetClient makes the component ready to perform RPC
// requests.
func (api *API) SetClient(c *rpc.Client) {
api.rpcClient = c
// One notification for http server and one for libp2p server.
api.rpcReady <- struct{}{}
api.rpcReady <- struct{}{}
}
func (api *API) idHandler(w http.ResponseWriter, r *http.Request) {
idSerial := types.IDSerial{}
err := api.rpcClient.Call("",
"Cluster",
"ID",
struct{}{},
&idSerial)
sendResponse(w, err, idSerial)
}
func (api *API) versionHandler(w http.ResponseWriter, r *http.Request) {
var v types.Version
err := api.rpcClient.Call("",
"Cluster",
"Version",
struct{}{},
&v)
sendResponse(w, err, v)
}
func (api *API) graphHandler(w http.ResponseWriter, r *http.Request) {
var graph types.ConnectGraphSerial
err := api.rpcClient.Call("",
"Cluster",
"ConnectGraph",
struct{}{},
&graph)
sendResponse(w, err, graph)
}
func (api *API) peerListHandler(w http.ResponseWriter, r *http.Request) {
var peersSerial []types.IDSerial
err := api.rpcClient.Call("",
"Cluster",
"Peers",
struct{}{},
&peersSerial)
sendResponse(w, err, peersSerial)
}
func (api *API) peerAddHandler(w http.ResponseWriter, r *http.Request) {
dec := json.NewDecoder(r.Body)
defer r.Body.Close()
var addInfo peerAddBody
err := dec.Decode(&addInfo)
if err != nil {
sendErrorResponse(w, 400, "error decoding request body")
return
}
mAddr, err := ma.NewMultiaddr(addInfo.PeerMultiaddr)
if err != nil {
sendErrorResponse(w, 400, "error decoding peer_multiaddress")
return
}
var ids types.IDSerial
err = api.rpcClient.Call("",
"Cluster",
"PeerAdd",
types.MultiaddrToSerial(mAddr),
&ids)
sendResponse(w, err, ids)
}
func (api *API) peerRemoveHandler(w http.ResponseWriter, r *http.Request) {
if p := parsePidOrError(w, r); p != "" {
err := api.rpcClient.Call("",
"Cluster",
"PeerRemove",
p,
&struct{}{})
sendEmptyResponse(w, err)
}
}
func (api *API) pinHandler(w http.ResponseWriter, r *http.Request) {
if ps := parseCidOrError(w, r); ps.Cid != "" {
logger.Debugf("rest api pinHandler: %s", ps.Cid)
err := api.rpcClient.Call("",
"Cluster",
"Pin",
ps,
&struct{}{})
sendAcceptedResponse(w, err)
logger.Debug("rest api pinHandler done")
}
}
func (api *API) unpinHandler(w http.ResponseWriter, r *http.Request) {
if ps := parseCidOrError(w, r); ps.Cid != "" {
logger.Debugf("rest api unpinHandler: %s", ps.Cid)
err := api.rpcClient.Call("",
"Cluster",
"Unpin",
ps,
&struct{}{})
sendAcceptedResponse(w, err)
logger.Debug("rest api unpinHandler done")
}
}
func (api *API) allocationsHandler(w http.ResponseWriter, r *http.Request) {
var pins []types.PinSerial
err := api.rpcClient.Call("",
"Cluster",
"Pins",
struct{}{},
&pins)
sendResponse(w, err, pins)
}
func (api *API) allocationHandler(w http.ResponseWriter, r *http.Request) {
if ps := parseCidOrError(w, r); ps.Cid != "" {
var pin types.PinSerial
err := api.rpcClient.Call("",
"Cluster",
"PinGet",
ps,
&pin)
if err != nil { // errors here are 404s
sendErrorResponse(w, 404, err.Error())
return
}
sendJSONResponse(w, 200, pin)
}
}
func (api *API) statusAllHandler(w http.ResponseWriter, r *http.Request) {
queryValues := r.URL.Query()
local := queryValues.Get("local")
if local == "true" {
var pinInfos []types.PinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"StatusAllLocal",
struct{}{},
&pinInfos)
sendResponse(w, err, pinInfosToGlobal(pinInfos))
} else {
var pinInfos []types.GlobalPinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"StatusAll",
struct{}{},
&pinInfos)
sendResponse(w, err, pinInfos)
}
}
func (api *API) statusHandler(w http.ResponseWriter, r *http.Request) {
queryValues := r.URL.Query()
local := queryValues.Get("local")
if ps := parseCidOrError(w, r); ps.Cid != "" {
if local == "true" {
var pinInfo types.PinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"StatusLocal",
ps,
&pinInfo)
sendResponse(w, err, pinInfoToGlobal(pinInfo))
} else {
var pinInfo types.GlobalPinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"Status",
ps,
&pinInfo)
sendResponse(w, err, pinInfo)
}
}
}
func (api *API) syncAllHandler(w http.ResponseWriter, r *http.Request) {
queryValues := r.URL.Query()
local := queryValues.Get("local")
if local == "true" {
var pinInfos []types.PinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"SyncAllLocal",
struct{}{},
&pinInfos)
sendResponse(w, err, pinInfosToGlobal(pinInfos))
} else {
var pinInfos []types.GlobalPinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"SyncAll",
struct{}{},
&pinInfos)
sendResponse(w, err, pinInfos)
}
}
func (api *API) syncHandler(w http.ResponseWriter, r *http.Request) {
queryValues := r.URL.Query()
local := queryValues.Get("local")
if ps := parseCidOrError(w, r); ps.Cid != "" {
if local == "true" {
var pinInfo types.PinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"SyncLocal",
ps,
&pinInfo)
sendResponse(w, err, pinInfoToGlobal(pinInfo))
} else {
var pinInfo types.GlobalPinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"Sync",
ps,
&pinInfo)
sendResponse(w, err, pinInfo)
}
}
}
func (api *API) recoverAllHandler(w http.ResponseWriter, r *http.Request) {
queryValues := r.URL.Query()
local := queryValues.Get("local")
if local == "true" {
var pinInfos []types.PinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"RecoverAllLocal",
struct{}{},
&pinInfos)
sendResponse(w, err, pinInfosToGlobal(pinInfos))
} else {
sendErrorResponse(w, 400, "only requests with parameter local=true are supported")
}
}
func (api *API) recoverHandler(w http.ResponseWriter, r *http.Request) {
queryValues := r.URL.Query()
local := queryValues.Get("local")
if ps := parseCidOrError(w, r); ps.Cid != "" {
if local == "true" {
var pinInfo types.PinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"RecoverLocal",
ps,
&pinInfo)
sendResponse(w, err, pinInfoToGlobal(pinInfo))
} else {
var pinInfo types.GlobalPinInfoSerial
err := api.rpcClient.Call("",
"Cluster",
"Recover",
ps,
&pinInfo)
sendResponse(w, err, pinInfo)
}
}
}
func parseCidOrError(w http.ResponseWriter, r *http.Request) types.PinSerial {
vars := mux.Vars(r)
hash := vars["hash"]
_, err := cid.Decode(hash)
if err != nil {
sendErrorResponse(w, 400, "error decoding Cid: "+err.Error())
return types.PinSerial{Cid: ""}
}
pin := types.PinSerial{
Cid: hash,
}
queryValues := r.URL.Query()
name := queryValues.Get("name")
pin.Name = name
pin.Recursive = true // For now all CLI pins are recursive
rplStr := queryValues.Get("replication_factor")
rplStrMin := queryValues.Get("replication_factor_min")
rplStrMax := queryValues.Get("replication_factor_max")
if rplStr != "" { // override
rplStrMin = rplStr
rplStrMax = rplStr
}
if rpl, err := strconv.Atoi(rplStrMin); err == nil {
pin.ReplicationFactorMin = rpl
}
if rpl, err := strconv.Atoi(rplStrMax); err == nil {
pin.ReplicationFactorMax = rpl
}
return pin
}
func parsePidOrError(w http.ResponseWriter, r *http.Request) peer.ID {
vars := mux.Vars(r)
idStr := vars["peer"]
pid, err := peer.IDB58Decode(idStr)
if err != nil {
sendErrorResponse(w, 400, "error decoding Peer ID: "+err.Error())
return ""
}
return pid
}
func pinInfoToGlobal(pInfo types.PinInfoSerial) types.GlobalPinInfoSerial {
return types.GlobalPinInfoSerial{
Cid: pInfo.Cid,
PeerMap: map[string]types.PinInfoSerial{
pInfo.Peer: pInfo,
},
}
}
func pinInfosToGlobal(pInfos []types.PinInfoSerial) []types.GlobalPinInfoSerial {
gPInfos := make([]types.GlobalPinInfoSerial, len(pInfos), len(pInfos))
for i, p := range pInfos {
gPInfos[i] = pinInfoToGlobal(p)
}
return gPInfos
}
func sendResponse(w http.ResponseWriter, rpcErr error, resp interface{}) {
if checkRPCErr(w, rpcErr) {
sendJSONResponse(w, 200, resp)
}
}
// checkRPCErr takes care of returning standard error responses if we
// pass an error to it. It returns true when everythings OK (no error
// was handled), or false otherwise.
func checkRPCErr(w http.ResponseWriter, err error) bool {
if err != nil {
sendErrorResponse(w, 500, err.Error())
return false
}
return true
}
func sendEmptyResponse(w http.ResponseWriter, rpcErr error) {
if checkRPCErr(w, rpcErr) {
w.WriteHeader(http.StatusNoContent)
}
}
func sendAcceptedResponse(w http.ResponseWriter, rpcErr error) {
if checkRPCErr(w, rpcErr) {
w.WriteHeader(http.StatusAccepted)
}
}
func sendJSONResponse(w http.ResponseWriter, code int, resp interface{}) {
w.Header().Add("Content-Type", "application/json")
w.WriteHeader(code)
if err := json.NewEncoder(w).Encode(resp); err != nil {
panic(err)
}
}
func sendErrorResponse(w http.ResponseWriter, code int, msg string) {
errorResp := types.Error{
Code: code,
Message: msg,
}
logger.Errorf("sending error response: %d: %s", code, msg)
sendJSONResponse(w, code, errorResp)
}