open-consul/agent/consul/client_serf.go

195 lines
5.5 KiB
Go
Raw Normal View History

package consul
import (
"fmt"
"path/filepath"
"strings"
"github.com/hashicorp/go-hclog"
"github.com/hashicorp/serf/serf"
"github.com/hashicorp/consul/agent/metadata"
New ACLs (#4791) This PR is almost a complete rewrite of the ACL system within Consul. It brings the features more in line with other HashiCorp products. Obviously there is quite a bit left to do here but most of it is related docs, testing and finishing the last few commands in the CLI. I will update the PR description and check off the todos as I finish them over the next few days/week. Description At a high level this PR is mainly to split ACL tokens from Policies and to split the concepts of Authorization from Identities. A lot of this PR is mostly just to support CRUD operations on ACLTokens and ACLPolicies. These in and of themselves are not particularly interesting. The bigger conceptual changes are in how tokens get resolved, how backwards compatibility is handled and the separation of policy from identity which could lead the way to allowing for alternative identity providers. On the surface and with a new cluster the ACL system will look very similar to that of Nomads. Both have tokens and policies. Both have local tokens. The ACL management APIs for both are very similar. I even ripped off Nomad's ACL bootstrap resetting procedure. There are a few key differences though. Nomad requires token and policy replication where Consul only requires policy replication with token replication being opt-in. In Consul local tokens only work with token replication being enabled though. All policies in Nomad are globally applicable. In Consul all policies are stored and replicated globally but can be scoped to a subset of the datacenters. This allows for more granular access management. Unlike Nomad, Consul has legacy baggage in the form of the original ACL system. The ramifications of this are: A server running the new system must still support other clients using the legacy system. A client running the new system must be able to use the legacy RPCs when the servers in its datacenter are running the legacy system. The primary ACL DC's servers running in legacy mode needs to be a gate that keeps everything else in the entire multi-DC cluster running in legacy mode. So not only does this PR implement the new ACL system but has a legacy mode built in for when the cluster isn't ready for new ACLs. Also detecting that new ACLs can be used is automatic and requires no configuration on the part of administrators. This process is detailed more in the "Transitioning from Legacy to New ACL Mode" section below.
2018-10-19 16:04:07 +00:00
"github.com/hashicorp/consul/agent/structs"
"github.com/hashicorp/consul/lib"
libserf "github.com/hashicorp/consul/lib/serf"
"github.com/hashicorp/consul/logging"
"github.com/hashicorp/consul/types"
)
// setupSerf is used to setup and initialize a Serf
func (c *Client) setupSerf(conf *serf.Config, ch chan serf.Event, path string) (*serf.Serf, error) {
conf.Init()
conf.NodeName = c.config.NodeName
conf.Tags["role"] = "node"
conf.Tags["dc"] = c.config.Datacenter
conf.Tags["segment"] = c.config.Segment
conf.Tags["id"] = string(c.config.NodeID)
conf.Tags["vsn"] = fmt.Sprintf("%d", c.config.ProtocolVersion)
conf.Tags["vsn_min"] = fmt.Sprintf("%d", ProtocolVersionMin)
conf.Tags["vsn_max"] = fmt.Sprintf("%d", ProtocolVersionMax)
conf.Tags["build"] = c.config.Build
if c.config.AdvertiseReconnectTimeout != 0 {
conf.Tags[libserf.ReconnectTimeoutTag] = c.config.AdvertiseReconnectTimeout.String()
}
New ACLs (#4791) This PR is almost a complete rewrite of the ACL system within Consul. It brings the features more in line with other HashiCorp products. Obviously there is quite a bit left to do here but most of it is related docs, testing and finishing the last few commands in the CLI. I will update the PR description and check off the todos as I finish them over the next few days/week. Description At a high level this PR is mainly to split ACL tokens from Policies and to split the concepts of Authorization from Identities. A lot of this PR is mostly just to support CRUD operations on ACLTokens and ACLPolicies. These in and of themselves are not particularly interesting. The bigger conceptual changes are in how tokens get resolved, how backwards compatibility is handled and the separation of policy from identity which could lead the way to allowing for alternative identity providers. On the surface and with a new cluster the ACL system will look very similar to that of Nomads. Both have tokens and policies. Both have local tokens. The ACL management APIs for both are very similar. I even ripped off Nomad's ACL bootstrap resetting procedure. There are a few key differences though. Nomad requires token and policy replication where Consul only requires policy replication with token replication being opt-in. In Consul local tokens only work with token replication being enabled though. All policies in Nomad are globally applicable. In Consul all policies are stored and replicated globally but can be scoped to a subset of the datacenters. This allows for more granular access management. Unlike Nomad, Consul has legacy baggage in the form of the original ACL system. The ramifications of this are: A server running the new system must still support other clients using the legacy system. A client running the new system must be able to use the legacy RPCs when the servers in its datacenter are running the legacy system. The primary ACL DC's servers running in legacy mode needs to be a gate that keeps everything else in the entire multi-DC cluster running in legacy mode. So not only does this PR implement the new ACL system but has a legacy mode built in for when the cluster isn't ready for new ACLs. Also detecting that new ACLs can be used is automatic and requires no configuration on the part of administrators. This process is detailed more in the "Transitioning from Legacy to New ACL Mode" section below.
2018-10-19 16:04:07 +00:00
if c.acls.ACLsEnabled() {
// we start in legacy mode and then transition to normal
// mode once we know the cluster can handle it.
conf.Tags["acls"] = string(structs.ACLModeLegacy)
} else {
conf.Tags["acls"] = string(structs.ACLModeDisabled)
}
// We use the Intercept variant here to ensure that serf and memberlist logs
// can be streamed via the monitor endpoint
serfLogger := c.logger.
NamedIntercept(logging.Serf).
NamedIntercept(logging.LAN).
StandardLoggerIntercept(&hclog.StandardLoggerOptions{InferLevels: true})
memberlistLogger := c.logger.
NamedIntercept(logging.Memberlist).
NamedIntercept(logging.LAN).
StandardLoggerIntercept(&hclog.StandardLoggerOptions{InferLevels: true})
conf.MemberlistConfig.Logger = memberlistLogger
conf.Logger = serfLogger
conf.EventCh = ch
conf.ProtocolVersion = protocolVersionMap[c.config.ProtocolVersion]
conf.RejoinAfterLeave = c.config.RejoinAfterLeave
conf.Merge = &lanMergeDelegate{
dc: c.config.Datacenter,
nodeID: c.config.NodeID,
nodeName: c.config.NodeName,
segment: c.config.Segment,
}
conf.SnapshotPath = filepath.Join(c.config.DataDir, path)
if err := lib.EnsurePath(conf.SnapshotPath, false); err != nil {
return nil, err
}
addEnterpriseSerfTags(conf.Tags)
conf.ReconnectTimeoutOverride = libserf.NewReconnectOverride(c.logger)
return serf.Create(conf)
}
// lanEventHandler is used to handle events from the lan Serf cluster
func (c *Client) lanEventHandler() {
var numQueuedEvents int
for {
numQueuedEvents = len(c.eventCh)
if numQueuedEvents > serfEventBacklogWarning {
c.logger.Warn("number of queued serf events above warning threshold",
"queued_events", numQueuedEvents,
"warning_threshold", serfEventBacklogWarning,
)
}
select {
case e := <-c.eventCh:
switch e.EventType() {
case serf.EventMemberJoin:
c.nodeJoin(e.(serf.MemberEvent))
case serf.EventMemberLeave, serf.EventMemberFailed, serf.EventMemberReap:
c.nodeFail(e.(serf.MemberEvent))
case serf.EventUser:
c.localEvent(e.(serf.UserEvent))
case serf.EventMemberUpdate: // Ignore
c.nodeUpdate(e.(serf.MemberEvent))
case serf.EventQuery: // Ignore
default:
c.logger.Warn("unhandled LAN Serf Event", "event", e)
}
case <-c.shutdownCh:
return
}
}
}
// nodeJoin is used to handle join events on the serf cluster
func (c *Client) nodeJoin(me serf.MemberEvent) {
for _, m := range me.Members {
ok, parts := metadata.IsConsulServer(m)
if !ok {
continue
}
if parts.Datacenter != c.config.Datacenter {
c.logger.Warn("server has joined the wrong cluster: wrong datacenter",
"server", m.Name,
"datacenter", parts.Datacenter,
)
continue
}
c.logger.Info("adding server", "server", parts)
c.router.AddServer(types.AreaLAN, parts)
// Trigger the callback
if c.config.ServerUp != nil {
c.config.ServerUp()
}
}
}
// nodeUpdate is used to handle update events on the serf cluster
func (c *Client) nodeUpdate(me serf.MemberEvent) {
for _, m := range me.Members {
ok, parts := metadata.IsConsulServer(m)
if !ok {
continue
}
if parts.Datacenter != c.config.Datacenter {
c.logger.Warn("server has joined the wrong cluster: wrong datacenter",
"server", m.Name,
"datacenter", parts.Datacenter,
)
continue
}
c.logger.Info("updating server", "server", parts.String())
c.router.AddServer(types.AreaLAN, parts)
}
}
// nodeFail is used to handle fail events on the serf cluster
func (c *Client) nodeFail(me serf.MemberEvent) {
for _, m := range me.Members {
ok, parts := metadata.IsConsulServer(m)
if !ok {
continue
}
c.logger.Info("removing server", "server", parts.String())
c.router.RemoveServer(types.AreaLAN, parts)
}
}
// localEvent is called when we receive an event on the local Serf
func (c *Client) localEvent(event serf.UserEvent) {
// Handle only consul events
if !strings.HasPrefix(event.Name, "consul:") {
return
}
switch name := event.Name; {
case name == newLeaderEvent:
c.logger.Info("New leader elected", "payload", string(event.Payload))
// Trigger the callback
if c.config.ServerUp != nil {
c.config.ServerUp()
}
case isUserEvent(name):
event.Name = rawUserEventName(name)
c.logger.Debug("user event", "name", event.Name)
// Trigger the callback
if c.config.UserEventHandler != nil {
c.config.UserEventHandler(event)
}
default:
if !c.handleEnterpriseUserEvents(event) {
c.logger.Warn("Unhandled local event", "event", event)
}
}
}