consul/command/acl/token/update/token_update.go

321 lines
9.2 KiB
Go
Raw Normal View History

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
package tokenupdate
import (
"flag"
"fmt"
"strings"
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/api"
"github.com/hashicorp/consul/command/acl"
"github.com/hashicorp/consul/command/acl/token"
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/command/flags"
"github.com/mitchellh/cli"
)
func New(ui cli.Ui) *cmd {
c := &cmd{UI: ui}
c.init()
return c
}
type cmd struct {
UI cli.Ui
flags *flag.FlagSet
http *flags.HTTPFlags
help string
tokenID string
policyIDs []string
policyNames []string
roleIDs []string
roleNames []string
serviceIdents []string
description string
mergePolicies bool
mergeRoles bool
mergeServiceIdents bool
showMeta bool
upgradeLegacy bool
format 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
}
func (c *cmd) init() {
c.flags = flag.NewFlagSet("", flag.ContinueOnError)
c.flags.BoolVar(&c.showMeta, "meta", false, "Indicates that token metadata such "+
"as the content hash and raft indices should be shown for each entry")
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
c.flags.BoolVar(&c.mergePolicies, "merge-policies", false, "Merge the new policies "+
"with the existing policies")
c.flags.BoolVar(&c.mergeRoles, "merge-roles", false, "Merge the new roles "+
"with the existing roles")
c.flags.BoolVar(&c.mergeServiceIdents, "merge-service-identities", false, "Merge the new service identities "+
"with the existing service identities")
2019-04-30 15:45:36 +00:00
c.flags.StringVar(&c.tokenID, "id", "", "The Accessor ID of the token to update. "+
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
"It may be specified as a unique ID prefix but will error if the prefix "+
"matches multiple token Accessor IDs")
c.flags.StringVar(&c.description, "description", "", "A description of the token")
c.flags.Var((*flags.AppendSliceValue)(&c.policyIDs), "policy-id", "ID of a "+
"policy to use for this token. May be specified multiple times")
c.flags.Var((*flags.AppendSliceValue)(&c.policyNames), "policy-name", "Name of a "+
"policy to use for this token. May be specified multiple times")
c.flags.Var((*flags.AppendSliceValue)(&c.roleIDs), "role-id", "ID of a "+
"role to use for this token. May be specified multiple times")
c.flags.Var((*flags.AppendSliceValue)(&c.roleNames), "role-name", "Name of a "+
"role to use for this token. May be specified multiple times")
c.flags.Var((*flags.AppendSliceValue)(&c.serviceIdents), "service-identity", "Name of a "+
"service identity to use for this token. May be specified multiple times. Format is "+
"the SERVICENAME or SERVICENAME:DATACENTER1,DATACENTER2,...")
c.flags.BoolVar(&c.upgradeLegacy, "upgrade-legacy", false, "Add new polices "+
"to a legacy token replacing all existing rules. This will cause the legacy "+
"token to behave exactly like a new token but keep the same Secret.\n"+
"WARNING: you must ensure that the new policy or policies specified grant "+
"equivalent or appropriate access for the existing clients using this token.")
c.flags.StringVar(
&c.format,
"format",
token.PrettyFormat,
fmt.Sprintf("Output format {%s}", strings.Join(token.GetSupportedFormats(), "|")),
)
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
c.http = &flags.HTTPFlags{}
flags.Merge(c.flags, c.http.ClientFlags())
flags.Merge(c.flags, c.http.ServerFlags())
flags.Merge(c.flags, c.http.NamespaceFlags())
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
c.help = flags.Usage(help, c.flags)
}
func (c *cmd) Run(args []string) int {
if err := c.flags.Parse(args); err != nil {
return 1
}
if c.tokenID == "" {
c.UI.Error(fmt.Sprintf("Cannot update a token without specifying the -id parameter"))
return 1
}
client, err := c.http.APIClient()
if err != nil {
c.UI.Error(fmt.Sprintf("Error connecting to Consul agent: %s", err))
return 1
}
tokenID, err := acl.GetTokenIDFromPartial(client, c.tokenID)
if err != nil {
c.UI.Error(fmt.Sprintf("Error determining token ID: %v", err))
return 1
}
t, _, err := client.ACL().TokenRead(tokenID, nil)
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 err != nil {
c.UI.Error(fmt.Sprintf("Error when retrieving current token: %v", err))
return 1
}
if c.upgradeLegacy {
if t.Rules == "" {
// This is just for convenience it should actually be harmless to allow it
// to go through anyway.
c.UI.Error(fmt.Sprintf("Can't use -upgrade-legacy on a non-legacy token"))
return 1
}
// Reset the rules to nothing forcing this to be updated as a non-legacy
// token but with same secret.
t.Rules = ""
}
if c.description != "" {
// Only update description if the user specified a new one. This does make
// it impossible to completely clear descriptions from CLI but that seems
// better than silently deleting descriptions when using command without
// manually giving the new description. If it's a real issue we can always
// add another explicit `-remove-description` flag but it feels like an edge
// case that's not going to be critical to anyone.
t.Description = c.description
}
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
parsedServiceIdents, err := acl.ExtractServiceIdentities(c.serviceIdents)
if err != nil {
c.UI.Error(err.Error())
return 1
}
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.mergePolicies {
for _, policyName := range c.policyNames {
found := false
for _, link := range t.Policies {
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 link.Name == policyName {
found = true
break
}
}
if !found {
// We could resolve names to IDs here but there isn't any reason why its would be better
// than allowing the agent to do it.
t.Policies = append(t.Policies, &api.ACLTokenPolicyLink{Name: policyName})
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
}
}
for _, policyID := range c.policyIDs {
policyID, err := acl.GetPolicyIDFromPartial(client, policyID)
if err != nil {
c.UI.Error(fmt.Sprintf("Error resolving policy ID %s: %v", policyID, err))
return 1
}
found := false
for _, link := range t.Policies {
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 link.ID == policyID {
found = true
break
}
}
if !found {
t.Policies = append(t.Policies, &api.ACLTokenPolicyLink{ID: policyID})
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
}
}
} else {
t.Policies = nil
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
for _, policyName := range c.policyNames {
// We could resolve names to IDs here but there isn't any reason why its would be better
// than allowing the agent to do it.
t.Policies = append(t.Policies, &api.ACLTokenPolicyLink{Name: policyName})
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
}
for _, policyID := range c.policyIDs {
policyID, err := acl.GetPolicyIDFromPartial(client, policyID)
if err != nil {
c.UI.Error(fmt.Sprintf("Error resolving policy ID %s: %v", policyID, err))
return 1
}
t.Policies = append(t.Policies, &api.ACLTokenPolicyLink{ID: policyID})
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.mergeRoles {
for _, roleName := range c.roleNames {
found := false
for _, link := range t.Roles {
if link.Name == roleName {
found = true
break
}
}
if !found {
// We could resolve names to IDs here but there isn't any reason why its would be better
// than allowing the agent to do it.
t.Roles = append(t.Roles, &api.ACLTokenRoleLink{Name: roleName})
}
}
for _, roleID := range c.roleIDs {
roleID, err := acl.GetRoleIDFromPartial(client, roleID)
if err != nil {
c.UI.Error(fmt.Sprintf("Error resolving role ID %s: %v", roleID, err))
return 1
}
found := false
for _, link := range t.Roles {
if link.ID == roleID {
found = true
break
}
}
if !found {
t.Roles = append(t.Roles, &api.ACLTokenRoleLink{Name: roleID})
}
}
} else {
t.Roles = nil
for _, roleName := range c.roleNames {
// We could resolve names to IDs here but there isn't any reason why its would be better
// than allowing the agent to do it.
t.Roles = append(t.Roles, &api.ACLTokenRoleLink{Name: roleName})
}
for _, roleID := range c.roleIDs {
roleID, err := acl.GetRoleIDFromPartial(client, roleID)
if err != nil {
c.UI.Error(fmt.Sprintf("Error resolving role ID %s: %v", roleID, err))
return 1
}
t.Roles = append(t.Roles, &api.ACLTokenRoleLink{ID: roleID})
}
}
if c.mergeServiceIdents {
for _, svcid := range parsedServiceIdents {
found := -1
for i, link := range t.ServiceIdentities {
if link.ServiceName == svcid.ServiceName {
found = i
break
}
}
if found != -1 {
t.ServiceIdentities[found] = svcid
} else {
t.ServiceIdentities = append(t.ServiceIdentities, svcid)
}
}
} else {
t.ServiceIdentities = parsedServiceIdents
}
t, _, err = client.ACL().TokenUpdate(t, nil)
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 err != nil {
c.UI.Error(fmt.Sprintf("Failed to update token %s: %v", tokenID, err))
return 1
}
formatter, err := token.NewFormatter(c.format, c.showMeta)
if err != nil {
c.UI.Error(err.Error())
return 1
}
out, err := formatter.FormatToken(t)
if err != nil {
c.UI.Error(err.Error())
return 1
}
if out != "" {
c.UI.Info(out)
}
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
return 0
}
func (c *cmd) Synopsis() string {
return synopsis
}
func (c *cmd) Help() string {
return flags.Usage(c.help, nil)
}
const synopsis = "Update an ACL token"
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
const help = `
Usage: consul acl token update [options]
This command will update a token. Some parts such as marking the token local
cannot be changed.
Update a token description and take the policies from the existing token:
$ consul acl token update -id abcd -description "replication" -merge-policies
Update all editable fields of the token:
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
$ consul acl token update -id abcd \
-description "replication" \
-policy-name "token-replication" \
-role-name "db-updater"
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
`