Vault will now register itself with Consul. The active node can be found using `active.vault.service.consul`. All standby vaults are available via `standby.vault.service.consul`. All unsealed vaults are considered healthy and available via `vault.service.consul`. Change in status and registration is event driven and should happen at the speed of a write to Consul (~network RTT + ~1x fsync(2)).
Healthy/active:
```
curl -X GET 'http://127.0.0.1:8500/v1/health/service/vault?pretty' && echo;
[
{
"Node": {
"Node": "vm1",
"Address": "127.0.0.1",
"TaggedAddresses": {
"wan": "127.0.0.1"
},
"CreateIndex": 3,
"ModifyIndex": 20
},
"Service": {
"ID": "vault:127.0.0.1:8200",
"Service": "vault",
"Tags": [
"active"
],
"Address": "127.0.0.1",
"Port": 8200,
"EnableTagOverride": false,
"CreateIndex": 17,
"ModifyIndex": 20
},
"Checks": [
{
"Node": "vm1",
"CheckID": "serfHealth",
"Name": "Serf Health Status",
"Status": "passing",
"Notes": "",
"Output": "Agent alive and reachable",
"ServiceID": "",
"ServiceName": "",
"CreateIndex": 3,
"ModifyIndex": 3
},
{
"Node": "vm1",
"CheckID": "vault-sealed-check",
"Name": "Vault Sealed Status",
"Status": "passing",
"Notes": "Vault service is healthy when Vault is in an unsealed status and can become an active Vault server",
"Output": "",
"ServiceID": "vault:127.0.0.1:8200",
"ServiceName": "vault",
"CreateIndex": 19,
"ModifyIndex": 19
}
]
}
]
```
Healthy/standby:
```
[snip]
"Service": {
"ID": "vault:127.0.0.2:8200",
"Service": "vault",
"Tags": [
"standby"
],
"Address": "127.0.0.2",
"Port": 8200,
"EnableTagOverride": false,
"CreateIndex": 17,
"ModifyIndex": 20
},
"Checks": [
{
"Node": "vm2",
"CheckID": "serfHealth",
"Name": "Serf Health Status",
"Status": "passing",
"Notes": "",
"Output": "Agent alive and reachable",
"ServiceID": "",
"ServiceName": "",
"CreateIndex": 3,
"ModifyIndex": 3
},
{
"Node": "vm2",
"CheckID": "vault-sealed-check",
"Name": "Vault Sealed Status",
"Status": "passing",
"Notes": "Vault service is healthy when Vault is in an unsealed status and can become an active Vault server",
"Output": "",
"ServiceID": "vault:127.0.0.2:8200",
"ServiceName": "vault",
"CreateIndex": 19,
"ModifyIndex": 19
}
]
}
]
```
Sealed:
```
"Checks": [
{
"Node": "vm2",
"CheckID": "serfHealth",
"Name": "Serf Health Status",
"Status": "passing",
"Notes": "",
"Output": "Agent alive and reachable",
"ServiceID": "",
"ServiceName": "",
"CreateIndex": 3,
"ModifyIndex": 3
},
{
"Node": "vm2",
"CheckID": "vault-sealed-check",
"Name": "Vault Sealed Status",
"Status": "critical",
"Notes": "Vault service is healthy when Vault is in an unsealed status and can become an active Vault server",
"Output": "Vault Sealed",
"ServiceID": "vault:127.0.0.2:8200",
"ServiceName": "vault",
"CreateIndex": 19,
"ModifyIndex": 38
}
]
```
No signal handler was setup to receive SIGINT. I didn't investigate to
see if signal(2) mask was setup (ala `SIG_IGN`) or if sigprocmask(2) is
being used, but in either case, the correct behavior is to capture and
treat SIGINT the same as SIGTERM. At some point in the future these two
signals may affect the running process differently, but we will clarify
that difference in the future.
One thing that has been a point of confusion for users is Vault's
response when deleting a key that does not actually exist in the system.
For example, consider:
$ vault delete secret/foo
Success! Deleted 'secret/foo'
This message is misleading if the secret does not exist, especially if
the same command is run twice in a row.
Obviously the reason for this is clear - returning an error if a secret
does not exist would reveal the existence of a secret (the same reason
everything on S3 is a 403 or why GitHub repos 404 instead of 403 if you
do not have permission to view them).
I think we can make the UX a little bit better by adding just a few
words to the output:
$ vault delete secret/foo
Success! Deleted 'secret/foo' if it existed
This makes it clear that the operation was only performed if the secret
existed, but it does not reveal any more information.
This came up twice, in two different training courses. The UX is a
little confusing here on the CLI. Users are used to running:
$ vault auth abcd-1234...
So when they auth using a method, the output leads them to believe the
need to "re-auth" as the generated token:
$ vault auth -method=userpass username=foo password=bar
Successfully authenticated!
token: defg-5678...
A number of users then run:
$ vault auth defg-5678
I've added some helpful text to hint this is not required if the method
is not "token".
This was a flawed test. Previously the test passed in a fixture that
corresponded to a CLI config file, not an actual policy. The test
_should_ have been failing, but it wasn't. This commit adds a new
fixture.
This does NOT apply to the backend config, since each backend config
could have a variation of options that differ based off of the
configured backend itself. This may be an optimization that can be made
in the future, but I think each backend should be responsible for
performing its own configuration validation instead of overloading the
config itself with this functionality.
In some situations, it can be impossible to revoke leases (for instance,
if someone has gone and manually removed users created by Vault). This
can not only cause Vault to cycle trying to revoke them, but it also
prevents mounts from being unmounted, leaving them in a tainted state
where the only operations allowed are to revoke (or rollback), which
will never successfully complete.
This adds a new endpoint that works similarly to `revoke-prefix` but
ignores errors coming from a backend upon revocation (it does not ignore
errors coming from within the expiration manager, such as errors
accessing the data store). This can be used to force Vault to abandon
leases.
Like `revoke-prefix`, this is a very sensitive operation and requires
`sudo`. It is implemented as a separate endpoint, rather than an
argument to `revoke-prefix`, to ensure that control can be delegated
appropriately, as even most administrators should not normally have
this privilege.
Fixes#1135
renew-self endpoint. Otherwise it will use the renew endpoint, even if
the token matches the client token.
Adds an -increment flag to allow increments even with no token passed
in.
Fixes#1150
This endpoint causes the node it's hit to step down from active duty.
It's a noop if the node isn't active or not running in HA mode. The node
will wait one second before attempting to reacquire the lock, to give
other nodes a chance to grab it.
Fixes#1093
auth-enable command help in the "Auth Enable Options" is suggesting
the usage of a non-existing command called 'auth-list' instead of
the correct one "auth -methods"
This change is almost perfectly compatible with the existing code,
except it's a little shorter because it uses a list of a available
formatters that must implement a `command.Formatter` interface.
Also added some basic formatting tests.
backends for the moment. This is pretty simple; it just adds the actual
capability to make a list call into both the CLI and the HTTP handler.
The real meat was already in those backends.
Functionality is split into ExternalTokenHelper, which is used if a path
is given in a configuration file, and InternalTokenHelper which is used
otherwise. The internal helper no longer shells out to the same Vault
binary, instead performing the same actions with internal code. This
avoids problems using dev mode when there are spaces in paths or when
the binary is built in a container without a shell.
Fixes#850 among others
