Vault 1.10.0 release notes
Software Release date: Mar 23, 2022
Summary: Vault version 1.10.0 offers features and enhancements that improve the user experience while closing the loop on key issues previously encountered by our customers. We are providing a summary of these improvements in these release notes.
We encourage you to upgrade to the latest release to take advantage of the new benefits that we are providing. Additionally, with this latest release, we offer solutions to critical feature gaps that have been identified previously. For further information on product improvements, including a comprehensive list of bug fixes, please refer to the Changelog within the Vault 1.10.0 release.
Some of these enhancements and changes in this release include:
- Ability to view client counts per auth and changes to clients over months, therefore, providing more granular visibility into clients.
- Extended the
sys/remountAPI endpoint to support moving secrets engines and auth method mounts from one location to another, within a namespace or across namespaces.
- Improved security posture that includes MFA on login for Vault Community Edition customers.
- Ability to implicitely achieve consistency via tokens.
- Support of PKCE on Vault’s OIDC auth method with Telemetry support for the Vault Agent.
- Improvement of key areas and parity to support using Terraform Provider with Vault.
This section describes the new features introduced as part of Vault
Vault has had support for the Step-up Enterprise MFA as part of its Enterprise edition. The Step-up Enterprise MFA allows having an MFA on login, or for step-up access to sensitive resources in Vault.
With Vault 1.10.0, MFA as part of login is now supported for Vault Community Edition. This demonstrates HashiCorp’s thought leadership in security and its continued endeavor to enable all Vault users to employ strong security policies with Vault.
Note: The Legacy MFA in Vault Community Edition is a deprecated feature and will be removed in Vault 1.11.
Refer to the Login MFA FAQ to understand the various MFA workflows that are supported in Vault 1.10.0.
Vault’s support to act as an OIDC provider is now generally available. Furthermore, Vault’s OIDC provider functionality can now support PKCE for authorization code flow as well. Thanks to all the excellent community feedback received, we have simplified the user experience around configuration of OIDC provider functionality.
With 0.6.0, Vault Lambda Extension supports caching in the local proxy server to avoid proxying every request to enable setting expiry time and invalidate cache, as needed.
We have introduced three new resources to enable configuration of the KMIP secrets engine using the Terraform Provider for Vault. In addition, frequent releases on the Terraform Provider for Vault have been incorporating the ability to configure newer resources and data sources. Please read the documentation for more details.
We now support an additional method for managing KV v2 secrets to maintain least privilege security in certain types of automated environments. This feature creates a new PATCH capability that enables partial updates to KV v2 secrets without requiring the READ privilege to the entire endpoint for an entity.
Vault operators can leverage the openldap secrets engine to manage credentials for IBM DB2 and the LDAP security plugin for Db2. This allows Db2 to offload authentication and authorization to the LDAP security plugin and allows Vault to manage static credentials or even generate dynamic users. For more details, refer to the For more details, refer to the IBM Db2 Credentials Management tutorial.
Proper key management includes occasionally rotating encryption keys to reduce the risks of a nonce reuse and opportunities for keys to be compromised. Previously, there was no automated way to rotate keys that is native to Vault. Now, we have provided a new configuration element on transit keys and tokenization transform configurations where a time interval triggers the keys to automatically rotate after the interval has lapsed.
To address security and compliance needs, customers may require that keys be either created or stored within Hardware Security Models (HSMs). Vault 1.10.0 introduces an accommodation for this requirement with regards to the PKI Secrets Engine. We now support offloading selected PKI operations to HSMs, in particular allowing customers to both generate new PKI key pairs and sign/verify some certificate workflows. All of these operations are conducted in a way that never allows the private key material to leave the secure confines of the HSM itself.
The work done above to support HSM-backed PKI operations inspired us to consider what other key possession paradigms we could support. This led us to extend the implementation to support Cloud Key Management Systems in addition to HSMs. In Vault 1.10.0, users may generate new PKI pairs and perform sign/verify certificate workflows, all with those keys never leaving the cloud KMS itself. Vault 1.10.0 provides support for AWS Key Management Service and Azure Key Vault Key Management Service.
Vault’s eventual consistency model precludes read-after-write guarantees when clients interact with performance standbys or performance replication clusters. The Client Controlled Consistency mitigations supported with Vault 1.7 provide ways to achieve consistency through client modifications or by using the agent for proxied requests, which is not possible in all cases. The Server Side Consistent Tokens feature provides an implicit way to achieve consistency by embedding the minimum Write-Ahead-Log state information in the Service tokens returned from logins or token-create requests. This feature introduces changes in the token format and the new tokesn will be the default tokens starting in Vault 1.10.0. Vault 1.10.0 is backwards compatible with old tokens.
See Replication, Vault Eventual Consistency, Upgrade to 1.10.0 and Server Side Consistent Token FAQ to understand the various consistency options available with Vault 1.10.0 and the considerations to be aware of prior to selecting an option for your use case.
Starting with Vault 1.10.0, the Vault Agent supports a new metrics endpoint and Telemetry metrics around run time, authentication success, authentication failures, cache hits, cache misses, proxy succes, and proxy client errors. This Vault Agent Telemetry should greatly help with the retrieval of key operational insights for Vault Agent deployments.
With this enhancement, users can specify user-assigned managed identities via the
client_id when configuring Vault agent auto-auth for Azure. This enables users that have more than one user-assigned managed identity associated with their VM to specify which one they'd like to use when authenticating via the Vault's Azure auth method. Note that providing these parameters is an "exclusive or" operation.
Previously, for instances where the Agent is a sidecar in a Kubernetes job and the job hangs, you must either use
shareProcessNamespace: true for the container so that the process kill signals can be sent, or avoid the sidecar container entirely and solely rely on an init container. With this enhancement, we have added support for a Quit API endpoint to automatically shut down the Vault Agent, therefore eliminating the need to perform the workarounds.
This section describes other features and enhancements introduced as part of the Vault 1.10.0 release.
We have introduced auth mount-based attribution of clients to help better understand where clients are being used within a cluster. This is available via UI and API. This is an enhancement on top of the namespace attribution capability we introduced in Vault 1.9.
We have also introduced the ability to view changes to clients month over month via the client count API, and made other UI enhancements. Refer to What is a Client? and Client Count FAQ for more details.
We have made improvements to the
sys/remount API endpoint to simplify the complexities of moving data, such as secret engine and authentication method configuration from one mount to another, within a namespace or across namespaces. This can help with restructuring namespaces and mounts for various reasons, including migrating mounts from root to other namespaces when transitioning to using namespaces for the first time. For step-by-step instructions, refer to the Mount Move tutorial.
Database plugins can now implement plugin multiplexing which allows a single plugin process to be used for multiple database connections. Database plugin multiplexing will be enabled on the Oracle Database plugin starting in v0.6.0. We will extend this functionality to additional database plugins in subsequent releases.
Any external database plugins that want to adopt multiplexing support will have to update their main.go call from dbplugin.Serve() to dbplugin.ServeMultiplex(). Multiplexable database plugins are compatible with older versions of Vault down to Vault 1.6. Refer to this Oracle Database PR as an example of the upgrade process.
Consul has supported namespace, admin partitions and ACL roles for some time now. In this release we have added enhancements to the Consul Secrets engine to support namespace awareness and add admin partition and role support for Consul ACL tokens. This significantly simplifies the integrations for customers who want to achieve a zero trust security posture with both Vault and Consul.
Prior to Vault 1.10.0, the Vault UI used localStorage to store authentication information. The data in localStorage was persisted in browsers and removed only on demand. Now, we have switched the Vault UI to use sessionStorage instead, which ensures that the authentication information is stored in the current browser tab alone, thereby improving security.
The Transform Secrets Engine allows users to securely encrypt data while providing control over the output format. In Vault 1.9, we introduced additional format fields on the templates used for this workflow. In Vault 1.10.0, we have now added those two new fields,
decode_format, to the Create Template page on the UI under Advanced Templating.
The following section details breaking changes introduced in Vault 1.10.0.
In Vault 1.9, we added support to provide custom user filters through the userfilter parameter. This support changed the way that entity alias was mapped to an entity. Prior to Vault 1.9, alias names were always based on the login username (which in turn is based on the value of the userattr). In Vault 1.9, alias names no longer mapped to the login username. Instead, the mapping depends on other config values as well, such as updomain, binddn, discoverydn, and userattr.
With Vault 1.10.0, we re-introduced the option to force the alias name to map to the login username with the optional parameter username_as_alias. Users that have the LDAP auth method enabled prior to Vault 1.9 may want to consider setting this to true to revert back to the old behavior. Otherwise, depending on the other aforementioned config values, logins may generate a new and different entity for an existing user with a previous entity associated in Vault. This in turn affects client counts since there may be more than one entity tied to this user. The username_as_alias flag was also made available in subsequent Vault 1.8.x and Vault 1.9.x releases to allow for this to be set prior to a Vault 1.10.0 upgrade.
We now support Server Side Consistent Tokens (See Replication, Vault Eventual Consistency, and Upgrade to 1.10.0.), which introduces a new token format that can only be used on nodes of 1.10 or higher version. This new format is enabled by default upon upgrading to the new version. Old format tokens can be read by Vault 1.10.0, but the new format Vault 1.10 tokens cannot be read by older Vault versions.
For more details, see the Server Side Consistent Tokens FAQ.
Since service tokens are always created on the leader, as long as the leader is not upgraded before performance standbys, service tokens will be of the old format and still be usable during the upgrade process. However, the usual upgrade process we recommend can't be relied upon to always upgrade the leader last. Due to this known issue, a Vault cluster using Integrated Storage may result in a leader not being upgraded last, and this can trigger a re-election. This re-election can cause the upgraded node to become the leader, resulting in the newly created tokens on the leader to be unusable on nodes that have not yet been upgraded. Note that this issue does not impact Vault Community Edition users.
We will have a fix for this issue in Vault 1.10.1. Until this issue is fixed, you may be at risk of having performance standbys unable to service requests until all nodes are upgraded. We recommended that you plan for a maintenance window to upgrade.
When a user has a policy that allows creating a secret engine but not reading it, after successful creation, the user sees a message
n is undefined instead of a permissions error. We will have a fix for this issue in an upcoming minor release.
When adding or modifying a Duo MFA method for step-up Enterprise MFA using the
sys/mfa/method/duo endpoint, a panic gets triggered due to a missing schema field. We will have a fix for this in Vault 1.10.1. Until this issue is fixed, avoid making any changes to your Duo configuration if you are upgrading Vault to v1.10.0.
Signing in to the Vault UI using an OIDC auth mount listed in the "tabs" of the form will result
in the following error: "Authentication failed: role with oidc role_type is not allowed".
The auth mounts listed in the "tabs" of the form are those that have listing_visibility
There is a workaround for this error that will allow you to sign in to Vault using the OIDC auth method. Select the "Other" tab instead of selecting the specific OIDC auth mount tab. From there, select "OIDC" from the "Method" select box and proceed to sign in to Vault.
When trying to start Vault server 1.10.0 on Windows, and there is less than 100GB of free disk space, there is an initialization error with raft DB related to insufficient space on the disk. See this issue for details. Windows users should wait till 1.10.1 to upgrade.
Please refer to the Deprecation Plans and Notice page for up-to-date information on feature deprecations and plans. An Feature Deprecation FAQ page is also available to address questions concerning decisions made about Vault feature deprecations.