--- layout: docs page_title: Lease problems description: >- Troubleshoot lease problems in Vault. --- # Troubleshoot lease problems Explanations, workarounds, and solutions for common lease problems in Vault. ## `429 - Too Many Requests` ### Problem Vault returns a `429 - Too Many Requests` response when users try to authenticate. For example: ```text Error making API request. URL: PUT https://127.0.0.1:61555/v1/auth/userpass/login/foo Code: 429. Errors: * 1 error occurred: * request path "auth/userpass/login/foo": lease count quota exceeded ``` ### Cause Vault returns a `429 - Too Many Requests` response if a new lease request violates the configured lease quota limit. To guard against [lease explosions](/vault/docs/troubleshoot/lease-explosions), Vault rejects authentication requests if completing the request would violate the configured lease quota limit. ### Solution 1. Correct any client-side errors that may cause excessive lease creation. 1. Determine if your resource needs have changed and complete the [Protecting Vault with Resource Quotas](/vault/tutorials/operations/resource-quotas) tutorial to determine new, appropriate defaults. 1. Use the [`vault lease`](/vault/docs/commands/lease) CLI command or [lease count quota endpoint](/vault/api-docs/system/lease-count-quotas) to tune your lease count quota. Consider making short-term changes to your lease quotas when you expect a significant increase in lease creation. For example, when you release a new feature or complete a marketing push to increase your user base. ## Lease explosion (degraded performance) ### Problem Your Vault nodes are out of memory and unresponsive to new lease requests. ### Cause Clients have caused a lease explosion with consistent, high-volume API requests. Unchecked lease explosions create cascading denial-of-service issues for the active node that can result in denial-of-service issues for the entire cluster. ### Solution To resolve a lease explosion, you need to mitigate the problem to stabilize Vault and provide space for cluster recovery then clean up your Vault environment. 1. Mitigate resource stress by adjusting TTL values for your Vault instance: Config level | Parameter | Precedence -------------------- | ---------------------- | ----------- Database plugin | `ttl` or `default_ttl` | first Database plugin | `max_ttl` | first AuthN/secrets plugin | `ttl` or `default_ttl` | second AuthN/secrets plugin | `max_ttl` | second Vault | `default_lease_ttl` | last Vault | `max_lease_ttl` | last **Granular TTLs on a role, group, or user level always override plugin and system-wide TTL values**. 1. Use firewalls or load balancers to limit API calls to Vault from aberrant clients and reduce load on the struggling cluster . 1. Once the cluster stabilizes, check the active node to determine if you can wait for it to purge leases automatically or if you need to speed up the process by manually revoking leases. 1. If the cluster requires manual intervention, confirm you have a recent, valid snapshots of the cluster. 1. Once you confirm a valid snapshot of the cluster exists, use [`vault lease revoke`](/vault/docs/commands/lease/revoke) to manually revoke the offending leases. Revoking or forcefully revoking leases is potentially a dangerous operation. Do not proceed without a valid snapshot. If you have a valid Vault Enterprise license, consider contacting the [HashiCorp Customer Support team](https://support.hashicorp.com/) for help. ### Related tutorials - [Troubleshoot irrevocable leases](/vault/tutorials/monitoring/troubleshoot-irrevocable-leases)