Changing the cluster configuration data with IBM Storage Scale native REST API
Use the scalectl config command or cluster config REST API endpoints to change the cluster configuration attributes.
| Attribute name and description | -i option allowed1 | -N option allowed2 |
|---|---|---|
| log.file The name of the log file for the administration daemon logs that is written in log.path. The default value is scaleadmd.log. |
no | yes |
| log.level Log level for the administration daemon. The possible values are error, info, and debug. The default value is info. |
yes | yes |
| log.path The directory path where the administration daemon log file
is written. The |
no | yes |
| log.utc Specify whether the administration daemon logs are in UTC timestamps or local timestamps. The default value is true. |
yes | yes |
| lro.cleanup_job_age The duration to retain LRO metadata files. LROs older than this duration are permanently removed. The default value is 240h. |
yes | yes |
| lro.cleanup_number_jobs The total number of LRO jobs that are saved in the local filesystem. When this value is reached, the oldest jobs are removed from the cache until the number of jobs is reduced. The default value is 5000. |
yes | yes |
| lro.replicate_num_jobs The number of LRO jobs sent to other LRO managers in a single gRPC message. This setting is used for internal communication to improve network call efficiency. A value that is too large might cause out-of-memory (OOM) issues. The default value is 1000. |
yes | yes |
| lro.dir The directory path that the LRO manager uses for persistent storage of LRO data. The default value is 2s. |
yes | yes |
| lro.output_limit The maximum number of bytes to display in the JSON output field when listing LROs. The default value is 32. |
yes | yes |
| lro.replicate_wait_time The duration to wait before replicating new and updated LRO jobs to other LRO managers. Longer durations allow for more efficient communication but increase the risk of data loss if a quorum node goes down. The default value is 30s. |
yes | yes |
| lro.rest_wait_time The amount of time the long-running operation (LRO) manager waits for an LRO to complete before returning the current LRO status to the caller. After the specified time, if the LRO is not complete, the user needs to poll the LRO manager for the job status to determine its current running state. The default value is 2s. |
yes | yes |
| lro.stale_wait_time The duration to wait for an LRO job to report its running status before the LRO manager moves the job as stale. The default value is 300s. |
yes | yes |
| nodeid.cachain The TLS certificate chain that is used for internal communication between administration daemons. The value is /var/mmfs/ssl/cachain.pem. |
no | yes |
| nodeid.certificate The TLS certificate that is used for internal communication between administration daemons. The value is /var/mmfs/ssl/cert.pem. |
no | yes |
| nodeid.key The TLS certificate key that is used for internal communication between administration daemons. The value is /var/mmfs/ssl/key.pem. |
no | yes |
| nsd.list_cache_age The amount of time that the results from an NSD list are stored in an internal cache before being discarded. In large NSD environments, pagination calculations are expensive. Caching helps reduce CPU and memory usage. The cache time applies to each individual list request and is not system-wide. Setting a value that is too large might not improve performance and could result in OOM situations. The default value is 30s. |
yes | yes |
| remote_grpc_server.host The interface that the administration daemon
uses to listen for communication from other administration daemons. Use |
no | yes |
| remote_grpc_server.port The port that is used by administration daemon to listen for communication from other administration daemons. The default value is 50052. |
no | no |
| remote_grpc_server.scheme The scheme that is used for communicating between administration daemons. Only https is allowed. The default value is https. |
no | no |
| rest_server.authentication_type The global authentication type. This value determines the other authentication configuration attributes, which must set for authentication to function correctly. Only one authentication method can be active at a time. The possible values are pam, ldap, oidc-userinfo, and cert. The default value is pam. |
no | no |
| rest_server.cert.cachain Specifies the certificate authority (CA) certificate that signs the client certificates. If this field is not specified, the cluster node identity cachain is used. The default value is empty. |
no | no |
| rest_server.cert.allowed_cns Specifies a list of common names (CNs) that are allowed. The CN from the authenticating certificate must exactly match one of the CNs in this list. The default value is empty. |
no | no |
| rest_server.cert.allowed_cns_regex Specifies a regular expression to match the CN of the authenticating certificate. For example, local….$ matches the localhost but not localhost1. |
no | no |
| rest_server.host The interface that is used to listen for REST API
requests. Use |
no | no |
| rest_server.ldap.attributes Sort the entries returned based on the attribute. The default value is empty. |
no | no |
| rest_server.ldap.basedn The LDAP base name to use when authenticating users. The default value is empty. |
no | no |
| rest_server.ldap.binddn The fully qualified domain name to bind to LDAP as. This field can be left blank for anonymous binds. The default value is empty. |
no | no |
| rest_server.ldap.bind_password The password for the binddn user. The default value is empty. |
no | no |
| rest_server.ldap.filter The default search filter for users. The default value is empty. |
no | no |
| rest_server.ldap.url The LDAP server name or IP. The value must start with the prefix of ldap:// or ldaps://. The default value is empty. |
no | no |
| rest_server.oidc.authorization_url The URL of the OIDC provider where the user is redirected to initiate the authentication process. The default value is empty. |
no | no |
| rest_server.oidc.callback_url The URL to which the native REST API redirects after authentication. You must also configure this value in the identity provider. The default value is empty. |
no | no |
| rest_server.oidc.discovery_url The URL that allows the native REST API to retrieve the necessary information from the identity provider. The default value is empty. |
no | no |
| rest_server.oidc.client_id The public identifier for the application. The default value is empty. |
no | no |
| rest_server.oidc.client_secret A secret key provided by the identity provider to ensure that only trusted applications, such as the native REST API, can authenticate. The default value is empty. |
no | no |
| rest_server.oidc.scopes The scope that is used by the native REST API during authentication to authorize access to user details. This is an optional field, and by default,native REST API appends openid to every request. The default value is empty. |
no | no |
| rest_server.oidc.token_url The URL where IBM Storage Scale native REST API requests access and identity tokens from the identity provider. The default value is empty. |
no | no |
| rest_server.oidc.userinfo_url The URL of the OIDC provider where native REST API retrieves user information after the user is authenticated. The default value is empty. |
no | no |
| rest_server.port The port that is used to listen for the incoming REST API requests. |
no | no |
| rest_server.scheme The scheme that is used to connect to the administration daemon for REST API requests. Only https is allowed. The default value is https. |
no | no |
| timeout.api_health_grace The grace period for the admin daemon when computing API health status. |
yes | yes |
| timeout.api_health_recompute The amount of time between each full API health refresh. The default value is 5s. |
yes | yes |
| timeout.api_health_response_timeout The timeout value when retrieving the API health from remote admin daemons. The default value is 5s. |
yes | yes |
| timeout.mmccr The timeout value for the administration daemon when running the mmccr command in the backend. The default value is 10s. |
yes | yes |
| timeout.mmchconfig The timeout value for the administration daemon when running the mmchconfig command in the backend. The default value is 60s. |
yes | yes |
| timeout.mmcmi The timeout value for the administration daemon when running the mmcmi command in the backend. The default value is 10s. |
yes | yes |
| timeout.mmlsconfig The timeout value for the administration daemon when running the mmlsconfig command in the backend. The default value is 10s. |
yes | yes |
| timeout.mmlsfs The timeout value for the administration daemon when running the mmlsfs command in the backend. The default value is 15s. |
yes | yes |
| timeout.mount The timeout value for the administration daemon when running the mount command on the local node. The default value is 120s. |
yes | yes |
| timeout.pam The timeout value for the administration daemon when PAM authentication is enabled to interact with the pam_unix_library. The default value is 10s. |
yes | yes |
| timeout.remote_rpc The timeout value for the administration daemon to daemon remote procedure calls (RPC). The default value is 120s. |
yes | yes |
| timeout.remote_unmount The timeout value for the administration daemon when running the unmount command on the remote node. The default value is 15m. |
yes | yes |
| timeout.systemd The timeout value for the administration daemon when running systemd commands in the backend. The default value is 15m. |
yes | yes |
| timeout.unmount The timeout value for the administration daemon when running the unmount command on the local node. The default value is 120s. |
yes | yes |
|
1 If the -i option is specified, the changes take effect immediately without requiring a restart.
2 If the -N option is specified, the changes apply only to the specified nodes or node classes. |
||