Runtime parameters
The following table provides the internal Portworx CSI runtime options that can be used to customize the behavior of your environment during runtime. Use the pxctl cluster options update command to configure the runtime options. For information, see the pxctl cluster CLI page.
| Parameter | Description | Default value |
|---|---|---|
| kvdb- defrag- frequency- in-hours | Interval in hours at which a kvdb defrag will be run. | Default is 2 weeks. Set to 0 to disable KVDB defragmentation. |
| min-kvdb- defrag- limit-mb | Minimum KVDB database size in MBs to trigger a defrag. | 100 MBS |
| auto- decommission- timeout | Timeout (in minutes) after which storage-less nodes will be automatically decommissioned. Timeout cannot be set to zero. | 20 minutes |
| concurrent- api- limit | Maximum number of concurrent api invocations allowed. | 20 |
| default- rpc- timeout | Default RPC timeout (in minutes) for all client communications. | 5 |
| fb- lock- timeout | Timeout (in minutes) for which the KVDB lock will be held during FlashBlade create volume call. | Minimum duration 3 mins. |
| fb- stats- expiry- duration | Duration (in minutes) after which cached stats for FB volumes will expire. | Minimum duration 1 min. |
| incremental- uploads- disabled | Disable Pure1 incremental uploads of PX information | false |
| incremental- uploads- interval-mins | Interval for Pure1 incremental uploads of PX information. | 30 |
| license- expiry- check | Number of days to raise alert before license expires. Set to zero to disable alerts. | 7 |
| license- expiry- check- interval | Interval for license expiry checks. Valid only if 'license-expiry-check' is defined. | 6 h |
| ro- vol- pod- bounce- interval | Interval (in seconds) at which the background task should check for read-only (ro) volume mounts which were supposed to be read-write (rw), so it can bounce pods using such volumes. | 30s |
| disabled-temporary -kvdb-loss-support | Enable or disable temporary kvdb loss support | on |