Index Index for
Section 5
Index Alphabetical
listing for S
Bottom of page Bottom of
page

sys_attrs_drd(5)

NAME

sys_attrs_drd - drd subsystem attributes

DESCRIPTION

This reference page lists and describes attributes for the device request dispatcher (drd) kernel subsystem. Refer to the sys_attrs(5) reference page for an introduction to the topic of kernel subsystem attributes. Module_name Name of this subcomponent: DeviceIO Request Dispatcher Supported operations: Query drd_max_hash_length Specifies the length of the largest chain in the device request dispatcher subsystem's member/disk map entry hash table. Supported operations: Query Default value: 1 Minimum value: 0 Maximum value: 5000 drd_hash_size Determines the number of hash entires allocated to keep lists of device request dispatcher devices. The default is to set drd_hash_size equal to the largest hash list since boot. Supported operations: Query; Configure at boot time Default value: 256 Minimum value: 1 Maximum value: 5000 drd_served_disks If this attribute is set to to 1, the device request dispatcher subsystem treats all disks as singly served disks, even direct-access I/O disks. A value of 0 (zero) enables direct-access I/O functionality. Do not modify the default setting for this attribute unless instructed to do so by Compaq Computer Corporation support personnel or by patch kit documentation. Supported operations: Query; Configure at boot time Default value: 0 Minimum value: 0 Maximum value: 1 drd_daio_disks If this attribute is set to to 1, the device request dispatcher subsystem treats all disks as direct-access I/O disks, even single- server disks. A value of 0 (zero) enables single-server disks. Do not modify the default setting for this attribute unless instructed to do so by Compaq Computer Corporation support personnel or by patch kit documentation. Supported operations: Query; Configure at boot time Default value: 0 Minimum value: 0 Maximum value: 1 drd_panic_all When set to 1, causes any generated panic within the device request dispatcher to cause all cluster node members to panic as well. This can be useful in determining (relative) clusterwide conditions at the time of a crash. Supported operations: Query; Configure at boot time; Reconfigure at run time Default value: 0 Minimum value: 0 Maximum value: 1 drd_nolwc When set to 1, DRD does not use lightweight context when handling buffered I/O completion for disks against which I/O barriers can be posed. Supported operations: Query; Configure at boot time; Reconfigure at run time Default value: 0 Minimum value: 0 Maximum value: 1 drd_remote_maxphys Specifies the size of the largest single I/O transfer that is supported when using the cluster interconnect. I/O requests that are larger than this value are fragmented. Supported operations: Query; Configure at boot time Default value: 524288 Minimum value: 262144 Maximum value: 1048576 drd_scan_rate Number of seconds between scans for restoration of failed disks. Setting drd_scan_rate to 0 disables scanning for restored paths. A single scan occurs when a device goes bad. Supported operations: Query; Configure at boot time Default value: 10 Minimum value: 0 Maximum value: 86400 (once per day) drd_iobarrier_to The number of seconds before timeouts on I/O barrier commands, including reservation and registration operations. If a device does not respond withing that time limit, the device will not be used by device request dispatcher from this member. The minimum is 0 (zero), which practically ensures that I/O barrier commands will fail, and so is of limited use. Supported operations: Query; Configure at boot time Default value: 120 Minimum value: 0 Maximum value: 86400 drd_nopr_list First entry on the list of hardware IDs for devices on which DRD will not perform persistent reservations. Supported operations: Query; Configure at boot time; Reconfigure at run time drd_target_reset_wait An interval, in seconds, during which all cluster I/O with shared devices is suspended. When a cluster member loses quorum, all of its I/O is suspended, and the remaining cluster nodes erect I/O barriers against nodes that have been removed from the cluster. This I/O barrier operation inhibits non-cluster members from performing I/O with shared storage devices. In the case of certain third-party SCSI storage devices, a temporary suspension of I/O is needed. See the Cluster Administration manual for information about using this attribute to manage third-party storage. Supported operations: Query; Configure at boot time; Reconfigure at run time Default value: 30 Minimum value: 0 Maximum value: 100000

SEE ALSO

sys_attrs(5), drdmgr(8) TruCluster Server Cluster Administration

Index Index for
Section 5
Index Alphabetical
listing for S
Top of page Top of
page