6.5 HP StoreAll OS CLI Reference Guide

Table 2 Fusion Manager configuration tuning options (continued)
DescriptionOption name
The minimum number of seconds that must elapse before attempting
another failback. Value is an integer. Default: 60 sec.
minFailbackInterval
The maximum number of network block device ports allowed.
Range: 0-128. Default: 32 ports.
nbdPortsLimit
The maximum network block device timeout. Range: 10-300.
Default: 60 sec.
nbdTimeout
Quota monitor loop interval. Value is an integer. Default: 300.quotaMonitorLoopInterval
Segment storage utilization threshold. When exceeded, generates
an event. Value is a percentage. Default: 90 percent.
segmentSpaceUsedEventThreshold
Sets the Fusion Manager failover policy based on how many
segments are unavailable. The following values are:
segmentUnavailableFailoverPolicy
• 0 (default value). If one or more segments is unavailable, failover
is initiated.
• 1. If all segments are unavailable, failover is initiated.
The segmentUnavailableFailoverPolicy option works with
the following related segment unavailable configuration options:
• clearSegmentUnavailableUponFailover
• enableSegmentUnavailableFailover
• segmentUnavailableScanInterval
Sets the interval (in seconds) at which the Fusion Manager checks
for unavailable segments. The default is 30 seconds; the range is
0-100.
segmentUnavailableScanInterval
The segmentUnavailableScanInterval option works with
the following related segment unavailable configuration options:
• clearSegmentUnavailableUponFailover
• enableSegmentUnavailableFailover
• segmentUnavailableFailoverPolicy
Specifies the number of seconds that must pass before a server is
marked down. Value is an integer. Default: 5 minutes (300
seconds).
serverDownTime
Port used for cluster network communication. Range: 0-65535.serverPort
Sets the maximum number of consecutive days that any node in a
StoreAll cluster can be up. The default value is 400 days. HP
serverUptimeEventThreshold
strongly recommends that you do not change the default to a
number higher than 400. When a node reaches the number of
days set by this parameter, an alert
(server.uptime.overshoot) is generated, and it indicates that
the node must be rebooted. Also, the server state that is displays
when executing the ibrix_server –l command is changed to
Up, UptimeOverThreshold as another indication that the node
must be rebooted. If the node is not rebooted, file systems eventually
becomes unresponsive.
NOTE: You can reboot the node at any time. StoreAll ensures
that the maximum uptime of a node does not exceed 400 days,
thereby preventing file system performance issues. To reboot the
node, see "Powering file serving nodes on or off" in the
Administrator Guide for your model.
Sets the number of health reporting cycles that must fail before a
client is considered to have failed. Value is an integer. Default:
120.
skippedCyclesBeforeClientFailure
76 StoreAll commands