epadmin-availabilityzone

epadmin availabilityzone target — Manages StreamBase Runtime availability zones.

SYNOPSIS

epadmin [globalparameters] command availabilityzone [commandparameters]

DESCRIPTION

Use the availabilityzone target to display, rebalance, report on metrics, or restore an availability zone, which is a high availability feature of a Spotfire Streaming cluster. Configure availability zones in a node deployment configuration file; there is a default zone if not configured. See availabilityzone in the Architecture Guide.

The availabilityzone target has these commands:

The designation [sn|ad] in syntax examples is a reminder that this command requires either the global parameter ‑‑servicename or the ‑‑adminport and ‑‑hostname combination to identify the node or cluster of interest.

epadmin display availabilityzone

Displays basic information about an availability zone.

For example, for an availability zone named uk1-az in a cluster named nf with two nodes running in a group named uk1: 

epadmin --servicename=.uk1.nf display availabilityzone --name=uk1-az

Returns: 

[A.uk1.nf] Availability Zone = uk1-az
[A.uk1.nf] Data Distribution Policy = dynamic-data-distribution-policy
[A.uk1.nf] Data Distribution Policy Type = Dynamic
[A.uk1.nf] Quorum State = Not Enabled
[A.uk1.nf] Rebalances Requested on Node = 1
[A.uk1.nf] Rebalances Executed on Node = 1
[A.uk1.nf] Rebalancing on Node = false
[A.uk1.nf] Last Rebalance on Node = Jul 3 2024 12:04:29 PM PDT
[A.uk1.nf] Last Rebalance Duration = less than one millisecond
[A.uk1.nf] Primary Nodes = B.uk1.nf,A.uk1.nf
[A.uk1.nf] Backup Nodes = B.uk2.nf,A.uk2.nf
[A.uk1.nf] Primary Node Pattern = [^.]+.uk1.nf
[A.uk1.nf] Backup Node Pattern = [^.]+.uk2.nf
[B.uk1.nf] Availability Zone = uk1-az
[B.uk1.nf] Data Distribution Policy = dynamic-data-distribution-policy
[B.uk1.nf] Data Distribution Policy Type = Dynamic
[B.uk1.nf] Quorum State = Not Enabled
[B.uk1.nf] Rebalances Requested on Node = 1
[B.uk1.nf] Rebalances Executed on Node = 0
[B.uk1.nf] Rebalancing on Node = false
[B.uk1.nf] Last Rebalance on Node = No rebalance has occurred
[B.uk1.nf] Last Rebalance Duration = No rebalance has occurred
[B.uk1.nf] Primary Nodes = A.uk1.nf,B.uk1.nf
[B.uk1.nf] Backup Nodes = A.uk2.nf,B.uk2.nf
[B.uk1.nf] Primary Node Pattern = [^.]+.uk1.nf
[B.uk1.nf] Backup Node Pattern = [^.]+.uk2.nf

epadmin [sn|ad] display availabilityzone
epadmin [sn|ad] display availabilityzone --name=zone-one
Command Parameter Description Required
name Only display this availability zone name No. The default is to display all availability zones.

epadmin rebalance availabilityzone

If the data distribution policy for a zone allows it, you can dynamically rebalance the distribution of data across an availabilityzone. You might need to do this if a node fails or is taken out of service, or in a multi-master situation (described in the Node Quorum section of the High Availability page of the Architecture Guide). 

epadmin [sn|ad] rebalance availabilityzone
epadmin [sn|ad] rebalance availabilityzone --wait
epadmin [sn|ad] rebalance availabilityzone --name=cluster-availzone
epadmin [sn|ad] rebalance availabilityzone --threads=4 --chunksize=500
Command Parameter Description Required
chunksize The number of objects to lock per transaction when sending data to remote nodes. No. The default value is set in data distribution policy.
name The name of an availability zone to rebalance No. The default is to rebalance all availability zones.
threads The number of threads to use when sending data to remote nodes. No. The default value is set in data distribution policy.
wait Boolean value controlling whether to wait for the rebalance to complete before returning. A value of false causes the command to return immediately with the rebalance completing asynchronously. A value of true causes the command to block until the rebalance completes. No. The default value is false.

epadmin report availabilityzone

Report on distribution of objects across partitions and nodes in an availability zone. For example, for an availability zone named zone-one: 

epadmin --servicename=A.sbuser report availabilityzone --name=zone-one

Returns: 

[A.sbuser] Name = zone-one
[A.sbuser] Number of Partitions = 4
[A.sbuser] Average Partition Object Cardinality = 1.25
[A.sbuser] Maximum Partition Object Cardinality = 3
[A.sbuser] Minimum Partition Object Cardinality = 0
[A.sbuser] Partition Cardinality Standard Deviation = 1.30
[A.sbuser] Average Node Object Cardinality = 2.50
[A.sbuser] Maximum Node Object Cardinality = 5
[A.sbuser] Minimum Node Object Cardinality = 0
[A.sbuser] Node Cardinality Standard Deviation = 2.50
[A.sbuser] Per-Node Object Cardinality = A.sbuser:0, B.sbuser:5

epadmin [sn|ad] report availabilityzone
epadmin [sn|ad] report availabilityzone --name=zone-one
Command Parameter Description Required
name Only report on this availability zone name No. The default is to display all availability zones.

epadmin restore availabilityzone

Use this command to restore an availabilityzone after a multi-master event, as described in the Node Quorum section of the High Availability page of the Architecture Guide). 

epadmin [sn|ad] restore availabilityzone
epadmin [sn|ad] restore availabilityzone --name=cluster-availzone
epadmin [sn|ad] restore availabilityzone --threads=4 --chunksize=500
epadmin [sn|ad] restore availabilityzone --action=replace --from=B.X
Command Parameter Description Required
action One of combine, replace, or merge.

  • combine copies all data from a remote node to the local node, and any data conflicts causes the restore to fail.

  • replace removes all data on the local node and replaces it with data from the remote node.

  • merge compares the data on the local and remote nodes and resolves any conflicts.

No. The default is merge.
chunksize The number of objects to lock per transaction when sending data to remote nodes. No. The default value is set in the data distribution policy.
from Restore data from this node. No. The default is determined using a cluster wide broadcast to locate an active node to restore from.
name The name of the availability zone being restored. No. The default is to restore all availability zones.
threads The number of threads to use when sending data to remote nodes. No. The default value is set in the data distribution policy.

epadmin wait availabilityzone

Wait for an availability zone to be ready. An availability zone is considered ready when the specified number of primary and backup nodes are active. 

epadmin [sn|ad] wait availabilityzone --name=zone-one --primary=4
epadmin [sn|ad] wait availabilityzone --name=zone-one --primary=4 --backup=2
epadmin [sn|ad] wait availabilityzone --name=zone-one --primary=4 --seconds=10
Command Parameter Description Required
name Availability zone name Yes.
primary Number of active primary nodes. Must be a value >= 0. Yes.
backup Number of active backup nodes. Must be a value >=0. No. Default value is 0.
seconds Time to wait in seconds. Must be a value >=0. No. Default value is 0, which waits forever.

SEE ALSO

Help System for epadmin
epadmin(1)
epadmin-globals(1)