Introduction

Version: 3.3.3

The opHA 3 has a cli tool to perform the same operations than the CLI, but with some additional debugging information and it also allows the task automation. 

/usr/local/omk/bin/opha-cli.pl 

Usage: opha-cli.pl act=[action to take] [options...]
opha-cli.pl act=discover url_base=... username=... password=.... role=... mirror=...
opha-cli.pl act=<import_peers|export_peers|list_peers>
opha-cli.pl act=delete_peer {cluster_id=...|server_name=...}
opha-cli.pl act=pull [data_types=X...] [peers=Y] [force=t]
	    pull data types except nodes
	    primary <-- peers
opha-cli.pl act=sync-all-nodes [peers=Y]
	    sync all node data from the primary to the pollers
	    primary --> peers
opha-cli.pl act=sync-processed-nodes
	    sync node data based on changes done by NMIS9 node_admin.pl
	    primary --> peers
opha-cli.pl act=import_config_data
	    for firsts installation, provide initial data (groups)


opha-cli.pl act=cleanup simulate=f/t
	    clean metadata and files
opha-cli.pl act=clean_orphan_nodes simulate=f/t
	    remove nodes with unknown cluster_id
opha-cli.pl act=resync_nodes peer=server_name
	    remove the nodes from the poller in the primary and pull the nodes from the poller
	    primary <-- peers
opha-cli.pl act=clean_data peer=server_name [all=true]
	    Like resync data but with all the data types
	    primary <-- peers
	    By default, cleanup just pull data
	    all=true includes nodes
opha-cli.pl act=cleanup_poller simulate=f/t
	    from the pollers, clean duplicate configuration items and files
opha-cli.pl act=check_duplicates
	    check for duplicate nodes
opha-cli.pl act=get_status 
opha-cli.pl act=setup-db
opha-cli.pl act=show_roles
opha-cli.pl act=data_verify
opha-cli.pl act=lock_peer {cluster_id=...|server_name=...}
opha-cli.pl act=unlock_peer {cluster_id=...|server_name=...}
opha-cli.pl act=peer_islocked {cluster_id=...|server_name=...}
	    Encryption key
opha-cli.pl act=push_encryption_key 

Core functionality

Discover Peer

You can discover a new peer with the following command:

opha-cli.pl act=discover url_base=... username=... password=.... role=... mirror=...

Where: 

• role: Poller or Mirror
• Mirror: Should we specify the mirror of which poller when role=mirror

Import and Export Peers

We can import, export and list all the peers information with:

opha-cli.pl act=<import_peers|export_peers|list_peers>

Delete Peer

It is possible to delete a peer with the command:

opha-cli.pl act=delete_peer {cluster_id=...|server_name=...}

This command will remove the peer and all the associated data: Nodes, inventory, latest data, etc. 

We need to specify OR the cluster_id OR the server name. 

Pull

With pull, we will sync the inventory, latest data, events, status and registry data.

opha-cli.pl act=pull [data_types=X...] [peers=Y] [force=t]

Where: 

• data_types: We can specify specific data types, if nothing is specified, all data types will be pulled. 
• peers: List of peers to pull. If no peers are specified, all the peers will be pulled. 
• force: By default, just the data changed from the last pull is synchronised. With this parameter, all the data will be moved. 

The pull is running in the opha cron job. 

Mirror and poller pulls

When we pull from a mirror, if its poller is active, just the registry and status data will be pulled.

It will happen the opposite: If the mirror is active, the poller data won't be pulled.  

Nodes Synchronisation

To synchronise the nodes, we can run the polling: 

opha-cli.pl act=sync-all-nodes [peers=Y]

Where: 

• peers: List of peers to sync. If no peers are specified, all the peers will be pulled. 

The sync-all-nodes is running in the opha cron job. 

More information.

sync-processed-nodes

Will sync the nodes processed by NMIS9 node_admin.pl:

opha-cli.pl act=sync-processed-nodes

More information.

Import Initial data

for firsts installations, provide initial data, basically setup the groups for pollers and primary and add the peers.

opha-cli.pl act=import_config_data

Cleanup Functions

cleanup

Function to clean metadata for files and files with no metadata information. This is mainly for configuration files:

opha-cli.pl act=cleanup

By default, it will run in simulation mode. 

Use simulate=f to perform the cleanup function.

clean orphan nodes

It is possible to check with nodes are not associated with any cluster id with the command:

opha-cli.pl act=clean_orphan_nodes simulate=f/t

By default, it will run in simulation mode. 

Use simulate=f to remove the nodes (And associated data).

resync nodes

By default, the Primary pushes the nodes to the pollers. Running this command, it is possible to update the nodes from the pollers:

opha-cli.pl act=resync_nodes peer=server_name

Where:

• peer: Specify the server name. 

Clean data

Will remove all the data from the peer and pull the data again. 

By default, it is not removing/resync the nodes. It is possible to do it with:

• all=true 

Cleanup poller

This operation should be run on a poller. And will clean duplicate configuration items and files:

opha-cli.pl act=cleanup_poller simulate=f/t

By default, it will run in simulation mode. 

Use simulate=f to remove the nodes (And associated data).

Diagnosis information

get status

Get all the peer status information as an array of perl hashes:

opha-cli.pl act=get_status

This is the same information that we see in the opHA front page. 

Show roles

Show the roles defined in the system:

opha-cli.pl act=show_roles

Data Verify

Show how many data do we have for each peer:

opha-cli.pl act=data_verify

How many inventory records, roles, which peer is active or enabled, also duplicate nodes and catchall inventory records duplicated. 

Check Duplicates

opha-cli.pl act=check_duplicates

Similar to data_verify, but will report just the duplicate data. 

Lock Peer

(V. >= 3.3.3) When a peer is doing a critical operation, it will be locked. We can see the lock status of a peer:

opha-cli.pl act=peer_islocked {cluster_id=...|server_name=...}

We can change the lock status of a peer with:

opha-cli.pl act=lock_peer {cluster_id=...|server_name=...}
opha-cli.pl act=unlock_peer {cluster_id=...|server_name=...}

Setup DB

Setup DB indexed. This is run by the installer during installation or upgrade:

opha-cli.pl act=setup-db 

Encryption

The primary can push the encryption key to all the pollers by running the following command: 

opha-cli.pl act=push_encryption_key 

It will run just if the server has the role primary and the key was not modified since the last time it was sent. 

To force send it anyway, you can run it with the force=true argument:

opha-cli.pl act=push_encryption_key force=1