[-]
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
  
 [+]
  
  
  
  
  
  
 [+]
  
  
  
[+]
[+]
[+]
[+]
[+]
[+]
[+]
[+]
[+]
Updated on 8/4/2019
Administration Guides
ECA (Eyeglass Clustered Agent) CLI Commands
Direct link to topic in this publication:
Home

ECA (Eyeglass  Clustered Agent) CLI Commands

Key commands used for operations of ECA clusters

ecactl cluster up - start cluster

ecactl cluster down - stop cluster

ecactl cluster exec <command> - run a command on all nodes

  1. ecactl cluster exec ecactl version  (checks version on all nodes)
  2. ecactl cluster exec ecactl cluster hdfs-status (check ports and dns resolution of HDFS url)
  3. ecactl cluster exec 'ping -c3 172.22.1.81'  (ping an ip from all nodes needs single quote)
  4. Very Useful:  ecactl cluster exec "ecactl stats --no-stream" (this will snapshot stats on all containers on all nodes)

ecactl stats - show container stats

ecactl db shell - show hbase shell to run commands example status, count, scan tables

ecactl components install eca - used one time to setup ssh keys between nodes

ecactl containers ps - list all running containers, ports, names, status

ecactl containers (start, stop, restart) container-name (restart a container, stop or start if stopped)

ecactl components clear -  removes old container images from previous upgrades


ecactl cluster services list  (lists all  containers configured to run on each node in the cluster ) 

ecactl cluster service nodesfor <container name>  returns all nodes configured to run the named service

  • example ecactl cluster service nodesfor kafka

 ecactl cluster services  launch <container name list>  (starts the container on the current node if not already running, space separated list of container names)

  • ecactl cluster service launch kafka dns

 ecactl cluster service restart     (on current node stop, remove ALL containers, NOTE must use launch command to start them again

  • add --pull (command to stop, remove and pull latest image of this container onto this node)
    • ecactl cluster service restart --pull
  • add --all command flag stops, removes all containers on all nodes but requires launch command to start container again)
    • ecactl cluster service restart --all

ecactl zk --help  (commands for zookeeper)

ecactl zk shell (opens shell)

ls /superna/eca/turboaudit  (will list sub folders to get status)

ls /superna/eca/turboaudit/ecanodes

ls /superna/eca/turboaudit/auditfolders  (list the cluster folders and select a folder name with /owner and use the get command to find the assigned ECA node of this audit folder on the Isilon cluster)

example

 get /superna/eca/turboaudit/auditfolders/00505698f0793f8bbb56fc176e2f7b6e204c_node001/owner 

ecactl kafka topics --help

This will list the available flags for kafka.

ecactl kafka topics --describe

Which returns info about all kafka topics. If these commands to not error out, this can be considered a successful test.

ecactl cluster services list  (lists each node and the services running on that node in a distributed cluster with different services running on each node.)

Schedule cron to restart containers


  1. ecactl cluster exec "sudo -E ecactl components restart-cron set fastanalysis,evtarchive,turboaudit 0 0,6,12,18 \'*\' \'*\' \'*\'"
    1. This command will set cron on three containers using the cron string shown above.     The above cron string is the default that should be used. Note: the outer double " is required for the ecactl cluster exec command
    2. password for each node will be required 
  2. sudo -E ecactl components restart-cron set <container>[,<container>[...]] <cron interval>

  3. sudo -E ecactl components restart-cron remove <container>[,<container>[...]]   (this removes the cron setting for  1 or more containers)
  4. NOTE: can be executed on a single node


Copyright Superna LLC