Command | Arguments | Description |
---|---|---|
ess version | print out the current version of Essentia and the AQ tools. | |
ess select | source
[--region REGION] [--credentials file] [--aws_access_key publickey] [--account_name AzureAccount] [--aws_secret_access_key secretkey] [--account_key AzureKey] [--label name] |
Select a data source |
ess summary | [category] [--label name] [--scan] |
Provide a summary of the categories in the current datastore, or a deep summary of a single category |
ess purge | label | Delete the reference to the datastore (not the datastore itself) |
ess ls | [pattern] [--cat pattern] [--label name] [-r] |
list the contents of the source datastore |
ess probe | category
[--file filename] [--label name] [--pcmd ProbeCommand] [--size Bytes] |
scans a file to determine its compression, file format, etc. |
ess category add | name pattern
[--dateformat auto|none|custom] [--archive pattern] [--compression type|none] [--delimiter delimiter] [--columnspec NewColumnSpec] [--allx] [--alls] [--overwrite] [--label name] [--comment comment] [--noprobe] |
Add a category to the datastore |
ess category delete | ExistingCategory
|
Remove the reference to this category |
ess category copy | ExistingCategory NewCategory | Create a copy of an existing category reference |
ess category change | columnspec|dateformat|comment
NewSpec|NewFormat|NewComment
|
Modify or override details about a category |
ess cluster create | [--type] [--number] |
Create worker nodes |
ess cluster terminate | Shutdown all worker nodes | |
ess cluster status | Summarizes the state of all workers | |
ess cluster stop | Suspend worker nodes (i.e. stop the EC2 instances) | |
ess cluster start | Restart suspended worker nodes | |
ess cluster set | local|cloud | Force essentia into master node only (local) or not. |
ess query | SQL style query on raw logs | |
ess stream | category start stop command
[--exclude pattern] [--master] [--debug] [--bulk] [--threads number] [--archive pattern] [--s3out label:path] [--progress] [--limit number] |
stream data in given range to given command (cat being default) |
ess exec | command
[--master] [--debug] [--s3out label:path] |
execute given command on all worker nodes |
ess server reset | Delete all database definitions | |
ess server restart | Clear database contents, but maintain schemas | |
ess server commit | Upload database spec files to workers | |
ess server summary | Summarize the databases and objects available for processing | |
ess create | database|table|vector|variable name spec | create an object. Name not required for variable |
ess drop | database|table|vector|variable name | delete the object |
ess use | database | select the given database for queries |
ess cat | filename
[--label name] |
read file and print to stdout |
ess lsa | filename
[--label name] [--pattern] |
List the files within an archive file with optional filtering. |
ess file push | Send a local file to worker nodes | |
ess file get | Fetch a file from workers to the master | |
ess file mkdir | create a directory on worker nodes | |
ess redshift register | clusterName | Look up your redshift cluster and find its connection information |
ess redshift stream | category startdate stopdate command
[--debug] [--master] [--threads] [--bulk] [-U userName] [-d redshiftTableName] [-p password] |
Import data from specified category between (and including) specified dates using the specified filter. Default is to execute on worker nodes. Data is streamed, one file at a time, to the given command and then into your redshift table |
ess redshift status | Get information about the host and port | |
ess udbd | start|stop|restart|status | direct manipulation of UDB |
ess udbd | ckmem|cklog | check memory or logs of UDB daemons on workers |