outputlookup

Description

Writes search results to a static lookup table, or KV store collection, that you specify.

Syntax

The required syntax is in bold.

| outputlookup

[append=<bool>]

[create_empty=<bool>]

[override_if_empty=<bool>]

[max=<int>]

[key_field=<field>]

[createinapp=<bool>]

[output_format=<string>]

<filename> | <tablename>

Required arguments

You must specify one of the following required arguments, either filename or tablename.

filename

Syntax: <string>

Description: The name of the lookup file. The file must end with .csv or .csv.gz.

tablename

Syntax: <string>

Description: The name of the lookup table as specified by a stanza name in transforms.conf. The lookup table can be configured for any lookup type (CSV, external, or KV store).

Optional arguments

append

Syntax: append=<bool>

Description: The default setting, append=false, writes the search results to the .csv file or KV store collection. Columns that are not in the current search results are removed from the file. If set to true, attempts to append search results to an existing .csv file or KV store collection. Otherwise it creates a file. If there is an existing .csv file, the outputlookup command writes only the fields that are present in the previously existing .csv file. An outputlookup search that is run with append=true might result in a situation where the lookup table or collection is only partially updated. This means that a subsequent lookup or inputlookup search on that lookup table or collection might return stale data along with new data. The outputlookup command cannot append to .gz files.

Default: false

create_empty

Syntax: create_empty=<bool>

Description: If set to true and there are no results, a zero-length file is created. When set to false and there are no results, no file is created. If the file previously existed, the file is deleted.

For example, suppose there is a system-level lookup called test with the lookup defined in test.csv. There is also an app-level lookup with the same name. If an app overrides that test.csv in its own app directory with an empty file create_empty=true, the app-level lookup behaves as if the lookup is empty. However, if there's no file at all create_empty=false at the app level, then the lookup file in the system-level is used.

Default: false

createinapp

Syntax: createinapp=<bool>

Description: If set to false, or if there is no current application context, the command creates the file in the system lookups directory.

Default: true

key_field

Syntax: key_field=<field>

Description: For KV store-based lookups, uses the specified field name as the key to a value and replaces that value. An outputlookup search using the key_field argument might result in a situation where the lookup table or collection is only partially updated. A subsequent lookup or inputlookup search on that collection might return stale data along with new data. A partial update only occurs with concurrent searches, one with the outputlookup command and a search with the inputlookup command. It is possible that the inputlookup occurs when the outputlookup is still updating some of the records.

max

Syntax: max=<int>

Description: The number of rows to output.

Default: no limit

output_format

Syntax: output_format=splunk_sv_csv | splunk_mv_csv

Description: Controls the output data format of the lookup. Use output_format=splunk_mv_csv when you want to output multivalued fields to a lookup table file, and then read the fields back into Splunk using the inputlookup command. The default, splunk_sv_csv outputs a CSV file which excludes the _mv_<fieldname> fields.

Default: splunk_sv_csv

override_if_empty

Syntax: override_if_empty=<bool>

Description: If override_if_empty=true and no results are passed to the output file, the existing output file is deleted, If override_if_empty=false and no results are passed to the output file, the command does not delete the existing output file.

Default: true

Usage

The lookup table must be a CSV or GZ file, or a table name specified with a lookup table configuration in transforms.conf. The lookup table can refer to a KV store collection or a CSV lookup. The outputlookup command cannot be used with external lookups.

For CSV lookups, if the lookup file does not exist, it is created in the lookups directory of the current application. If the lookup file already exists, it is overwritten with the results of the outputlookup command. If the createinapp option is set to false or if there is no current application context, then the file is created in the system lookups directory.

For permissions in CSV lookups, use the check_permission field in transforms.conf and outputlookup_check_permission in limits.conf to restrict write access to users with the appropriate permissions when using the outputlookup command. Both check_permission and outputlookup_check_permission default to false. Set to true for Splunk software to verify permission settings for lookups for users. You can change lookup table file permissions in the .meta file for each lookup file, or Settings > Lookups > Lookup table files. By default, only users who have the admin or power role can write to a shared CSV lookup file.

For more information about creating lookups, see About lookups in the Knowledge Manager Manual.

For more information about App Key Value Store collections, see About KV store in the Admin Manual.

Appending results

Suppose you have an existing CSV file which contains columns A, D, and J. The results of your search are columns A, C, and J. If you run a search with outputlookup append=false, then columns A, C, and J are written to the CSV file. Column D is not retained.

If you run a search with outputlookup append=true, then only the columns that are currently in the file are preserved. In this example columns A and J are written to the CSV file. Column C is lost because it does not already exist in the CSV file. Column D is retained.

You can work around this issue by using the eval command to add a column to your CSV file before you run the search. For example, if your CSV file is named foo you would do something like this:

| inputlookup foo | eval c=null | outputlookup foo append=false ....

Then run your search and pipe the results to the fields command for the columns you want to preserve.

... | fields A C J | outputlookup append=true foo

Multivalued fields

When you output to a static lookup table, the outputlookup command merges values in a multivalued field into single space-delimited value. This does not apply to a KV store collection.

Examples

1. Write to a lookup table using settings in the transforms.conf file

Write to usertogroup lookup table as specified in the transforms.conf file.

| outputlookup usertogroup

2. Write to a lookup file in a specific system or app directory

Write to users.csv lookup file under $SPLUNK_HOME/etc/system/lookups or $SPLUNK_HOME/etc/apps/*/lookups.

| outputlookup users.csv

3. Specify not to override the lookup file if no results are returned

Write to users.csv lookup file, if results are returned, under $SPLUNK_HOME/etc/system/lookups or $SPLUNK_HOME/etc/apps/*/lookups. Do not delete the users.csv file if no results are returned.

| outputlookup users.csv override_if_empty=false

4. Write to a KV store collection

Write food inspection events for Shalimar Restaurant to a KV store collection called kvstorecoll. This collection is referenced in a lookup table called kvstorecoll_lookup.

index=sf_food_health sourcetype=sf_food_inspections name="SHALIMAR RESTAURANT" | outputlookup kvstorecoll_lookup

5. Write from a CSV file to a KV store collection

Write the contents of a CSV file to the KV store collection kvstorecoll using the lookup table kvstorecoll_lookup. This requires usage of both inputlookup and outputlookup commands.

| inputlookup customers.csv | outputlookup kvstorecoll_lookup

6. Update field values for a single KV store collection record

To update field values for a single KV store collection record you must specify the internal key ID for the record.

To learn how to obtain the internal key ID values of the records in a KV store collection, see "Return the internal key ID values for the KV store collection" in the Examples section of the inputlookup command.

You must use the inputlookup, outputlookup, and eval commands to update field values for a single KV store collection record.

In the following example, the KV store collection record is indicated by the value of its internal key ID, the _key field, and is updated with a new customer name and customer city. The record belongs to the KV store collection kvstorecoll, which is accessed through the lookup table kvstorecoll_lookup.

| inputlookup kvstorecoll_lookup | search _key=544948df3ec32d7a4c1d9755 | eval CustName="Vanya Patel" | eval CustCity="Springfield" | outputlookup kvstorecoll_lookup append=True key_field=_key

The outputlookup command appends the search results to the specified lookup file, kvstorecoll_lookup. The key_field argument identifies the field in the collection that contains the key ID for the values that you want to append.

See also

Commands

collect

inputlookup

lookup

inputcsv

mcollect

meventcollect

outputcsv

outputtext