save Search Operator

The save operator allows you to save the results of a query to a lookup table you have already created, as described in Create a Lookup Table. You can use the lookup and cat operator to access the saved data.

You can use the append option with save to merge new and changed rows into a lookup table. If you use save without append, any existing rows in the lookup table will be overwritten by your search results. 

If your lookup table is configured with a time-to-live (TTL), creating or updating a row sets or resets the TTL for that row.   

Either raw or aggregated results can be saved with the save operator.

Syntax 

save [append] path://”<path-to-table>”

Where: 

• path-to-table is the path to the lookup table in the Sumo Logic Library.

note

Be sure to specify the path to the table in in this format: path://”<path-to-table>”

To determine the path to a lookup table, highlight the row for the table in the Sumo Logic Library, and select Copy path to clipboard from the three-dot more options menu for the table.

Rules 

• Your search schema must match the schema of the Lookup Table that you are writing to, including the data types of the fields you want to save to the lookup table. Make sure your search returns all of the fields defined for the lookup table and no additional fields. Additional fields will be dropped and not saved to the lookup table. If your search returns fewer fields than that defined for the lookup table, the search will fail.
• The file size limit for lookup tables is 100 MB.
• You can't create a lookup table with the save operator. You must first create a lookup table, as described in Create a Lookup Table. 
• The save operator is not supported with Scheduled Views.
• Queries that use the save operator can't be pinned.

Example 

Using save without append

This example saves search results data about new user accounts to the lookup table on Sumo Logic  .

| parse "name=*," as name
| parse "action=*," as action
| parse "date=*," as date
| where action="sign-up"
| first(date) as date, first(action) as action by name
| save path://"/Library/Users/myusername@sumologic.com/Users"

The above search would populate the lookup table to have these rows:

| Name | Action | Date | |-||| | John | sign-up | 2012-08-20 | | Bill | sign-up | 2012-08-21 | | Bob | sign-up | 2012-08-21 |

Using save with append

You can use the append option to add rows to a lookup table and to update existing rows. For example, you'd run a scheduled search once a day and use save append to merge new and changed rows into the table. If a row in your search results has the same primary key as a row in the lookup, the lookup table row will be updated. If the primary key in a row in the search results does not match a row in the lookup table, the new row will be added to the lookup. Fields returned by your search that are not in the Lookup Table schema will be dropped and not saved to the lookup table. If your search returns fewer fields than that defined for the lookup table, the search will fail.

Let's say that you'd like to append your lookup file each day by scheduling this search to run every 24 hours:

| parse "name=*," as name
| parse "action=*," as action
| parse "date=*," as date
| where action="sign-up"
| first(date) as date, first(action) as action by name
| save append path://"/Library/Users/myusername@sumologic.com/Users"

Each day the query runs and the new and changed rows are written to the table.

You can also append data to a saved file from different queries. For example, say we have two sources, "bill" that includes billing information, and "config" that contains account information, and we'd like to be able to search for some values from each source. These searches would populate a table with information from both sources:

_source=bill | parse "user_id=*," as name
| parse "user_email=*," as email
| save path://"/Library/Users/myusername@sumologic.com/Users"
_source=config | parse "_user=[*]" as name
| parse "contact_info=[*]" as email
| save append path://"/Library/Users/myusername@sumologic.com/Users"