Snowflake Time Travel - A complete guide

In the recent years IT Roles such as Database administrators are becoming  less and less relevant as most cloud based databases and datawarehouses are managed by IT engineers needing very little or no administration.

Snowflake already has very little administration. You do not need to do typical DBA tasks like index tables, partitioning or deal with space related issues. Time travel is another feature that lets developers self serve dba related task of reverting or restoring the database after you do issue some inadvertent commands.

In this detailed hands on video I explained about parameters that affect time travel and how to utilize this feature to revert the accidental or intentional database changes. 

Here is the video link - https://www.youtube.com/watch?v=CRZQjxP10G0&list=PLUxVA4dlJgcz90Vbufh6o6p8Rrew-lnE3&index=3

Here are the list of commands used in the video

-- DATA_RETENTION_TIME_IN_DAYS

show parameters;
show parameters in database SALES;
show parameters in warehouse XMALLFORFINANCE;
show parameters in account;
show parameters like '%DATA%';
show parameters like 'DATA%' in account;
--CHANGES DATA RETENTION FOR THE DATABASEshow parameters in database SALES;alter DATABASE SALES set DATA_RETENTION_TIME_IN_DAYS = 1 ;

--CHANGES DATA RETENTION FOR THE WHOLE ACCOUNTshow parameters like 'DATA%' in account;alter ACCOUNT set DATA_RETENTION_TIME_IN_DAYS = 1 ;

select * from PATIENTS;
INSERT INTO PATIENTS values  (5,'TEST PATIENT','AMSTERDAM','Mayo Clinic');select * from PATIENTS at(offset => -60*5);
delete from PATIENTS where id=5;drop table sales.snow.PATIENTS ;
--Over a dayselect * from PATIENTS at(offset => -60*500000);

select * from PATIENTS at(timestamp => 'Sat, 16 Mar 2019 16:20:00 -0700'::timestamp);

select * from user_access;
--Before changes make by a specific queryselect * from patients before(statement => '025e545d-fc23-4e8d-9ac5-335943a1bec2');
create table restored_table clone user_access  at(timestamp => 'Mon, 09 May 2019 01:01:00 +0300'::timestamp);    create table restored_table clone user_access  at(offset => -60*10);    select * from restored_table;  drop table restored_table;    --Schema as existed 1 hour before  create schema restored_schema clone snow at(offset => -3600);  drop schema restored_schema;    create database restored_db clone sales  before(statement => '025e545d-fc23-4e8d-9ac5-335943a1bec2');  drop database restored_db;  show tables history  in sales.snow;show tables history  in sales;---Restoring objects
undrop table patients;
undrop schema snow;
undrop database sales;

More details are available in snowflake's documentation page - https://docs.snowflake.net/manuals/user-guide/data-time-travel.html

Check out my youtube channel for more videos - https://www.youtube.com/channel/UCT3bqK2QL93j-IFYFYbvjWQ