You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
A very common use case for transforms is pivoting historical data. But, when there is a large amount of past data, that data may not be very useful in the resulting index.
User's typically achieve this by adding a query filter to the transform:
{"range":{"timestamp": {"gte": "now-1d"}}}
But this may introduce additional issues around
search request caching
strange interactions with frequency, change detections && date histogram
What the user usually wants is:
When handling the initial batch of data (all historical), only worry about the last day
Then, take over when running continuously when data changes
It would be very useful for a new parameter to _start or something to indicate that the initial batch of the transform should be restricted to after some page of results, and then when running continuously don't worry about that filter any longer.
The text was updated successfully, but these errors were encountered:
We are simulating this feature with starting with a query for recent data and then will update the transform to relax the time to a much longer interval or no time query at all. It would be nice to have this as a core feature of transforms.
Description
A very common use case for transforms is pivoting historical data. But, when there is a large amount of past data, that data may not be very useful in the resulting index.
User's typically achieve this by adding a query filter to the transform:
But this may introduce additional issues around
What the user usually wants is:
It would be very useful for a new parameter to
_start
or something to indicate that the initial batch of the transform should be restricted to after some page of results, and then when running continuously don't worry about that filter any longer.The text was updated successfully, but these errors were encountered: