Cube.js: Option to renew queries in caching layer every defined interval period

[psharma0789]

Hi,

Is there an option to for a query to renew in the caching layer in the background by a defined interval. For example: the system would look at all cached queries and automatically renew every hour, day, etc. This would keep data fresh in the cache.

[vasilev-alex]

hey @psharma0789! I believe you're looking for https://cube.dev/docs/caching#keeping-cache-up-to-date

[psharma0789]

Hi @vasilev-alex ! This only seems to be for pre aggregations, correct?

[jcw-]

This only seems to be for pre aggregations, correct?

Nope! The CRON job technique in the first paragraph would also keep the redis cache fresh as well.

[vijsai]

@jcw- I think that the CRON job needs to execute the queries as if the frontend calls them with "renewQuery: true" flag. Is there any example available? I searched enough and couldn't find a solution.

[jcw-]

That's right - you would mimic the same queries that your dashboard is making, and include renewQuery: true, as documented here: https://cube.dev/docs/query-format#query-properties

An example of making a query is here: https://cube.dev/docs/@cubejs-client-core#cubejs-api-load

A more specific version of that would be something like this - note that this example is using the client-side library, you'd probably want to do something similar from server-side Node code on a cron schedule, but conceptually this is what is being suggested I believe:

import cubejs from '@cubejs-client/core';

const cubejsApi = cubejs('CUBEJS_TOKEN');

const resultSet = await cubejsApi.load({
 measures: ['Stories.count'],
 timeDimensions: [{
   dimension: 'Stories.time',
   dateRange: ['2015-01-01', '2015-12-31'],
   granularity: 'month'
  }],
 renewQuery: true
});