There was a bug causing misbehaviour of some sort labels settings in GoodData Writer. It is fixed by now and if you still experience this problem you just need to update model of the dataset from your writer (i.e. perform Upload Table or Upload Project action). We are sorry for the inconvenience.
Google Analytics Extractor now supports Segments.
"Segments allow you to select users and sessions to answer questions that are important to your business. For example, users that have completed at least 2 transactions with a lifetime revenue of over $1,000, or users who visited first on a mobile device followed by a desktop browser."
Read more about Segments here.
Redshift writer is the new addition to the growing list of writers. You can now write your data to any Redshift warehouse, or if your project includes Redshift, we can provision you with it's own database - that's useful when you need to perform read-only operations in an external application, such as chart.io.
Have an instant preview of the storage table right in place such as transformation mappings configuration, most of the extractors and writers configuration, job detail page etc. No need to switch between Storage Console in most of the cases.
Table id is displayed as a link:
Click on the link and get modal dialog pop up with general info about the table, its sample data, and the most recent events:
Keboola Connection now supports adding/removing the primary index on existing tables.
Some GoodData Writer loads may have failed with weird "Target model is invalid" errors in recent few hours. This had been caused by a bug in data modelling part of the application which we already fixed. We are sorry for any inconveniences, it should work as expected now.
There are some under the hood improvements to GoodData Writer which allow you to rename datasets already exported to GoodData and freely rename attributes and facts (to this moment GoodData Writer has been automatically adding name of dataset to attribute/fact title).
Also there is possibility to change default GoodData identifiers for datasets and their columns. This is suitable especially for projects migrated from CloudConnect, otherwise you won't need this feature probably.
This improvement is available immediately for new writers but existing ones will be migrated in several waves during upcoming days.
The current API version (2.3) our Facebook Ads Extractor uses will be deprecated on 7.10.2015.
We will update the extractor to current 2.4 version. Please review the changelog and see if it affects your counfiguration: https://developers.facebook.com/docs/marketing-api/changelog
If your extractor fails, or delivers wrong data after 30.9., please let us know and we'll investigate what caused the problem.
Amazon Web Services - our major backend cloud provider - announcing massive API error rate in their infrastructure
More details and updates can be found at their official status page http://status.aws.amazon.com/. AWS Status page archive is also here.
Thanks to our heavy dependency on Amazon AWS cloud, our Keboola Connection platform suffer by their errors, so please be patient, check AWS status page and keep your fingers crossed!
For more information, do not hesitate to contact us here in comments or by support@keboola.com.
UPDATE (2015-09-20 7pm CEST): Amazon API is back in business.
Media coverage - VentureBeat
We have introduced query limits for Redshift clusters to prevent deadlocks and keep the clusters in good shape:
- 5 queries in parallel (further queries will be waiting in a queue for 60 minutes)
- 60 minutes execution time per query
If the query time exceeds 60 minutes, it will be terminated with a user error "Query cancelled on user's request" and/or "An exception occurred while executing".
These limits will take place during the next maintenance window of each cluster (after it reboots).
If you need to change the limits, contact us at support@keboola.com.
EDIT September 22nd: Due to a high number of requests for a higher number of concurrent queries we've increased the limit from 2 to 5 concurrent queries on each cluster.