As we announced earlier, we have decided to split the current Google Drive Writer into two separate components to simplify their usage:

• Google Sheets Writer - as its name implies, is designed to upload tabular data from Storage to Google Sheets.

• (New) Google Drive Writer - will handle uploading general files to Google Drive.

Google Sheets Writer has already been available for some time and now we are proud to release the new version of Google Drive Writer.

The Google Drive Writer is best for uploading general files into Drive. It allows you to create a new file on each run and therefore is ideal for backup purposes.

Features

• Advanced Input Mapping
• choose which columns will be uploaded
• filter data by date or by column values

• Update existing files or create new files

Migration

The existing Google Drive Writer will be deprecated soon and a migration tool will be made available to help you transfer your existing configurations to new Google Drive or Google Sheets Writer respectively.

Don't hesitate to give us feedback or ask a question, write to support@keboola.com.

RStudio and Jupyter Sandboxes

• HTTPS Support.
• Configurable (pay as you go) memory and disk limits.
• Creating better UI and integration with transformations.
• Allowing tables and files to be added to a running sandbox.

Keboola Connection in EU

We have launched Keboola Connection in a new region outside of the USA. You can now create your projects in the EU-Frankfurt region, please contact support@keboola.com if you are interested. This new region is now in beta.

Minor improvements

• Improved performance of PostgreSQL extractor.  
  

Friday June 23rd, 20:30 PST

We're experiencing a large number of application errors when loading events from Storage in the UI. 

We're trying to debug and fix this issue, no other operations should be affected.

Friday June 23rd, 20:48 PST

We have rolled out new servers which seems to mitigate the issue. All operations back to normal.

We're sorry for this inconvenience.

Today (Jun 21, 2017) between 16:00 and 18:30 CEST all Redshift Writer jobs created in this period failed with an application error. We have identified and fixed the issue.

We're sorry for this inconvenience.

Updated Components

• GoodData Writer supports login parameter in the Create User API call to specify different mailing address
• GoodData Writer supports full parameter in the Synchronize Dataset API call to perform full sync without data preservation
• DB2 Extractor supports SSH tunneling

New Features

• Documentation to generate Google OAuth 2.0 tokens for Generic Extractor including token refreshing
  

Community News

• ZOOT developed extractor to download data from CJAffiliate’s Commission detail service API
  

  

Minor Improvements

• Storage API Client for R supports incremental writer to Storage

Fixes

• Snowflake Extractor supports large files
  

We released a new AdWords Extractor updated to use the latest version of AdWords API v201705. The previous Extractor is marked as deprecated. It uses API v201607 which will be switched off by June 27. You are encouraged to migrate soon during the June. 

Configuration migration is automatic but you need to rewrite fetching of deprecated values yourself according to the migration guide (https://developers.google.com/adwords/api/docs/guides/migration/v201705). AdWords developer token must be entered to the configuration again because it will be encrypted for this app only. You also have to reauthorize OAuth access to your data.

Recently (this week) we've begun hitting request quota limits with our Google Analytics Extractor.  The process of increasing the quota has begun, but until that happens we will unfortunately be required to put a hard cap on the extractor runtime of 30 minutes.  This means that only extractors that finish their jobs in under 30 minutes will complete successfully.

If you are affected by this limit, please try adjusting your configuration to reduce the amount of data being extracted.

Sorry for this inconvenience, we'll update here when/if the quota gets increased and we can remove this limitation.

Some database extractor timeout errors have occurred in a few projects due to yesterday's switch to static IPs. 

We have temporarily reverted back from from new Fixed IPs as we attempt to resolve these issues.

We will post an update when the fixed IPs return and the timeout issues resolved.
 

We're updating datatype handling in the Transformations UI and backend. The update features datatype validation and new datatype configuration format. 

In rare cases errors may occur. If it happens in your project, please get in touch with us at support@keboola.com. 

Thanks for your understanding.

It is our pleasure to announce that as of today, all our outgoing network connections in the US region are using a pool of four static IP addresses. 

This can help you meet your company security standards. To find out more, please visit the IP Addresses page in our documentation.

Be aware that IP addresses can change in the future. For your convenience, you can programatically fetch and parse the list of existing IP addresses in JSON format at https://help.keboola.com/extractors/ip-addresses/kbc-public-ip.json. 

PLEASE NOTE: 

Some of you were employing a very old concept where we performed an on-demand network source routing which allowed us to force the source IP under syrup-out.keboola.com. This was deprecated almost a year ago. Today, we are also announcing that old source routing is deprecated and will be turned off at the end of this month. If you rely on source IP, please move all your existing firewall rules to our new addresses before June 30, 2017.