Yes, if you have a Server license - you could create a workflow that will refresh data from the source every x hours/minutes and keep doing it until data is returned successfully.

Hi Paola, thanks for a speedy reply - do you have an example please?

Matt

Hi Matt, there is no ready-made example, but we could re-create the pattern.
The question is what the routine should look like, as we need to create a loop, that will refresh and try to retrieve the data from the API in regular intervals.

What happens once the data is extracted? Should the refreshing and extraction attempts stop?

Hi Matt,

The DoubleClick connector should not be failing. What version of Omniscope are you using?

Hi guys -

Chris, the connector doesn't fail - very occasionally our pull from DoubleClick can experience connectivity issues, this is more than likely something on our end. When it does fail it causes us to miss that data from our processes.

Paola, as above - if we keep with the DoubleClick example - we would need the workflow to loop back around if the feed does not download new data each day. Looping around say 3 times before sending an error email us by way of notification. If we set up one example we can apply this to all our data feeds ( we have many across many different projects).

Matt

Are you directly using the connector to configure the report? It might help for consistency to save the report onto the DoubleClick platform (there's an option in the Omniscope to save the report settings and give it a name). From their platform, refresh the report on a schedule (find the saved report on the DoubleClick web tool and just edit/configure there), then set another connector option to call a pre-loaded report reducing the time it takes to obtain the report. Reports executed on DoubleClick side will appear to be pre-loaded so won't have timeouts issues as it is on their servers. You can alternatively extract/download the report in Excel/csv from Doubleclick too if you set it to send it out via email and build an alternative block in Omniscope to read your emails and load the files it sends out.

These are just some other ways to explore, my past experience with the DoubleClick API was that the API would face high traffic periods making it difficult to request a report at certain times of the day. Basically if DoubleClick is under high traffic, the connector request on a new report will often timeout as it puts you on a queue (This I suspect is due to the increase in the API usage for other applications that shares the API for programmatic purchasing or other companies pulling big data extracts for their large media campaigns etc). It is worth considering not relying on the API as the demand for it seemed to be increasing and down times was getting more frequent. Please note I have not used the DoubleClick connector in over a year so things may have changed!

There are also other advantages to schedule on DoubleClick side, configuring the report on DoubleClick platform there is a few more options for different types of reports which the Omniscope API cannot access such as cross device reporting and more custom options and drill downs by different splices. These are basically the newer reports and API improvements they make incrementally, You can better manage large data requests if you are pulling data for a longer period than a year. You can also set the schedule on DoubleClick side to be as frequent as you want without waiting for Omniscope to initiate the report refresh.

Hi Daniel - we create the reports within the DoubleClick UI. My question wasn't really related to a complaint about the DoubleClick API per se, it was more to retry failed API feeds - DC was my example. The chances of our internal server having connectivity or memory issues is by far the biggest threat to our automation.

If anyone could point me in the direction of some basic documentation relating to server workflows I can set this up to try and try again in the event of failure.

Thanks,
M

http://kb.visokio.com/scheduler-help

It just really using the scheduler and picking "File Action" to open your iok file,
then "Refresh from Source" and save/publish.

You then pick the times and schedule for when to try your 3 runs.

You can use the delay tick box to send out emails, if the file fails to refresh and complete those file actions after a certain amount of time.