It doesn't work when creating an event trigger for the first time.
But, if I delete /blobs/, it will start.

"blobPathBeginsWith": "/XXXXX/blobs/YYYYY.csv",

change

”blobPathBeginsWith": "/XXXXX/YYYYY.csv”

is it Bug??
and , It is safe to insert again after deleting

dataset policy (see below) in azure data factory - can it be applied to azure data lake store? it currenyly only applies to Blob.

"policy":

{

"validation":

{

    "minimumSizeMB": 10.0

}

}

Typically if we use Union operation in SQL, it removes duplicate row.

But the equivalent transformation named as Union in the Azure Data Flow does not remove duplicate row.

Ideally there should be option to do so.

Documentation for VSTS integration with ADF v2 is missing one thing. The "Default" account isn't the "Microsoft Account" when using a MSDN subscription. So with each new MSDN account, there is a nice {GUID}.visualstudio.com address. Someone can create a new organization and specify a "vanity" address.

The second thing is there is a "Microsoft" account and a "Default" account. ADFv2 observes the "Default" account and doesn't even list the "Microsoft" account.

Once you navigate to the "Default" account and the {GUID}.visualstudio.com address, create a new project for ADFv2, then you are good to go.

Considering if developers are working on enterprise version, it is difficult on small workspace and limitations of web has it. Why not, have a ADF client version instead similar to Azure * Explorer which does lot more than online version (available through portal.azure)

User Wants to know Estimated execution time of pipeline. Do we have any predefined API's for this?

Now, when I do [Import ARM Template], it just only can deploy new Data Factory.
Please add function that can import arm template at current Data factory.

When a task is triggered via ADF, this gives a system generated ID which we cannot control. The current task history's default is sorted by task name. There should be a better way of sorting task history, ie., sort by rundate.

There is NO information on HOW ADF writes the activity, pipeline and trigger logs!!! I have run multiple jobs and the data is never consistent. Sometimes it writes that the activity has started, in progress but will NOT log the error! Sometimes, it just does not write data at ALL (unless it only logs data every x minutes which is DUMB if it is not near real time). I just wish I knew:
1) How it wrote the logs
2) common messgages for ALL executions (e.g. started, in progress, failure, success) and not have sporadic ones

The current email notification system is based on alerts and therefore you have to set the min/average/max/total occurrences of an event/events over a period. This means that an activation and resolution emails are generated when all I want is one simple email stating whether a trigger was successful or not.

Currently ADF can write to AzureDataLakeStorage as a e.g. CSV file.

Please create functionality so that ADF can write directly to AzureDataLakeAnalytics database. For example, if I am reading from a SQL table... I can write directly to an ADLA database. Without first dropping it as a CSV.

Let it be more flexible to changes in tables, pipelines, etc. Also, let us download all json files easily

I could not find the limitation of ADF V2 in its documentation. I found it on Github or in Express route documentaion.
It would be very usefull to have in the ADF documenation as section with at least a link to limitations

Currently ADF isn't realtime, can you please help make the sync realtime so if someone makes a change in one database it appears in another.