Alteryx Server Ideas

I know the Alteryx server supports auto-scaling up and out as of now. Alteryx Servers should also support auto-scaling down and in.  

Links included in the notification emails sent to users (e.g. links to collections/workflows in our Private Gallery) or copied from the browser address bar only works for public workflows or for workflows in your private studio.

If the workflow is in a collection shared with you, clicking on the link takes you to what seems to be the gallery page for the workflow asking you to sign in, but after you sign-in you are redirected to the gallery home page and not back to the workflow page.

The same occurs regardless if you're already signed-in or not.

This was related to support case: 00352021

Alteryx server APIs are associated with Subscription ( private or shared studio) and an Artisan could be part a of either a private or shared studio. In an enterprise environment you want the ability of scheduling workflows from private studio ( automation of private workflows) and shared studio ( production automation). You can achieve both in an enterprise environment with governance and control if Artisans ( along with Curators) have the ability to switch subscriptions. This will enable them to publish workflow in their private studio for automating personal workflows and in shared studio for production automation. In this way admin or curator still control the data connection creation and assignment , run as credentials but switching subscriptions can be provided to Artisans as well.

Shared studio's private api key and secret can be setup in other scheduling systems such as Ctrl-M and individual leaving or moving departments will not have any impact  on the scheduling configuration.

At present Gallery API events are captured without User details..

Eg:

if an Admin user changes the "Default permissions" under "Permissions" this will be captured as a galley API event in server Gallery logs..
but it will not capture the user details, who has changed the same..

Looking forward to a feature which can accommodate this..

In Gallery, we can see the workflow run details for a scheduled workflow that just ran, and the only way to dive into these details is to query the database.  We're requesting these details being changed to a hyperlink that sends the administrator to the information it is referencing.

Currently, workflow history is available to specific user only but cannot see workflow history from other users even they are in same studio.

For example, in Private studio "MAIN_STD", user A to run workflow “WF1”. User A can see the running history while user B cannot see.

The workflow execution log file is locked while execution is in progress. Can we look at making the log file available in read only mode while execution is in progress. This will to triage any performance issues on run time and track the progress on run time, rather than waiting for completion of entire process.

I am leveraging the server as much as possible while working remote.  I usually use it mostly for scheduled workflows when I am in the office, but limitations come up with the efficiency of the VPN and poor bandwidth when I am remote.

A typical process is to save to server, validation cycle runs, then open the workflow on the server in a browser, click run.

An ideal would be a macro or option that would combine two steps; for example, instead of "Save to Server", "Save to Server and RUN".  At that point you can collect your workflow when it is down running. 

Hi, our analytics team has dozens of workflows saved to Gallery and scheduled to refresh at regular intervals--or at least we did up until this weekend when the scheduled refreshed terminated for some unknown reason.  We currently have a team investigating the cause. 

The last run for the workflows took place between 10 PM on Friday 3/6 until 2 AM Saturday 3/7.  We didn't notice the refreshes getting terminated until start of business Monday morning.  Immediately our internal customers started asking us why things were out of date and we quickly found the issue and we're now going back and refreshing the data and re-initiating the scheduled refreshes.  However, some of our workflows can't be retroactively restated, so we'll just have a gap in the data from this point on. 

Needless to say, this is unfortunate for our org, so I'm trying to think of ways to avoid it in the future.  Having a notification set to send an e-mail when a workflow errs is helpful only if the workflow gets kicked off to run to begin with.  However, this will not help in cases when something has gone wrong to prevent the running of the workflow to begin with.

What I think we need is a system to auto-generate an e-mail to a person/group whenever a schedule is termed for any reason whether if it's by a person actively terming it, or for any other reason.  Just as you get a confirmation e-mail whenever you term an e-mail subscription, getting a confirmation whenever a scheduled refresh is ended, would be extremely useful.  

Thanks, Kurt

Hi Team,

     Just like the workflow upload API, could you please provide an API end-point for deleting the workflow,Job results and it's related information. In a scenario we might use gallery for one time execution of workflow, once the output was generated there is no point to have some workflows in the Gallery hence in this scenario the Delete API helps as a clean up activity on the Gallery to avoid the Junk/Unused information (Workflow/Job/Outputs). 

regression

Judging by the Community, it seems like the Alteryx business model is to crowdsource documentation. Do the same for a QA checklist. The 2019.4 release of Server shipped with some major regressions, notably:

DE21143 Workflow events no sending Email

DE22751 32-bit OCI connections not working in 2019.4 release

The QA team should publish the QA checklist and allow feedback from the community. The entire point of the community is to make the users and the product better. Let the users see which scenarios are tested and add ideas here for the QA team to utilize.

When viewing the schedule viewer with more than the default per page workflow amount of 25 workflows, if you sort it only sorts in the 25 in the current page. I would expect it to first sort all workflows then show the first 25, and actually used the interface with this as my expectation for longer than i'd care to admit, leading to a large amount of confusion when i for some reason couldn't figure out why workflows that i scheduled were not showing up. Obviously this is down to user error and now that i understand it i don't have any issue, but at the same time, why would the sort not apply before population of the list?

Interested to know if i'm the only one who's been annoyed by this.

~ Clive

Currently we have built-in authentication for Alteryx Server.
In that state, the Default user role of Gallery is set to "Artisan",
If you sign up to create a new user and log in,
It will be registered as "Viewer".

I want the default user to be valid even with the built-in authentication.

Hi Alteryx,

When a staff leave the company, his/her AD account will be removed from the AD directory,

However, the user record is still exists in the Alteryx gallery. For example, still existing in the user list of data connection and the permission tab.

Can Alteryx provide the following features ?

1. Identify the user that do not exist in the AD directory anymore.

2. one button to remove all of the setting related to that user in the gallery.  

Best Regards,

Samuel

With Version 10 of Alteryx Server Gallery, if new user signup is disabled, people see the following when clicking the sign in button:

If someone tries to fill out the "Join Now" section they then get:

In Alteryx Server Version 9.5, when new user sign up was disabled, the join now section of the prompt was not present. This is our prefered behavior. Another option would be to change the "Access denied" message to something more descriptive such as "New user sign up disabled."

Thanks,

Ryan

I'm really loving the new data connections for 11.0. We have deployed them in a private gallery for our users and it's great. The only drawback is that the server itself cannot easily be configured to use these same credentials. I would like some way for the service account running our jobs on our server to use the data connections from the gallery. I can assign the credentials to that service account just fine, but it never picks them up since it never open up the Alteryx GUI when it's running jobs. This would allow us to have one spot with all of the credentials. Currently, we're going to have 2 locations with all of our credentials that we'll have to make sure stays in sync: 1 in the gallery for the users, and 1 on the sever box for the server.

Thanks!

Are there any plans to add support for the Spatial Object, File Browse and Tree questions for the Gallery REST API?

I see via the Interface Viewer in Designer that the "answers"/values to these questions are just (messy) XML I realize the File Browse could be difficult because it would involve accommodating some type of BLOB data but the Spatial Object (one or more lat/lngs describing a point or shape) and the Tree questions seems to be fairly straightforward.

We currently get around the Spatial Object limitation by specifying a latitude and longitude (for a Point) in text boxes. For shapes (polygons) we just stuff the WKT (Well Known Text) representation of the polygon in a text box. This works fine but requires special handling on the workflow side. 

I have a chained app that is running on the gallery/server. When I run it, I see status changing from Processing to Running, but the window doesn't show a meaningful message as to which app is running. Is it possible to show that in the Browser window which reflects what stage is currently running in the Chained App? 

See snapshot of window below.

This would be super useful to monitor the progress of the chained app currently executing on the server. 

It would be helpful to be able to customize the color scheme in Alteryx gallery to differentiate between the Production and Sandbox environment. This would be a quick check to ensure you are in the correct environment.