Alteryx Server Ideas

We have at the moment our server configured to work with SAML credentials. Whenever a user sets the credentials option to be "User must specify own credentials" the validation is performed using the servers own local user, giving out an error whenever some workflows with SSO data access are uploaded.

It would be great if a window would pop up or credentials could be configured for the validation to aviod seeing the error to show up.

Hi,

The schedules page on server needs an overhaul.  Sorry to be so blunt but it is awful.  It is by far the worst aspect of Alteryx and one of the most important at the same time.

Here are a few points about it...

1.) Sorting needs to be fixed as it doesn't work correctly (sorting on the first column, "Workflow", returns items that start with 'L', 'S' and a zero as the first three on the list), so looking for 1 scheduled item across 10+ pages of schedules becomes a "needle in a haystack" endeavor.

2.) There are no search and/or filter capabilities which would help alleviate the pain-point mentioned above.

3.) [Missing] items automatically removed from schedule when the workflow is removed from the server.  It just jumbles up the screen with useless information.

4.) The 'Schedule' column doesn't need to reflect the name again like the 'Workflow' column. is a hyperlink and the other isnt.  I cant tell you how many times I have clicked on the wrong hyperlink only to have to go back and deal with point #1 all over again.  Can't it just be a simple calendar icon or something like that next to the Workflow name?

5.) While not as critical, but more of a 'nice to have' function.  An ability to export the list of schedules would be helpful.  If we ever need to divy up work for an employee that leaves the company it would be a good way to see all of the work they have out there and mark up the list to distribute info for reassignment purposes, etc. 

I am sure there are other aspects I am missing, but here are the most important ones to me as an end-user (Not an Admin)

Thanks for listening.  🙂

The idea is to have tabs on separate pages in the Gallery (or the option to allow that type of behavior).  Right now, if you created an analytic app that has questions on different tabs they will be on separate pages in Designer, but when you push the workflow to the Gallery all the questions will be on the same page. The tabs act as a navigation bar instead of taking you to separate pages. 

Here's two different discussions that ask about this in case there is any confusion:

https://community.alteryx.com/t5/Alteryx-Server-Discussions/Interface-Designer-with-TABS-in-the-serv...

https://community.alteryx.com/t5/Alteryx-Server-Discussions/Analytic-App-Tabs-in-Gallery/m-p/846477/...

As the title says: please make the date modified data more granular e.g. yyyy-mm-dd HH:MM:SS and not 6 months ago

The current behaviour is the same in designer & browser, but is more important in browser (first screenshot) as there isn't an indicator or method or sorting the order.

System Administrator should be able to configure portal pages be being able to choose the columns, columns widths, and sort order.  For Example  in the collections view,  there are columns Inight Count, User Count, Studio count which add no value to my user community.  Yet the name column is not wide enought to show the full names. The collect page is sorted by Create Date which does not add value and once you get above 20 collections  

The recommendation would be to allow System Admins to be able to change these settings.  If the definations are stored a json or xml file or some other file that can be edited, allowing for the customizations.  Admins would edit at their own risk working with in parameters and documentation provided. 

Currently it's impossible to tell which worker a job is running on. Add a column to the jobs view that shows what worker a job is currently running on.

Currently, if we want to assign a worker node it has to be done at the workflow level.  It would be very helpful to be able to make a rule that auto-assigns the type of workflows to a specific node, so all apps go to one, all schedules go to another, etc.  

At the very least, if there was an API endpoint to update specific items on a workflow (such as node), this would work as well as we could then build a process that auto-assigns.

Thanks! 

The gallery shouldn't allow to create data connections with already existing names.

For example, in designer it's difficult to choose which data connection we want to work with if they have the same name, please verify the images in attachment.

Currently, Alteryx Server lacks rich job scheduling functionality to other products like Control-M and Informatica's Scheduler

Feature Requests:

1) Ability to configure job retries on failure without having to rerun the workflow based on a  event trigger: https://community.alteryx.com/t5/Alteryx-Designer-Knowledge-Base/How-to-run-a-workflow-from-an-event...

There are common occurrences where a workflow may fail to connect to Input sources and a retry would normally resolve the issue. 

2) More user friendly way to setup dependencies between scheduled workflows.  For example, if workflow A fails it triggers workflow B to run.  If workflow B runs without errors, it triggers workflow C to run.  Currently, we would need to configure this based upon events.  A complex chain of scheduled workflows becomes difficult to manage and scale.

I would suggest a service split to simplify maintenance actions in the Alteryx Server.

Split Alteryx Service into 5 services:

 Alteryx controller

• Dependencies: Alteryx Database
• Purpose: Controls job submissions
• Meaning: The Alteryx controller Service stopped means neither the users or the scheduler can submit new jobs

Alteryx Database

• Dependencies: N/A
• Purpose: Database
• Meaning: The Alteryx Database Service stopped means all the platform is down

Alteryx Gallery

• Dependencies: Alteryx Database
• Purpose: Controls the Web UI
• Meaning: Alteryx Gallery Service stopped means the gallery web page is down

Alteryx Scheduler

• Dependencies: Alteryx controller, Alteryx Database
• Purpose: Controls the scheduled jobs
• Meaning: The Alteryx Scheduler Service stopped means the users can still submit jobs manually but scheduled jobs wont be submited

Alteryx Worker

• Dependencies: Alteryx controller, Alteryx Database (only applicable if the server where the Alteryx Worker Service is running the Alteryx controller)
• Purpose: Executes the jobs
• Meaning: The Alteryx Worker Service stopped means no job will be executed on the node

This service split will help perform some maintenance tasks and enables the Platform Administrators to shutdown the platform correctly shutting down the services in the correct order.

Usage Example 1 (Cold Backup):

• Stop the controller so no more jobs can be submitted
• Stop all the workers
• After all the workers have stopped the Alteryx Database Service and the rest of the remaining services can be shutdown
• Preform the Cold Backup
• Start the services

Usage Example 2 (Changes in the worker configuration - By doing this the users will still have access to the gallery and their jobs):

• Make changes in the worker configuration
• Restart only the worker service

Usage Example 3 (Changes in the Gallery Page or Configurations):

• Stop the Gallery Service
• Make the changes
• Start the Gallery Service

One of our Server customers has a need to disable all schedules temporarily across all worker nodes.  I do not see an easy way to do this unless we go into each worker's settings and disable from there, but this is not ideal.

The reason is for scheduled maintenance windows during a large upgrade or migration.  Currently stopping the Alteryx Service on each worker is time consuming.

Thank you! 

More of our customers (we are a premier partner) have Servers with multiple worker nodes.  If a custom macro is needed, currently it has to be added manually on each worker node. 

Would love to see a feature in the Admin portal that allows for the upload and distribution of custom macros to all or select workers. 

currently, there isn't any way to fully delete (safely) a user from the gallery. So if there is a previous client or previous employee in gallery, we cannot clean up the gallery settings - only "disable". It would be great to have a "trash can" symbol next to the users where we can remove them from the server and keep all of the users clean. 

The Gallery currently only enables scheduling of the latest version of a workflow, regardless of the published version - for instance, if there are two versions of a workflow and Version 1 is the published version, only Version 2 can be scheduled since it is more recent. There is no straightforward way to schedule Version 1.

Currently two workarounds are to:

1) download the older (published) version, export to Designer, and then save as an entirely new workflow in Gallery.

2) re-save the local Designer file of the older (published) workflow as the most recent version in Gallery, for example re-saving Version 1 as Version 3 so that "Version 1" can be scheduled.

The ability to publish and schedule older workflow versions would be a helpful new feature to add to Gallery, as this would avoid the need to re-create or duplicate workflows simply because the most recent version has an unexpected issue that the previous version does not have.

When a render tool is run on Designer pointed to a temporary file the log gives you a clickable link to access the output.  When deployed to the Gallery unless you have given it a specific output destination with %temp% in the filename you get... nothing.  

What did I expect?  The pdf or other temp file to render on the screen or at least download for me.  Or give me a dropdown selection of the various files render could create for me.

When a user publishes and schedules a workflow on the gallery, they have the option to specify their time zone to ensure that the execution of the workflow occurs at the appropriate time. However, if the workflow invokes the DateTimeNow() function, the captured time is in UTC. If it is important for the end-user to have that time that is captured be in their local time zone, they will need to manually (either within the workflow or as a user input for an Analytic App) specify the offset from UTC.

Instead, there should be an Engine variable that captures the offset (or the time zone) that the workflow was executed from (or scheduled in), which can dynamically adjust the DateTimeNow() output appropriately.

The current Gallery API executes a job until it reaches a completion state with either success or failure.  Once reaching that state, the callout to /v1/jobs/{jobId}/ will return an array of messages that can then be parsed to gather information about the execution of the job, similar to the output you see in designer or via the CLI when running a job directly.

It would be nice if the callout to the jobs endpoint would constantly update and return the messages array with each callout.  With jobs that take longer to execute, our Gallery API wrapper could then report back to us and let us know what messages have come out as the jobs runs.  For example, if a process takes 2 hours to execute, we might want to know how far along the job is or if it has met a special condition that will cause it to run longer, etc.  To do that, we could add in message tools that report info messages to us.  Unfortunately, these messages are ineffective if we are running via the API because we won't see them until the job has fully completed.

Very Simple Request.

Just like how you can search for a collection, we would like to be able to search for a schedule. One team has over 5 pages of schedules and it would be nice for them to search for one to update vs having to scroll through 5 pages to find it

Currently, we have the functionality to set a recurring schedule with daily frequency with an option to 'Run only during work week (Mon-Fri)' I would like to add this 'Run only during work week (Mon-Fri)' option to the Hourly frequency interface. 

No current option to run on weekdays only: