Sharepoint Sync (Input Data tool) Run on Server

[bertal34]

Has anyone gotten the Sharepoint Sync method to run successfully on Alteryx Server?  The sync method I followed is in the link below and uses the regular Input Data tool (not the Sharepoint Connector tools).  This method works great in Designer, however I wish I could get it working on Server.  I have found the Sharepoint connector tools to be inefficient and I cannot satisfy all my uses cases with these tools.  Using performance profiler, the SP connector takes 99% of processing time while the sync input is much more efficient.  Anyways, if someone could share in detail if they know how to use the Sync on Server, I would appreciate it.

https://knowledge.alteryx.com/index/s/article/How-To-Connect-to-Sharepoint-with-Alteryx

[caltang]

We have the same problem. I have been going back and forth with Alteryx on this, and the SharePoint tool has been making my Server life very difficult to scale. 

[bertal34]

@caltang 

Oh no, as Server admin for my organization, that's not what I wanted to hear.  We have not started to transition workflows to Sharepoint yet however I'm trying to learn and understand how to connect to Sharepoint and what pitfalls we might encounter.  Then relay those to other users and provide guidance on what tools to use in certain situations.

Can I ask what specifically about the SP tool is not working well with Server?

[caltang]

The main reason we are opting for SharePoint input/output connections is because the data does not persist in MongoDB.

Because of our current setup, we cannot directly connect with SharePoint - we have to use a Service Principal which involves Client IDs, Client Secret, and Tenant IDs. Now these credentials expire too, so that's an added layer of complexity. 

Since we ride on thes credentials for every individual to access SharePoint, the main problems I have seen (and have not been answered by experts yet):

1. If the developers build the workflow to be server-ready, end users who enter collections to run the workflows (assuming with interface), will need to have the file in SharePoint at the time of running to be ingressed automatically on Server side.
2. But if the developers built the workflow, wouldn't that mean that the SharePoint site is connected only if the Developer has access? What about new users who don't have access to that particular SharePoint site? It's all very confusing for me and I have yet to test it properly because of Point 3.
3. The SharePoint at some points in time fails to run, and sometimes it can run. So it fluctuates, sometimes the input runs then the output fails. Sometimes both fail. It's really clunky and unreliable to me, and I really wish we could just use the interface tools without this limitation. 

Now, the next bit of concern is that if your organization's setup is restrictive, then you may encounter issues like me. There's not one wide net solution to this as of yet. 

I've been thinking a lot about this as well, which prompted me to consider OneDrve connection instead - or even just build a SharePoint/OneDrive connector of my own via Python. I've had great success using Python to connect with Google BigQuery without the need of using the deprecated GBQ in/out tool and/or the normal Input tool/DB tools. The performance is better, and more flexible too.

[bertal34]

Wow, I can tell you are deep in the weeds with this subject!  Glad to hear you have success using Python with GBQ and the performance is better.

I'm still trying to figure out what the best Authentication method is for our company's setup.  Alteryx users want this to be a seamless authentication experience...not having to create DCM connections/credentials and wondering if each user is synced & shared to server properly.

OneDrive sync method seems like it would work for all users IF there was a way to reference a synced Sharepoint site through UNC instead of absolute ie: C:\Users\jalbert\SPName\SiteName - Documents\

[caltang]

Honestly it has been a nightmare for me to balance both building on designer with my current projects and to get the Server working with my limitations in place. 

I have been asking around everywhere and I just can’t get a clear answer or path forward. With regards to your storage of DCM credentials, I think the Server itself could be a place where you can list them out. Though I’ve not tested that myself. 

If you find anything or have a path forward, keep my in the loop! I’m digging deeper at every corner of the server to get it working to scale. Otherwise it would have been all for naught.

[bertal34]

FYI, I brought this up as a talking point for today's Dallas Ft Worth user group meeting.  Several ACE's will be on hand so I'm hoping to have good discussion.  Link below if you can make it.

https://community.alteryx.com/t5/Dallas-Ft-Worth-TX/DFW-Alteryx-User-Group-2024-Kickoff-Meeting/ec-p...

[tbuenaflor]

Bumping this thread since I've been pulling my hair on how to make Sharepoint connections work in Server as well. We have a connection string created to connect to a SharePoint site/page but I'm not clear on how to actually use this in our workflows. Do I have to create a separate connection string for each input and output file?