Is anyone else having the problem that they can’t save routines?
Someone I know asked me about a week ago if I could replicate the issue he’s having and as it turns out, I have the same problem.

Is anyone else having the problem that they can’t save routines?
Someone I know asked me about a week ago if I could replicate the issue he’s having and as it turns out, I have the same problem.

I tested and was able to
Hi, @Andreas_Roedl
Can you share the configuration you set for this routine?
I’ve seen this in the past, but I seem to remember it was when there was an issue with a custom capability.
I’m having an issue with Routines as well.
When I create a Routine and select a category, after pressing save, it will eventually time out with a “There was a problem connecting” error. Despite the error, the Routine was actually created, but not placed in the proper category. It winds up in the “All” category.
If I try to move the Routine into another category using the Category menu from Routines, it also times out and the Routine is not moved.
@nayelyz One that fails is IF a Virtual Switch turns on, THEN play a message on a Speaker, Delay 5 Seconds, then turn off the same Virtual Switch.
Another is IF an Aqara FP2 detects presence and the thermostat is set to Cool, THEN turn off and Inovelli Fan Switch.
There are others, but these two are a representative sample.
Very simple routines with very simple production capabilities, like a Matter on/off switch and nothing else.
This seems to happen when I choose a category. The other user reported the same: he tried to give his existing routines categories.
Now I have tried to create a routine without a category and that worked, tried it again and it didn’t. I have to do more tests to figure out the exact combinations of those that (sometimes) work and those that always fail to save.
All I know for certain is that I’m not alone. The other user is also German and maybe it has something to do with Umlauts in the list of categories?
Now I remember that this is what the other user reported about a week ago.
I didn’t had the time to investigate it further and today I found out that I’m having the same issue.
I can generate an app log later today or tomorrow.
I created a category with these symbols (¨) and I cannot replicate the issue yet.
So, please check if you observe anything else and
@nayelyz I’m in the US, so English with no Umlauts. I agree that it may be related to the Category. I was able to create a similar Routines to the ones that failed, but I started in the All category on the Routines page and did NOT select a category. However, once the Routine was created, I tried moving it to a category using the Category menu and that failed as before.
Where have all the icons gone? Do they disappear when you hit the Save?
Good question…
Do they disappear when you hit the Save?
Sometimes.
I can create a routine.
Creating a category either way fails.
Done.
[more characters]
I think it would be a good idea to do the same as mentioned here:
From the app log:
05-21 13:01:10.602 26555 26862 E [+0200]oneconnect[1.8.45.24_01][MAIN_UI]: [ATM]AutomationTitleViewModel.create - error create rule to cloud: 500 (Path: /categories/****)
05-21 13:01:10.607 26555 26555 E [+0200]oneconnect[1.8.45.24_01][MAIN_UI]: [ATM]AutomationTitleFragment.onErrorState - [SmartClientError] HTTP 500
05-21 13:01:10.607 26555 26555 E [+0200]oneconnect[1.8.45.24_01][MAIN_UI]: @PUT https://client.smartthings.com/categories/0e08d1b1-ff90-42e7-a28e-3c977ac1a6b3?locationId=10cea42f-b555-xxxx-xxxx-xxxxxxxxx
05-21 13:01:10.607 26555 26555 E [+0200]oneconnect[1.8.45.24_01][MAIN_UI]: --> Request Body : {"name":"Belüftung","categoryType":"ROUTINE","rules":["1584b69c-485e-43f1-b60c-85cdb801f532","42f998c8-80b2-4530-808c-763c340d925d","5ae4bd78-53e1-4355-a040-87d209bf56e1","8f1aa228-bbee-44b4-a213-0847dc6b5f4f","b06380ce-6312-44f5-8807-3167a47b5c91","c5d3cbb8-8daf-49b0-95ed-7ee083baea82","d092b03f-365a-47df-a70c-b23942be511c","e8e81798-30c7-4150-8cde-7fc7c5b47686","521571b0-df1f-42bd-b39a-8848d5be89b6"],"scenes":["50fce808-ba3a-4772-90fc-3f5ad6a4da46"]}
05-21 13:01:10.607 26555 26555 E [+0200]oneconnect[1.8.45.24_01][MAIN_UI]: --> Correlation Id : 8c8a5eb3-ec3e-4a98-a797-xxxxxxxxx
05-21 13:01:10.607 26555 26555 E [+0200]oneconnect[1.8.45.24_01][MAIN_UI]: <-- Response Body : {"requestId":"8c8a5eb3-ec3e-4a98-a797-540de09661ea","error":{"code":"UnexpectedError","target":null,"message":"A non-recoverable error condition occurred.","details":[]}}
Screen capture and logs sent.
Looks like the same error as @Andreas_Roedl
05-21 11:16:47.967 29186 29186 E [-0400]oneconnect[1.8.45.24_01][MAIN_UI]: → Request Body : {“name”:“Turn Off Fans”,“categoryType”:“ROUTINE”,“rules”:[“28c8a23b-3484-49cd-a599-6baae9c1c999”,“a91462cf-bb01-4525-a3cd-a58df7249050”,“c8d63a14-90e8-4dc0-881a-c0234a77960b”,“d25c3bbe-d6b3-427b-b3e6-60440c388675”,“e9f43129-abbc-4648-9bf0-7e23eb3ec70e”,“f5ace350-6510-47ef-b8b8-fd3ce97c5fb0”,“a763f75e-80ad-4cfb-892e-ef5cfb17c4c6”],“scenes”:}
05-21 11:16:47.967 29186 29186 E [-0400]oneconnect[1.8.45.24_01][MAIN_UI]: → Correlation Id : aea99cab-6861-43c3-b9eb-723280b149bf
05-21 11:16:47.967 29186 29186 E [-0400]oneconnect[1.8.45.24_01][MAIN_UI]: ← Response Body : {“requestId”:“aea99cab-6861-43c3-b9eb-723280b149bf”,“error”:{“code”:“UnexpectedError”,“target”:null,“message”:“A non-recoverable error condition occurred.”,“details”:}}
The logs make this look less like a Routine problem and more like a Routine Category problem.
The failing request is:
PUT https://client.smartthings.com/categories/{categoryId}?locationId=...
The backend returns:
HTTP 500
UnexpectedError
A non-recoverable error condition occurred.
The request body is not the Routine itself, but the category object containing categoryType: "ROUTINE" plus arrays of Rule IDs and Scene IDs. So the Routine can be created, but the follow-up category update fails. That would also explain why the Routine may appear under All, but not in the selected category.
The same pattern has now been reproduced by users in different regions and without Umlauts, so this does not look like a German/encoding issue.
One additional observation from my side: for about one or two weeks, I have also been seeing broader SmartThings cloud weirdness in the EU region. Once a day, all of my devices appear to go offline - including cloud-connected cameras - and then they come back online one by one a few minutes later.
That may be unrelated, but together with this /categories HTTP 500 it makes me wonder whether there is currently an EU cloud/backend stability issue or some account/location synchronization problem that is not visible on the public status page.
The public status page currently shows EMEA as operational and lists no incidents for the last days, so I am not claiming this is a confirmed outage. But the behavior is consistent enough that it might be worth checking backend logs for the affected location/account, not just the Android app flow.
For now, the safest workaround seems to be: create Routines without assigning a category and avoid moving existing Routines between categories until SmartThings fixes or explains the /categories update failure.
“UnexpectedError”
It wasn’t the error we were expecting ![]()
Thank you for the information, @Andreas_Roedl and @Bry.
As I mentioned over email. We created the corresponding report. We’ll keep you updated here.