Attended RPA PAD flows triggered from the cloud have been working for us most of the time, but occasionally they fail for no apparent reason.
We get the green status bar icon that shows a flow is queued, but the flow never starts. Additionally, we have seen that the File System connector which requires a gateway instance is also breaking. It would seem that these failures coincide but I can't say for sure.
With the File System connector in a cloud flow, the error we get is as follows:
{
"status": 404,
"message": "Could not connect to your data gateway. Make sure your file system is active with the gateway running. The detailed error is 'Gateway on endpoint '...' is unreachable.'.\r\nclientRequestId: ...",
"error": {
"message": "Could not connect to your data gateway. Make sure your file system is active with the gateway running. The detailed error is 'Gateway on endpoint '...' is unreachable.'."
},
"source": "filesystem-eus.azconn-eus-01.p.azurewebsites.net"
}
Because we have a timeout set, the PAD flows triggered from the cloud fail with:
ActionTimedOut. The workflow action 'Run_a_flow_built_with_Power_Automate_Desktop' timed out while waiting for underlying operation to complete.
However the timeout is not the problem. If we did away with the timeout, the flow would apparently just never start at all.
This is extremely frustrating.
For reference, our PAD flows run in a Hyper-V Windows 10 Pro VM. Not sure if there could be some setting in the VM that could cause intermittent issues, or possibly our router/firewall settings could have some effect that only causes errors occasionally. But since it works maybe 80 or 90 percent of the time, it would be hard to find some local reason for these issues.
We assume Microsoft just sucks. We won't keep paying for our subscription if it continues to suck.
