Yes, it's ye old
"Internal error encountered. Please try again"
error, rich in meaning! I've read what posts I could find on this forum, and elsewhere, to no avail.
I'm new to GCP Vertex AI, and just tried to load a small dataset of 1k rows, for learning. (Yes, I know that that is only a minimal set.) The 'Generate Statictics' steps succeeded, and it proceeded to begin. Lingering a few minutes, without any substantive feedback, it then stops with the above error message. I've retried many times, tweaking what little I could. No, there are no 'date' or type or column issues; all of that is kosher.
No logs are shown in the main window, even when I click on a node. The only semblance of a log is on the RHS (right-hand side), at the
Debugging Info > View pipeline proto
link. All that I could decipher is that it seems to be installing some python package before it fails. But, I'm not placing any bets on that interpretation!
Yes, I ran the curl command which
https://cloud.google.com/vertex-ai/docs/general/troubleshooting
suggests, to no avail.
Next, I tried to pay for Support. That ran into a need for "organization resources", Google Workspace, Cloud Identity. Clearly, Google is unclear on the concept of making purchases pain-free!
The above is written as only a record. If I find something else, I'll try to update this.
(BTW, I'd read a search result which'd suggested that the cause might be a discrepancy in the storage location of the dataset. In my case, I'd sepecified a storage location different from the default. But, a new dataset, at the default location did not help.)
LinkedIn
Twitter
