1

I have run mlrun by docker-compose. But when I finished the sample training, I couldn't see the results graph on the management pages. I get 500 error and the details likes:

{
 "data": {
   "detail": "AttributeError(\"'K8sHelper' object has no attribute 'v1api'\")"
 },
 "status": 500,
 "statusText": "Internal Server Error",
 "headers": {
   "connection": "keep-alive",
   "content-length": "76",
   "content-type": "application/json",
   "date": "Fri, 07 Apr 2023 09:45:51 GMT",
   "server": "nginx/1.21.6"
 },
 "config": {
   "url": "projects/quick-tutorial-root/files",
   "method": "get",
   "headers": {
     "Accept": "application/json, text/plain, */*"
   },
   "params": {
     "path": "/auto-trainer-train/0/confusion-matrix.html"
   },
   "baseURL": "/mlrun/api/v1",
   "transformRequest": [
     null
   ],
   "transformResponse": [
     null
   ],
   "timeout": 0,
   "xsrfCookieName": "XSRF-TOKEN",
   "xsrfHeaderName": "X-XSRF-TOKEN",
   "maxContentLength": -1,
   "maxBodyLength": -1,
   "transitional": {
     "silentJSONParsing": true,
     "forcedJSONParsing": true,
     "clarifyTimeoutError": false
   }
 },
 "request": {}
}

I do not find any info for this question, please help me, thanks!

JIST
  • 1,139
  • 2
  • 8
  • 30
sequoia00
  • 111
  • 4

1 Answers1

1

Yes, I have found a method to solve this problem by myself. This problem is due to k8s config conflict. So I can change the code in the container of mlrun-api and then commit a new image. Now I add after the 66-line code in mlrun/api/api/endpoints/files.py with "use_secrets= False" temporarily. The show is on UI normally.

sequoia00
  • 111
  • 4