Cloud Run domain mapping issue

Google Cloud Serverless Applications
1

Hi,

For almost 48 hours I have been trying to add subdomains to a cloud run service (which works very well).

I was able to add the main domain without any problems, it took 10-15 minutes to take effect. However, for the subdomains I have no results. The addition remains blocked at the step “Waiting for certificate provisioning. You must configure your DNS records for certificates issuance too begin”.

The service is based on “europe-west4”, so compatible with the mapping and for the anecdote I was able to make the subdomain alliancefrancophone.frenchparadise.net work via load balancing, but I would rather use the mapping via Cloud Run.

Yet by checking on the DNS Checker site:

  1. Main domain (URL Removed by Staff)
  • Everything is OK
  1. Domain (URL Removed by Staff)
  • Everything is OK
  1. Subdomain (production) (URLs Removed by Staff)
  • Some zones are in errors (Canada, Mexico, Australia, China, India, Ireland)

Google Admin Toolbox Dig where everything seems OK

The DNS settings are as follows:

Capture d’écran 2024-09-17 131451.png
Capture d’écran 2024-09-17 131451.png539×543 27.1 KB

With all this I do not understand where the problem is, could someone enlighten me?

2

Hi!

The problem is still ongoing :/. The domain mapping request for alliancefrancophone.frenchparadise.net having exceeded 24 hours, I deleted it then waited about ten minutes before doing it again (40 minutes ago at the time of writing this topic).

Since then, I have looked through the service logs and found these lines

{
  "insertId": "66000a0725afv811",
  "httpRequest": {
    "requestMethod": "GET",
    "requestUrl": "http://alliancefrancophone.frenchparadise.net/.well-known/acme-challenge/pN18DY55gHhJA_92BPFivahG4wB",
    "requestSize": "337",
    "status": 302,
    "userAgent": "Google-Certificates-Bridge",
    "remoteIp": "xx.xxx.xx.xx",
    "serverIp": "xxx.xxx.x.xxx",
    "latency": "0s",
    "protocol": "HTTP/1.1"
  },
  "resource": {
    "type": "cloud_run_revision",
    "labels": {
      "configuration_name": "xxxxx",
      "project_id": "xxxxxx",
      "location": "europe-west4",
      "service_name": "xxxx",
      "revision_name": "xxxxxxxx"
    }
  },
  "timestamp": "2024-09-18T12:09:53.656033Z",
  "severity": "INFO",
  "logName": "projects/xxxx/logs/run.googleapis.com%2Frequests",
  "trace": "projects/xxxx/traces/8be1bcad2d711bcbee1164f",
  "receiveTimestamp": "2024-09-18T12:09:53.988892607Z",
  "spanId": "xxxxxxxx",
  "traceSampled": true
}

What worries me is the “302” error, could this be the cause of the problem?

Namely that my project under reactjs uses redirects (because it is a redesign of an existing site whose urls have changed, but not the domain and subdomain). could this come from there? Knowing that the redirected urls are deeper like “/viewtopic.php?id=302” => “/New_page” than the subdomain alone.

The craziest thing in this story is that I did a test with a fake subdomain (example plop.frenchparadise.net") and it didn’t work either. Given that the domain name worked quickly, I have the impression that the problem is at the subdomain level only.